Skip to main content
Download PDF

Associations of socioeconomic disparities with buccal DNA-methylation measures of biological aging

Clinical Epigenetics volume 15, Article number: 70 (2023) Cite this article

Abstract

Background

Individuals who are socioeconomically disadvantaged are at increased risk for aging-related diseases and perform less well on tests of cognitive function. The weathering hypothesis proposes that these disparities in physical and cognitive health arise from an acceleration of biological processes of aging. Theories of how life adversity is biologically embedded identify epigenetic alterations, including DNA methylation (DNAm), as a mechanistic interface between the environment and health. Consistent with the weathering hypothesis and theories of biological embedding, recently developed DNAm algorithms have revealed profiles reflective of more advanced aging and lower cognitive function among socioeconomically-at-risk groups. These DNAm algorithms were developed using blood-DNA, but social and behavioral science research commonly collect saliva or cheek-swab DNA. This discrepancy is a potential barrier to research to elucidate mechanisms through which socioeconomic disadvantage affects aging and cognition. We therefore tested if social gradients observed in blood DNAm measures could be reproduced using buccal-cell DNA obtained from cheek swabs.

Results

We analyzed three DNAm measures of biological aging and one DNAm measure of cognitive performance, all of which showed socioeconomic gradients in previous studies: the PhenoAge and GrimAge DNAm clocks, DunedinPACE, and Epigenetic-g. We first computed blood-buccal cross-tissue correlations in n = 21 adults (GEO111165). Cross-tissue correlations were low-to-moderate (r = .25 to r = .48). We next conducted analyses of socioeconomic gradients using buccal DNAm data from SOEP-G (n = 1128, 57% female; age mean = 42 yrs, SD = 21.56, range 0–72). Associations of socioeconomic status with DNAm measures of aging were in the expected direction, but were smaller as compared to reports from blood DNAm datasets (r = − .08 to r = − .13).

Conclusions

Our findings are consistent with the hypothesis that socioeconomic disadvantage is associated with DNAm indicators of worse physical health. However, relatively low cross-tissue correlations and attenuated effect sizes for socioeconomic gradients in buccal DNAm compared with reports from analysis of blood DNAm suggest that in order to take full advantage of buccal DNA samples, DNAm algorithms customized to buccal DNAm are needed.

Background

Individuals who are socioeconomically disadvantaged are at increased risk for aging-related diseases and exhibit lower average levels of cognitive function across the life course [1,2,3]. Studies of humans and other animals identify several biological pathways through which social factors drive disease, including dysregulation of immune and metabolic systems in response to chronic stress [4]. These pathways overlap substantially with the biology that mediates aging-related health declines [5]. This overlap is consistent with the weathering hypothesis, which proposes that social adversity accelerates biological processes of aging [6].

Biological aging can be conceptualized as the progressive loss of system integrity that occurs with advancing chronological age [7]. The current state-of-the-art for quantification of biological aging in epidemiological studies of humans is a family of DNA-methylation (DNAm) measurements. Epigenetic changes, including DNAm, are among the hallmarks of aging and are theorized to be key transducers of biological embedding of social adversity [5, 8]. Consistent with the hypothesis that DNAm measures of biological aging capture both of these epigenetic mechanisms, the DNAm measures of biological aging that show the most consistent and strongest prediction of aging-related disease, disability, and mortality also show consistent associations with social determinants of health (i.e., the GrimAge, PhenoAge, and DunedinPACE epigenetic clocks; [9, 10]. The current study, therefore, focuses on these DNAm measures. Consistent with the hypothesis that DNAm measures of biological aging capture both of these epigenetic mechanisms, the DNAm measures of biological aging that show the most consistent and strongest prediction of aging-related disease, disability, and mortality also show consistent associations with social determinants of health (i.e., the GrimAge, PhenoAge, and DunedinPACE epigenetic clocks) [9, 10]. In addition, there is evidence for social patterning of a DNAm measurement quantifying cognitive performance (i.e., Epigenetic-g) [11], which parallels well-documented socioeconomic disparities in cognitive function across the life course [2]. These DNAm measures open opportunities to study mechanisms of social disparities in physical and cognitive health and to guide the development and evaluation of interventions to address them.

A barrier to achieving this potential is that DNAm is specific to types of tissues and cells; it is a critical mechanism of cellular differentiation and determinant of cellular phenotype [12]. Most DNAm algorithms used to study social gradients in health were developed from analysis of DNA derived from blood samples. Therefore, the ideal setting for their application is blood-derived DNA methylation. However, collection of blood samples is not feasible in some studies. For these studies, alternative sources of DNA, such as saliva and buccal tissue (i.e., inner cheek), may be easier to obtain. The extent to which algorithms developed from blood-derived DNA can provide reliable and valid measurements in alternative tissues remains uncertain.

