Limits...
Characterization of age signatures of DNA methylation in normal and cancer tissues from multiple studies.

Kim J, Kim K, Kim H, Yoon G, Lee K - BMC Genomics (2014)

Bottom Line: Genes related to the normal signature were enriched for aging-related gene ontology terms including metabolic processes, immune system processes, and cell proliferation.The related gene products of the normal signature had more than the average number of interacting partners in a protein interaction network and had a tendency not to interact directly with each other.The genomic sequences of the normal signature were well conserved and the age-associated DNAm levels could satisfactorily predict the chronological ages of tissues regardless of tissue type.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Informatics, Ajou University School of Medicine, Suwon 443-380, South Korea. kiylee@ajou.ac.kr.

ABSTRACT

Background: DNA methylation (DNAm) levels can be used to predict the chronological age of tissues; however, the characteristics of DNAm age signatures in normal and cancer tissues are not well studied using multiple studies.

Results: We studied approximately 4000 normal and cancer samples with multiple tissue types from diverse studies, and using linear and nonlinear regression models identified reliable tissue type-invariant DNAm age signatures. A normal signature comprising 127 CpG loci was highly enriched on the X chromosome. Age-hypermethylated loci were enriched for guanine-and-cytosine-rich regions in CpG islands (CGIs), whereas age-hypomethylated loci were enriched for adenine-and-thymine-rich regions in non-CGIs. However, the cancer signature comprised only 26 age-hypomethylated loci, none on the X chromosome, and with no overlap with the normal signature. Genes related to the normal signature were enriched for aging-related gene ontology terms including metabolic processes, immune system processes, and cell proliferation. The related gene products of the normal signature had more than the average number of interacting partners in a protein interaction network and had a tendency not to interact directly with each other. The genomic sequences of the normal signature were well conserved and the age-associated DNAm levels could satisfactorily predict the chronological ages of tissues regardless of tissue type. Interestingly, the age-associated DNAm increases or decreases of the normal signature were aberrantly accelerated in cancer samples.

Conclusion: These tissue type-invariant DNAm age signatures in normal and cancer can be used to address important questions in developmental biology and cancer research.

Show MeSH

Related in: MedlinePlus

Age-associated DNA methylation signature independent of tissue type. (A, B) Examples of age-associated CpG loci with linear (A) or nonlinear (B) relationships identified in integrated normal samples. (C) Venn diagrams showing the numbers of age-associated CpG loci with three regression models (linear and second- and third-degree nonlinear) in integrated normal, cancer, or all samples. (D) Venn diagram showing the number of age-associated CpG loci among integrated normal, cancer, and all samples. (E) Manhattan plot of age-associated CpG loci in integrated normal samples by chromosome. Hypermethylated CpG loci with age are shown with a –log (P-value) and hypomethylated loci are shown with a log (P-value). The most significant P-values among linear and nonlinear models were chosen. Significant loci are marked as green (hypermethylated) or blue (hypomethylated) dots. The numbers of significant age-associated CpG loci by chromosome. Bar plots of P-values with hypergeometric tests for the degrees of significance of the numbers of the loci. (F) Manhattan plot of age-associated CpG loci in integrated cancer samples by chromosome.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4289351&req=5

Fig4: Age-associated DNA methylation signature independent of tissue type. (A, B) Examples of age-associated CpG loci with linear (A) or nonlinear (B) relationships identified in integrated normal samples. (C) Venn diagrams showing the numbers of age-associated CpG loci with three regression models (linear and second- and third-degree nonlinear) in integrated normal, cancer, or all samples. (D) Venn diagram showing the number of age-associated CpG loci among integrated normal, cancer, and all samples. (E) Manhattan plot of age-associated CpG loci in integrated normal samples by chromosome. Hypermethylated CpG loci with age are shown with a –log (P-value) and hypomethylated loci are shown with a log (P-value). The most significant P-values among linear and nonlinear models were chosen. Significant loci are marked as green (hypermethylated) or blue (hypomethylated) dots. The numbers of significant age-associated CpG loci by chromosome. Bar plots of P-values with hypergeometric tests for the degrees of significance of the numbers of the loci. (F) Manhattan plot of age-associated CpG loci in integrated cancer samples by chromosome.