In two prior projects, we followed up algorithms developed to measure biological aging and cognitive functioning from blood DNAm in saliva samples collected from a pediatric cohort [13, 14]. In those studies, we were able to replicate several observations made from blood samples. First, the DNAm measure of the pace of biological aging (i.e., a previous iteration of DunedinPACE) exhibited a parallel socioeconomic gradient in the pediatric saliva samples as had been observed previously in blood DNAm datasets from adults. Second, the DNAm measure of cognitive functioning Epigenetic-g exhibited parallel association with children’s performance on cognitive tests as had been observed previously in a blood DNAm dataset from adults. In contrast, the PhenoAge and GrimAge DNAm measures of biological age showed no social gradient in the pediatric saliva samples, in contrast to results from studies of blood samples [15].

Saliva is composed of a mix of leukocytes (which are also the source of blood-derived DNA samples) and epithelial cells. Buccal sample-derived DNA comes predominantly from epithelial cells. It is unclear whether DNAm measures computed in buccal DNAm will show similar evidence of trans-tissue validation. Here, we examined whether the same socioeconomic gradients in biological aging and DNAm-predicted cognitive performance apparent in blood DNAm analyses could be reproduced in analysis of buccal DNAm. The analysis we report is based on a pre-registration plan filed with OSF (https://osf.io/msjgc). Where our work has developed beyond this original pre-registration, we note it in the text. We first tested cross-tissue correlations of DNAm measures of biological aging (i.e., PhenoAge Accel., GrimAge Accel., DunedinPACE) and DNAm-predicted cognitive performance (i.e., Epigenetic-g) in buccal and blood DNAm datasets generated from the same individuals using the public dataset GEO111165 (n = 21). Next, we examined association of chronological age with buccal DNAm measures in n = 1128 participants from SOEP-G (57% female; age mean = 42 yrs, SD = 21.56, range 0–72). Finally, we tested associations of socioeconomic status with DNAm algorithms computed from buccal-cell DNAm in the same SOEP-G sample.

Results

Cross-tissue correlations between blood and buccal samples were low-to-moderate

We evaluated the correspondence between buccal and blood DNAm measures in an auxiliary dataset that collected both buccal and blood samples from the same n = 21 people [16], Illumina EPIC array dataset in Gene Expression Omnibus accession GSE11116, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE111165).

Cross-tissue correlations between blood and buccal samples of the DNAm measures were low-to-moderate across measures (r = 0.25 to r = 0.48). Means of DNAm measures were higher in buccal compared to blood samples, with the exception of Epigenetic-g, for which mean comparisons are not possible because beta-methylation values are standardized prior to computation (see Table 1).

Table 1 Blood-buccal cross-tissue correlations of blood-based DNA-methylation measures (n = 21)
Full size table

Chronological age gradients in biological aging are reproduced in buccal DNAm

We examined associations of chronological age with buccal DNAm algorithms. For PhenoAge, strong association with chronological age is expected. In SOEP-G, participants’ buccal DNAm PhenoAge values were highly correlated with their chronological ages (PhenoAge r = 0.89, 95% CI = 0.88, 0.90, p < 0.001). GrimAge calculations include information about participant chronological age and, as a result, show very strong correlations (r = 0.99, 95% CI = 0.99, 0.99, p < 0.001). In contrast to PhenoAge and GrimAge, which estimate biological age values, DunedinPACE estimates the pace of aging. Consistent with prior reports from blood DNAm datasets and with biodemography theory, which proposes that the pace of aging accelerates as we grow older [17, 18], participants’ DunedinPACE values were modestly correlated with their chronological ages (r = 0.24, 95% CI = 0.18, 0.29, p < 0.001). We also observed positive age trends for Epigenetic-g, mirroring known patterns of cognitive development; values increased across the first half of the lifespan and then stabilized in late middle age (r = 0.45, 95% CI = 0.40, 0.49, p < 0.001; age in years unstandardized b = 0.008, 95% CI = 0.006–0.011, p < 0.001; and age squared unstandardized b = − 0.001, 95% CI = − 0.001–0.000, p = 0.001). Age patterning of DNAm measures is shown in Fig. 1.

Fig. 1
figure 1

Chronological age and buccal DNAm algorithms. Panel A–B plot associations of chronological age with buccal DNAm algorithms of biological aging, for which strong associations are expected: A PhenoAge and B GrimAge. Panel C plots association of chronological age with the pace of aging, DunedinPACE. Panel D plots association of chronological age with a DNAm algorithm of cognitive performance, Epigenetic-g

Full size image

Socioeconomic disadvantage is associated with accelerated biological aging in Germany

We tested associations of socioeconomic status (SES) with DNAm measures of biological aging computed from buccal-cell DNAm in SOEP-G. SES was measured as a composite of household income and educational levels (highest in household). Consistent with reports from blood DNAm datasets, participants with higher SES had younger biological ages and slower pace of aging (r’s = − 0.08 to − 0.13, p’s < 0.011, Table 2).

Table 2 Associations of socioeconomic status with buccal DNA-methylation measures
Full size table

Next, according to our pre-registered analysis plan, we tested whether the association of SES with DNAm measures of aging differed by chronological age. This interaction was statistically significant for PhenoAge and GrimAge Acceleration (SES by continuous age interaction on PhenoAge std b = − 0.11, 95% CI = − 0.17, − 0.05, p < 0.001; GrimAge std b = − 0.07, 95% CI = − 0.13, − 0.02, p = 0.011). There were no age differences in the SES association with DunedinPACE (p-value for continuous age interaction = 0.916). To further illustrate the interaction, we stratified the sample into older and younger participants (mean split). Among the older participants (aged > 42 years, n = 576), the SES association with PhenoAge Acceleration was r = − 0.14, 95% CI = − 0.22, − 0.06, p < 0.001 and with GrimAge Acceleration was r = − 0.18, 95% CI = − 0.26, − 0.10, p < 0.001. In contrast, among younger participants (aged < 42 years, n = 482), the SES association with PhenoAge Acceleration was r = 0.03, 95% CI = − 0.06, 0.12, p = 0.494 and with GrimAge Acceleration was r = − 0.04, 95% CI = − 0.13, 0.05, p = 0.352. In sum, SES was associated with PhenoAge and GrimAge Acceleration only for older participants, whereas low SES was associated with DunedinPACE across age groups. Figure 2 shows the association of socioeconomic status with DNAm by age.

Fig. 2
figure 2

Socioeconomic status and buccal DNAm algorithms. Panel A–B plot associations of socioeconomic status with buccal DNAm algorithms of accelerated biological aging: A PhenoAge Acceleration and B GrimAge Acceleration. Panel C plots association of socioeconomic status with the pace of aging, DunedinPACE. Panel D plots association of socioeconomic status with a DNAm algorithm of cognitive performance, Epigenetic-g

Full size image

Association of socioeconomic status with Epigenetic-g was in the expected direction, but was small and not statistically different from zero at the alpha = 0.05 level (see Table S4 and Fig. 2D). Excluding smokers and accounting for body mass index did not substantially affect associations with SES (see Additional file 1: Figure S1).

Discussion

We tested if socioeconomic gradients in DNAm measurements of biological aging and cognitive performance, which are apparent in blood DNAm analyses, could be reproduced in analysis of buccal DNAm. Our findings are consistent with the weathering hypothesis that socioeconomic disadvantage is associated with accelerated biological aging. However, effect sizes were approximately 50% lower than those reported in previously published analyses of blood DNAm datasets. Such studies have reported associations of magnitude of approximately r = .20, ranging from r = .10 to r = .37 [10], whereas here we report associations of magnitude of approximately r = .10, ranging from r = .079 to r = .13. Similarly, associations of socioeconomic status with buccal DNAm-predicted cognitive performance were attenuated by approximately 50% and not statistically different from zero, in contrast to studies of blood and saliva DNAm datasets, which have reported associations with socioeconomic measures of magnitude r = .11 and r = .14 [11, 19] (note larger effect sizes for neighborhood-level socioeconomic contexts in 19). Moreover, cross-tissue correspondence of DNAm indices was low-to-moderate. Collectively, these findings suggest that in order to take full advantage of buccal DNA samples, it will be important to develop DNAm indices that are customized to buccal DNAm.

One observation from our buccal DNAm data is that SES was associated with more PhenoAge Acceleration and GrimAge Acceleration only for older participants, whereas in the case of DunedinPACE, the socioeconomic gradient was evident for both young and old participants. This pattern of results is consistent with findings from saliva DNAm in children and adolescents, which showed no association of PhenoAge and GrimAge with household SES, but did identify an association with DunedinPACE [13]. One possible explanation for this result is that measures of biological age, such as PhenoAge and GrimAge, which were designed to quantify differences in mortality risk among midlife and older adults, may be less sensitive to early stages in the biological embedding of social disadvantage. Replication of this result in other datasets and across tissues is needed.