Mentions: To identify tissue type-invariant age-associated CpG loci, we integrated the normalized DNAm levels of all the disease-free normal samples after removing noisy samples. Using this integrated data set, we first identified CpG loci with a linear relationship with age. For example, the CG19722847 site was linearly hypomethylated (30740381 on chr12; R = -0.65) and CG22736354 was linearly hypermethylated (18230698 on chr6; R = 0.8) according to age, regardless of tissue type (Figure 4A). However, DNAm levels of some loci showed nonlinear patterns according to age (Figure 4B). For these nonlinear relationships, we also observed more rapid changes in DNAm levels at younger ages, which is consistent with previous studies [9, 12]. We also observed similar phenomena for gene-level DNAm levels (Additional file 2: Figure S3A, B). We thus identified tissue type-invariant age-associated DNA methylation signatures using second- and third-degree nonlinear regression models in addition to the linear model. For the threshold, we used three measures, including false discovery rate (FDR) (<0.01), correlation coefficient (≥0.55), and residual error (<0.15), to reduce tissue-type variations. We identified 127 unique CpG loci in the combined normal samples, which we termed a “tissue type-invariant age-associated DNAm signature”. Among these, 80 CpG loci had a linear relationship with age and the other 47 loci were identified using nonlinear models (Additional file 1: Table S2). Seventy-seven loci were hypermethylated and 50 were hypomethylated with age. We also applied a similar approach to examine the DNAm levels of the combined 2181 cancer samples. Compared with normal samples, only 26 age-associated CpG loci were identified (Figure 4C). Interestingly, there was no CpG locus common to the normal and cancer age-associated signatures (Figure 4D). These epigenetic phenomena were also observed with the gene-level DNAm values (Additional file 2: Figure S3C, D). In case of the combined normal and cancer samples, only 18 CpG loci were identified as age-associated. We examined the positions on human chromosomes of the age-associated CpG loci of each of the signatures by separating hypomethylated (blue) and hypermethylated (green) loci, in normal (Figure 4E), cancer (Figure 4F), or combined samples (Additional file 2: Figure S4). Generally, the 127 age-associated loci in normal tissue were distributed throughout the human genome, except for chromosomes 18 and 21. In contrast to a previous study using male pediatric samples [12], the X chromosome had the largest number of age-associated loci. This difference may be caused by differences in sex, age range, and tissue types. We checked the significance of the numbers of loci by chromosome using hypergeometric tests (green bars for hypermethylation and blue bars for hypomethylation with age in Figure 4E). Chromosomes X (P = 8.1E–08), 22 (1.3E–03), 12 (1.7E–02), 1 (4.0E–02), and 16 (4.9E–02) were preferentially enriched for hypermethylated loci with age, whereas chromosomes Y (P = 3.4E–05), X (9.4E–04), 3 (9.5E–03), and 11 (4.3E–02) were enriched for hypomethylated loci. Thus, the sex chromosomes, especially X, were enriched for age-associated CpG loci in disease-free normal tissues. In cancer samples, chromosomes 3 (P = 0.03), 5 (1.7E–03), 6 (0.03), 7 (0.02), 10 (0.01), 11 (1.8E–04), and 21 (0.01) were enriched with age-associated hypomethylated loci (Figure 4F). Interestingly, there were no age-associated loci on the X or Y chromosomes in cancer samples. With the normal and cancer samples matched in age distribution, we also observed similar trends such as no overlap in signature between the normal and cancer samples (Additional file 2: Figure S5).Figure 4


Characterization of age signatures of DNA methylation in normal and cancer tissues from multiple studies.

Kim J, Kim K, Kim H, Yoon G, Lee K - BMC Genomics (2014)

Age-associated DNA methylation signature independent of tissue type. (A, B) Examples of age-associated CpG loci with linear (A) or nonlinear (B) relationships identified in integrated normal samples. (C) Venn diagrams showing the numbers of age-associated CpG loci with three regression models (linear and second- and third-degree nonlinear) in integrated normal, cancer, or all samples. (D) Venn diagram showing the number of age-associated CpG loci among integrated normal, cancer, and all samples. (E) Manhattan plot of age-associated CpG loci in integrated normal samples by chromosome. Hypermethylated CpG loci with age are shown with a –log (P-value) and hypomethylated loci are shown with a log (P-value). The most significant P-values among linear and nonlinear models were chosen. Significant loci are marked as green (hypermethylated) or blue (hypomethylated) dots. The numbers of significant age-associated CpG loci by chromosome. Bar plots of P-values with hypergeometric tests for the degrees of significance of the numbers of the loci. (F) Manhattan plot of age-associated CpG loci in integrated cancer samples by chromosome.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4289351&req=5