Conclusion

Our findings are consistent with the hypothesis that socioeconomic disadvantage is associated with accelerated biological aging in Germany. However, cross-tissue correspondence of DNAm indices was low-to-moderate, and effect sizes for SES associations estimated from buccal DNAm were attenuated by roughly 50% compared with reports from blood DNAm datasets. Development of DNAm measures of biological aging and cognitive performance that are customized to buccal DNAm should be a research priority.

Methods

Participants

SOEP-G participants were from the SOEP-IS cohort, which is based on a random sample of German households and contains a rich array of information on socioeconomic context, household dynamics, personality, and health [20]. Six thousand five hundred seventy-six people were originally invited to participate in the 2019 wave of the SOEP-IS with the aim to collect saliva for genotyping, 2598 of whom provided a valid genetic sample. ~ 98% of the genotyped SOEP-IS sample is of high genetic similarity to European reference groups. See Koellinger et al. [20] for more information on the genotyped SOEP-IS cohort called SOEP-G.

Residual frozen DNA samples from n = 1128 individuals from the n = 2598 genotyped SOEP-G cohort were selected for DNAm extraction based on the availability of funds (see Table 3 for descriptive statistics). Exclusion and inclusion criteria were: (1) exclusion of five samples due to sex mismatch between self-reported and genetic sex, (2) inclusion of all samples from children and adolescents (i.e., under or equal to 18 yrs) whose residual DNA samples contained at least 50 ng of DNA, (3) inclusion of adults that had (a) at least 250 ng of DNA left, (b) had a DNA call rate of at least 0.975, (c) were not parents of selected children and adolescents so that the maximum number of different households were included, and (d) extended the age distribution continuously past 18 years so that all younger adults were included. The ID list was randomized so that plate effects were not confounded with chronological age. In addition, 24 samples were randomly selected as technical duplicates. The final sample of n = 1128 unrelated participants (490 male, 638 female) consisted of 872 adults and 256 children and adolescents (age mean = 41.88 yrs, SD = 21.56, range 0–72, see Additional file 1: Figure S1 for density plot of age distribution). 95% of participants were born in Germany.

Table 3 Descriptive statistics of the analytic sample after DNA-methylation-based exclusions (N = 1058)
Full size table

Measures

DNA-methylation preprocessing and exclusions

DNA was extracted from buccal swabs collected using Isohelix IS SK-1S Dri-Capsules [20]. DNA extraction and methylation profiling was conducted by the Human Genomics Facility (HuGe-F) at the Erasmus Medical Center in Rotterdam, Netherlands. The Infinium MethylEPIC v1 manifest B5 kit (Illumina, Inc., San Diego, CA) was used to assess methylation levels at 865,918 CpG sites.

DNAm preprocessing was primarily conducted with Illumina’s GenomeStudio software and open-source R (version 4.2.0) packages ‘minfi’ [21] and ‘ewastools’ [22]. We generated 20 control metrics in GenomeStudio as described in the BeadArray Controls Reporter Software Guide from Illumina (note similar parameters can be computed using the ewastools ‘control_metrics()’ function). Samples falling below the Illumina-recommended cut-offs were flagged and further investigated. Flagged samples were classified as failed if 1. all types of poor bisulfite conversion and all types of poor bisulfite conversion background; 2. all types of bisulfite conversion background falling below 0.5; 3. all types of poor hybridization; and 4. all types of poor specificity (excluded n = 42).

As a second step, we identified unreliable data points resulting from low fluorescence intensities by filtering using detection p-values, calculated from comparing fluorescence intensities to a noise distribution. We removed probes with only background signal in a high proportion of samples (proportion of samples with detection p-value > 0.01 is > 0.1). We also removed probes for which a high proportion of samples had low bead numbers (proportion of samples with bead number < 3 is > 0.1). Further, we removed probes with SNPs at the CG or single base extension position as well as cross-reactive probes for EPIC arrays [23, 24].

We used minfi’s ‘preprocessNoob’ [25] to correct for background noise and color dye bias and ‘BMIQ’ to account for probe-type differences [26].

Cell composition was estimated using HEpiDISH, which is an iterative hierarchical version of the EpiDISH R package using robust partial correlations (https://github.com/sjczheng/EpiDISH). Because epithelial cell types are the dominant cell type in buccal samples, we applied a threshold of 0.5 for epithelial cell proportions to reliably call a ‘buccal sample’ and excluded samples that failed this metric (n = 28). All samples were from the same batch. Final analytic sample size after DNAm exclusions was n = 1058.

In GSE111165 blood samples, DNAm algorithms were residualized for reference-free cell composition and plate [27].

DNA-methylation algorithms

Our pre-registered analysis focused on two DNAm measures developed from blood DNAm data and which we had previously followed up in saliva DNAm data (i.e., DunedinPACE and Epigenetic-g) as well as a buccal-based algorithm of chronological age to be used as a data quality control measure (i.e., PedBE). For comparative purposes, we report additional results for two further DNAm measures developed from blood DNAm, the PhenoAge and GrimAge clocks [28, 29]. We include these measures, which are among the best-evidenced DNAm biomarkers of aging, to help contextualize findings for DunedinPACE and Epigenetic-g. See Table 4 for description of DNA-methylation algorithm computations.

Table 4 Description of DNA-methylation algorithm computations
Full size table

Socioeconomic status We deviated from our pre-registered analysis plan by testing associations with socioeconomic status (average z-scored household income and education) rather than examining income and education separately, to reduce the number of statistical comparisons. Monthly household net income in Euros from all sources (e.g., employment, pensions, unemployment benefits, maternity benefits, higher education grants, military or civil service pay, compulsory child support, etc.) was reported by the self-defined head of household. In the 2% of cases with missing income values, information about determinants of household income and past data were used to impute estimated values (for more information see page 27 https://www.diw.de/documents/publikationen/73/diw_01.c.787445.de/diw_ssp0844.pdf)/. Household income was divided by the number of persons in the household and sqrt transformed to correct for skew (this deviated from our pre-registration plan; sqrt-transformation improved normality of distribution more than log-transformation in Shapiro–Wilk test).

Given the wide age range of participants, we indexed educational attainment as the highest degree obtained by any individual in the household. Educational attainment was converted to number of educational years (no degree = 7 years, lower school degree = 9 years, intermediary school = 10 years, degree for, a professional coll. = 12 years, high school degree = 13 years, other = 10 years) with additional occupational training added (apprenticeship = + 1.5 years, technical schools (including health) = + 2 years, civil servants apprenticeship = + 1.5 years, higher technical college = + 3 years, university degree = + 5 years).

Covariates

Body mass index (BMI) Height (in cm) and weight (in kg) were measured via self-report and transformed to sex- and age-normed BMI z-scores.

Smoking Participant self-reported current or past smoking across multiple waves. Across questions and waves, if a participant ever responded that they smoked currently or in the past, they were identified as a smoker. If a participant ever responded that they never smoked and never responded that they did smoke, they were identified as a never-smoker.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

DNAm:

DNA methylation

SES:

Socioeconomic status

References

  1. Gkiouleka A, Huijts T, Beckfield J, Bambra C. Understanding the micro and macro politics of health: inequalities, intersectionality & institutions—a research agenda. Soc Sci Med. 2018;200:92–8. https://doi.org/10.1016/j.socscimed.2018.01.025.

    Article  PubMed  Google Scholar 

  2. Lövdén M, Fratiglioni L, Glymour MM, Lindenberger U, Tucker-Drob EM. Education and cognitive functioning across the life span. Psychol Sci Public Interest. 2020;21(1):6–41. https://doi.org/10.1177/1529100620920576.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Tucker-Drob EM. Cognitive aging and dementia: a life-span perspective. Annu Rev Dev Psychol. 2019;1(1):177–96. https://doi.org/10.1146/annurev-devpsych-121318-085204.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Snyder-Mackler N, Burger JR, Gaydosh L, Belsky DW, Noppert GA, Campos FA, Bartolomucci A, Yang YC, Aiello AE, O’Rand A, Harris KM, Shively CA, Alberts SC, Tung J. Social determinants of health and survival in humans and other animals. Science. 2020. https://doi.org/10.1126/science.aax9553.

    Article  PubMed  PubMed Central  Google Scholar 

  5. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194–217. https://doi.org/10.1016/j.cell.2013.05.039.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Geronimus AT, Hicken M, Keene D, Bound J. “Weathering” and age patterns of allostatic load scores among blacks and whites in the United States. Am J Public Health. 2006;96(5):8.

    Article  Google Scholar 

  7. Kirkwood TBL. Understanding the odd science of aging. Cell. 2005;120(4):437–47. https://doi.org/10.1016/j.cell.2005.01.027.