Fig4: Age-associated DNA methylation signature independent of tissue type. (A, B) Examples of age-associated CpG loci with linear (A) or nonlinear (B) relationships identified in integrated normal samples. (C) Venn diagrams showing the numbers of age-associated CpG loci with three regression models (linear and second- and third-degree nonlinear) in integrated normal, cancer, or all samples. (D) Venn diagram showing the number of age-associated CpG loci among integrated normal, cancer, and all samples. (E) Manhattan plot of age-associated CpG loci in integrated normal samples by chromosome. Hypermethylated CpG loci with age are shown with a –log (P-value) and hypomethylated loci are shown with a log (P-value). The most significant P-values among linear and nonlinear models were chosen. Significant loci are marked as green (hypermethylated) or blue (hypomethylated) dots. The numbers of significant age-associated CpG loci by chromosome. Bar plots of P-values with hypergeometric tests for the degrees of significance of the numbers of the loci. (F) Manhattan plot of age-associated CpG loci in integrated cancer samples by chromosome.
Mentions: To identify tissue type-invariant age-associated CpG loci, we integrated the normalized DNAm levels of all the disease-free normal samples after removing noisy samples. Using this integrated data set, we first identified CpG loci with a linear relationship with age. For example, the CG19722847 site was linearly hypomethylated (30740381 on chr12; R = -0.65) and CG22736354 was linearly hypermethylated (18230698 on chr6; R = 0.8) according to age, regardless of tissue type (Figure 4A). However, DNAm levels of some loci showed nonlinear patterns according to age (Figure 4B). For these nonlinear relationships, we also observed more rapid changes in DNAm levels at younger ages, which is consistent with previous studies [9, 12]. We also observed similar phenomena for gene-level DNAm levels (Additional file 2: Figure S3A, B). We thus identified tissue type-invariant age-associated DNA methylation signatures using second- and third-degree nonlinear regression models in addition to the linear model. For the threshold, we used three measures, including false discovery rate (FDR) (<0.01), correlation coefficient (≥0.55), and residual error (<0.15), to reduce tissue-type variations. We identified 127 unique CpG loci in the combined normal samples, which we termed a “tissue type-invariant age-associated DNAm signature”. Among these, 80 CpG loci had a linear relationship with age and the other 47 loci were identified using nonlinear models (Additional file 1: Table S2). Seventy-seven loci were hypermethylated and 50 were hypomethylated with age. We also applied a similar approach to examine the DNAm levels of the combined 2181 cancer samples. Compared with normal samples, only 26 age-associated CpG loci were identified (Figure 4C). Interestingly, there was no CpG locus common to the normal and cancer age-associated signatures (Figure 4D). These epigenetic phenomena were also observed with the gene-level DNAm values (Additional file 2: Figure S3C, D). In case of the combined normal and cancer samples, only 18 CpG loci were identified as age-associated. We examined the positions on human chromosomes of the age-associated CpG loci of each of the signatures by separating hypomethylated (blue) and hypermethylated (green) loci, in normal (Figure 4E), cancer (Figure 4F), or combined samples (Additional file 2: Figure S4). Generally, the 127 age-associated loci in normal tissue were distributed throughout the human genome, except for chromosomes 18 and 21. In contrast to a previous study using male pediatric samples [12], the X chromosome had the largest number of age-associated loci. This difference may be caused by differences in sex, age range, and tissue types. We checked the significance of the numbers of loci by chromosome using hypergeometric tests (green bars for hypermethylation and blue bars for hypomethylation with age in Figure 4E). Chromosomes X (P = 8.1E–08), 22 (1.3E–03), 12 (1.7E–02), 1 (4.0E–02), and 16 (4.9E–02) were preferentially enriched for hypermethylated loci with age, whereas chromosomes Y (P = 3.4E–05), X (9.4E–04), 3 (9.5E–03), and 11 (4.3E–02) were enriched for hypomethylated loci. Thus, the sex chromosomes, especially X, were enriched for age-associated CpG loci in disease-free normal tissues. In cancer samples, chromosomes 3 (P = 0.03), 5 (1.7E–03), 6 (0.03), 7 (0.02), 10 (0.01), 11 (1.8E–04), and 21 (0.01) were enriched with age-associated hypomethylated loci (Figure 4F). Interestingly, there were no age-associated loci on the X or Y chromosomes in cancer samples. With the normal and cancer samples matched in age distribution, we also observed similar trends such as no overlap in signature between the normal and cancer samples (Additional file 2: Figure S5).Figure 4

Bottom Line: Genes related to the normal signature were enriched for aging-related gene ontology terms including metabolic processes, immune system processes, and cell proliferation.The related gene products of the normal signature had more than the average number of interacting partners in a protein interaction network and had a tendency not to interact directly with each other.The genomic sequences of the normal signature were well conserved and the age-associated DNAm levels could satisfactorily predict the chronological ages of tissues regardless of tissue type.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Informatics, Ajou University School of Medicine, Suwon 443-380, South Korea. kiylee@ajou.ac.kr.

ABSTRACT

Background: DNA methylation (DNAm) levels can be used to predict the chronological age of tissues; however, the characteristics of DNAm age signatures in normal and cancer tissues are not well studied using multiple studies.

Results: We studied approximately 4000 normal and cancer samples with multiple tissue types from diverse studies, and using linear and nonlinear regression models identified reliable tissue type-invariant DNAm age signatures. A normal signature comprising 127 CpG loci was highly enriched on the X chromosome. Age-hypermethylated loci were enriched for guanine-and-cytosine-rich regions in CpG islands (CGIs), whereas age-hypomethylated loci were enriched for adenine-and-thymine-rich regions in non-CGIs. However, the cancer signature comprised only 26 age-hypomethylated loci, none on the X chromosome, and with no overlap with the normal signature. Genes related to the normal signature were enriched for aging-related gene ontology terms including metabolic processes, immune system processes, and cell proliferation. The related gene products of the normal signature had more than the average number of interacting partners in a protein interaction network and had a tendency not to interact directly with each other. The genomic sequences of the normal signature were well conserved and the age-associated DNAm levels could satisfactorily predict the chronological ages of tissues regardless of tissue type. Interestingly, the age-associated DNAm increases or decreases of the normal signature were aberrantly accelerated in cancer samples.

Conclusion: These tissue type-invariant DNAm age signatures in normal and cancer can be used to address important questions in developmental biology and cancer research.

Show MeSH
Related in: MedlinePlus