    Article  CAS  PubMed  Google Scholar 

  8. Hertzman C, Boyce T. How experience gets under the skin to create gradients in developmental health. Annu Rev Public Health. 2010;31(1):329–47. https://doi.org/10.1146/annurev.publhealth.012809.103538.

    Article  PubMed  Google Scholar 

  9. Oblak L, van der Zaag J, Higgins-Chen AT, Levine ME, Boks MP. A systematic review of biological, social and environmental factors associated with epigenetic clock acceleration. Age Res Rev. 2021;69:101348. https://doi.org/10.1016/j.arr.2021.101348.

    Article  CAS  Google Scholar 

  10. Raffington L, Belsky DW. Integrating DNA methylation measures of biological aging into social determinants of health research. Curr Environ Health Rep. 2022;15:196.

    Article  Google Scholar 

  11. McCartney DL, Hillary RF, Conole ELS, Banos DT, Gadd DA, Walker RM, Nangle C, Flaig R, Campbell A, Murray AD, Maniega SM, Valdés-Hernández MC, Harris MA, Bastin ME, Wardlaw JM, Harris SE, Porteous DJ, Tucker-Drob EM, McIntosh AM, Marioni RE. Blood-based epigenome-wide analyses of cognitive abilities. Genome Biol. 2022;23(1):26. https://doi.org/10.1186/s13059-021-02596-5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Bakulski KM, Halladay A, Hu VW, Mill J, Fallin MD. Epigenetic research in neuropsychiatric disorders: the “tissue issue.” Curr Behav Neurosci Rep. 2016;3(3):264–74. https://doi.org/10.1007/s40473-016-0083-4.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Raffington L, Belsky DW, Kothari M, Malanchini M, Tucker-Drob EM, Harden KP. Socioeconomic disadvantage and the pace of biological aging in children. Pediatrics. 2021. https://doi.org/10.1542/peds.2020-024406.

    Article  PubMed  Google Scholar 

  14. Raffington L, Tanksley PT, Sabhlok A, Vinnik L, Mallard T, King LS, Goosby B, Harden KP, Tucker-Drob EM. Socially stratified epigenetic profiles are associated with cognitive functioning in children and adolescents. Psychol Sci. 2023;16:170.

    Article  Google Scholar 

  15. Schmitz LL, Zhao W, Ratliff SM, Goodwin J, Miao J, Lu Q, Guo X, Taylor KD, Ding J, Liu Y, Levine M, Smith JA. The socioeconomic gradient in epigenetic aging clocks: evidence from the multi-ethnic study of atherosclerosis and the health and retirement study [Preprint]. Epidemiology. 2021. https://doi.org/10.1101/2021.03.01.21252660.

    Article  Google Scholar 

  16. Braun PR, Han S, Hing B, Nagahama Y, Gaul LN, Heinzman JT, Grossbach AJ, Close L, Dlouhy BJ, Howard MA, Kawasaki H, Potash JB, Shinozaki G. Genome-wide DNA methylation comparison between live human brain and peripheral tissues within individuals. Transl Psychiatry. 2019;9(1):47. https://doi.org/10.1038/s41398-019-0376-y.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Belsky DW, Caspi A, Corcoran DL, Sugden K, Poulton R, Arseneault L, Baccarelli A, Chamarti K, Gao X, Hannon E, Harrington HL, Houts R, Kothari M, Kwon D, Mill J, Schwartz J, Vokonas P, Wang C, Williams BS, Moffitt TE. DunedinPACE, a DNA methylation biomarker of the pace of aging. Elife. 2022;11:e73420. https://doi.org/10.7554/eLife.73420.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Finch CE, Crimmins EM. Constant molecular aging rates vs. The exponential acceleration of mortality. Proc Natl Acad Sci. 2016;113(5):1121–3. https://doi.org/10.1073/pnas.1524017113.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Raffington L, Tanksley PT, Sabhlok A, Vinnik L, Mallard T, King LS, Goosby B, Harden KP, Tucker-Drob EM. Socially stratified epigenetic profiles are associated with cognitive functioning in children and adolescents [Preprint]. Genetics. 2021. https://doi.org/10.1101/2021.08.19.456979.

    Article  Google Scholar 

  20. Koellinger PD, Okbay A, Kweon H, Schweinert A, Karlsson Linnér R, Goebel J, Richter D, Reiber L, Zweck BM, Belsky DW, Biroli P, Mata R, Tucker-Drob EM, Harden KP, Wagner G, Hertwig R. Cohort profile: genetic data in the german socio-economic panel innovation sample (Gene-SOEP) [Preprint]. Genetics. 2021. https://doi.org/10.1101/2021.11.06.467573.

    Article  Google Scholar 

  21. Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, Irizarry RA. Minfi: a flexible and comprehensive bioconductor package for the analysis of infinium DNA methylation microarrays. Bioinformatics. 2014;30(10):1363–9. https://doi.org/10.1093/bioinformatics/btu049.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Heiss JA, Just AC. Identifying mislabeled and contaminated DNA methylation microarray data: an extended quality control toolset with examples from GEO. Clin Epigenetics. 2018;10(1):73. https://doi.org/10.1186/s13148-018-0504-1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. McCartney DL, Walker RM, Morris SW, McIntosh AM, Porteous DJ, Evans KL. Identification of polymorphic and off-target probe binding sites on the illumina infinium MethylationEPIC BeadChip. Genomics Data. 2016;9:22–4. https://doi.org/10.1016/j.gdata.2016.05.012.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Pidsley R, Zotenko E, Peters TJ, Lawrence MG, Risbridger GP, Molloy P, Van Djik S, Muhlhausler B, Stirzaker C, Clark SJ. Critical evaluation of the Illumina MethylationEPIC BeadChip microarray for whole-genome DNA methylation profiling. Genome Biol. 2016;17(1):208. https://doi.org/10.1186/s13059-016-1066-1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Triche TJ, Weisenberger DJ, Van Den Berg D, Laird PW, Siegmund KD. Low-level processing of illumina infinium DNA methylation BeadArrays. Nucleic Acids Res. 2013;41(7):e90. https://doi.org/10.1093/nar/gkt090.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Teschendorff AE, Widschwendter M. Differential variability improves the identification of cancer risk markers in DNA methylation studies profiling precursor cancer lesions. Bioinformatics. 2012;28(11):1487–94. https://doi.org/10.1093/bioinformatics/bts170.

    Article  CAS  PubMed  Google Scholar 

  27. Houseman EA, Kile ML, Christiani DC, Ince TA, Kelsey KT, Marsit CJ. Reference-free deconvolution of DNA methylation data and mediation by cell composition effects. BMC Bioinf. 2016;17(1):259. https://doi.org/10.1186/s12859-016-1140-4.

    Article  CAS  Google Scholar 

  28. Levine ME, Lu AT, Quach A, Chen BH, Assimes TL, Bandinelli S, Hou L, Baccarelli AA, Stewart JD, Li Y, Whitsel EA, Wilson JG, Reiner AP, Aviv A, Lohman K, Liu Y, Ferrucci L, Horvath S. An epigenetic biomarker of aging for lifespan and healthspan. Aging. 2018;10(4):573–91. https://doi.org/10.18632/aging.101414.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Lu AT, Quach A, Wilson JG, Reiner AP, Aviv A, Raj K, Hou L, Baccarelli AA, Li Y, Stewart JD, Whitsel EA, Assimes TL, Ferrucci L, Horvath S. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging. 2019;11(2):303–27. https://doi.org/10.18632/aging.101684.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Higgins-Chen AT, Thrush KL, Wang Y, Minteer CJ, Kuo P-L, Wang M, Niimi P, Sturm G, Lin J, Moore AZ, Bandinelli S, Vinkers CH, Vermetten E, Rutten BPF, Geuze E, Okhuijsen-Pfeifer C, van der Horst MZ, Schreiter S, Gutwinski S, Levine ME. A computational solution for bolstering reliability of epigenetic clocks: Implications for clinical trials and longitudinal tracking. Nature Aging. 2022;2(7):644–61. https://doi.org/10.1038/s43587-022-00248-2.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Belsky DW, Caspi A, Houts R, Cohen HJ, Corcoran DL, Danese A, Harrington H, Israel S, Levine ME, Schaefer JD, Sugden K, Williams B, Yashin AI, Poulton R, Moffitt TE. Quantification of biological aging in young adults. Proc Natl Acad Sci. 2015;112(30):E4104–10. https://doi.org/10.1073/pnas.1506264112.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Elliott ML, Caspi A, Houts RM, Ambler A, Broadbent JM, Hancox RJ, Harrington H, Hogan S, Keenan R, Knodt A, Leung JH, Melzer TR, Purdy SC, Ramrakha S, Richmond-Rakerd LS, Righarts A, Sugden K, Thomson WM, Thorne PR, Moffitt TE. Disparities in the pace of biological aging among midlife adults of the same chronological age have implications for future frailty risk and policy. Nature Aging. 2021;1(3):295–308. https://doi.org/10.1038/s43587-021-00044-4.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Sugden K, Hannon EJ, Arseneault L, Belsky DW, Corcoran DL, Fisher HL, Houts RM, Kandaswamy R, Moffitt TE, Poulton R, Prinz JA, Rasmussen LJH, Williams BS, Wong CCY, Mill J, Caspi A. Patterns of reliability: assessing the reproducibility and integrity of DNA methylation measurement. Patterns. 2020. https://doi.org/10.1016/j.patter.2020.100014.

    Article  PubMed  PubMed Central  Google Scholar 

  34. McEwen LM, O’Donnell KJ, McGill MG, Edgar RD, Jones MJ, MacIsaac JL, Lin DTS, Ramadori K, Morin A, Gladish N, Garg E, Unternaehrer E, Pokhvisneva I, Karnani N, Kee MZL, Klengel T, Adler NE, Barr RG, Letourneau N, Kobor MS. The PedBE clock accurately estimates DNA methylation age in pediatric buccal cells. Proc Natl Acad Sci. 2019. https://doi.org/10.1073/pnas.1820843116.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank the participants of SOEP-G.

Funding

Open Access funding enabled and organized by Projekt DEAL. KPH and EMTD are Faculty Research Associates of the Population Research Center at the University of Texas at Austin, which is supported by a NIH grant P2CHD042849. EMTD is a member of the Center on Aging and Population Sciences (CAPS) at The University of Texas at Austin, which is supported by NIH grant P30AG066614. KPH and EMTD were also supported by Jacobs Foundation Research Fellowships. DWB received support from R01AG073402. KPH received support from R01HD092548. EMTD received support from RF1AG073593. The funding bodies had no role in the design, collection, analysis or interpretation of the study.

Author information

Authors and Affiliations

  1. Max Planck Research Group Biosocial – Biology, Social Disparities, and Development, Max Planck Institute for Human Development, Lentzeallee 94, 14195, Berlin, Germany

    L. Raffington & M. Aikins

  2. Department of Psychology, The University of Texas at Austin, Austin, TX, USA

    T. Schwaba, K. P. Harden & E. M. Tucker-Drob

  3. SHARE Berlin Institute, Berlin, Germany

    D. Richter

  4. Educational Science and Psychology, Free University Berlin, Berlin, Germany

    D. Richter

  5. Max Planck Institute for Human Development, Berlin, Germany

    G. G. Wagner

  6. Federal Institute for Population Research, Wiesbaden, Berlin, Germany

    G. G. Wagner

  7. German Socio-Economic Panel Study (SOEP), Berlin, Germany

    G. G. Wagner

  8. Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA

    D. W. Belsky

  9. Robert N Butler Columbia Aging Center, Columbia University Mailman School of Public Health, New York, NY, USA

    D. W. Belsky

Authors
  1. L. Raffington
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. T. Schwaba
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. M. Aikins
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. D. Richter
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. G. G. Wagner
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. K. P. Harden
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. D. W. Belsky
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. E. M. Tucker-Drob
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

LR developed the study concept and design, performed and supervised data analysis, and drafted the manuscript. TS performed data analysis and provided critical revisions. MA provided critical revisions. DR and GGW developed the study concept and provided critical revisions. KPH, DWB, and EMTD developed the study concept and design, supervised data analysis, drafted the manuscript, and provided critical revisions. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work. All authors read and approved the final manuscript.

Corresponding author

Correspondence to L. Raffington.

Ethics declarations

Ethics approval and consent to participate

Participants or the parents of minor participants consented to the archiving, extraction, and analysis of DNA-based measures. Ethical approval for DNA-based research was received by the Research Ethics Review Board of Vrije Universiteit Amsterdam, School of Business and Economics (application number 20181018.1.pkr730) and the Ethics Council of the Max Planck Society (application number 2019_16).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Fig. S1.

Correlation matrix of socioeconomic variables with buccal DNA-methylation algorithms in full sample, excluding self-reported smokers, and residualizing DNAm algorithms for body mass index. In all plots, DNAm measures were residualized for chronological age. BMI was age- and sex-normed.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Raffington, L., Schwaba, T., Aikins, M. et al. Associations of socioeconomic disparities with buccal DNA-methylation measures of biological aging. Clin Epigenet 15, 70 (2023). https://doi.org/10.1186/s13148-023-01489-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13148-023-01489-7

Keywords

Clinical Epigenetics

ISSN: 1868-7083

Contact us