Skip to main content
Download PDF

Causal association of epigenetic age acceleration and risk of subacute thyroiditis: a bidirectional Mendelian randomization study

Clinical Epigenetics volume 16, Article number: 133 (2024) Cite this article

Abstract

Background

Epigenetic age accelerations (EAAs) are a promising new avenue of research, yet their investigation in subacute thyroiditis (SAT) remains scarce. Our study endeavors to fill this void by exploring the potential causal association between EAAs and SAT.

Methods

Our study utilized publicly available genome-wide association study (GWAS) data of European ancestry to conduct a bidirectional Mendelian randomization (MR) study. Five MR methods were employed to measure causal association between EAAs and SAT multiple analyses were utilized to perform quality control.

Results

Our study evaluated causal association between SAT and four EAAs, included GrimAge acceleration (GrimAA), Hannum age acceleration (HannumAA), PhenoAge acceleration (PhenoAA), intrinsic epigenetic age acceleration (IEAA). Results showed that there is a significant causal association between PhenoAA and SAT (OR 1.109, 95% CI 1.000–1.228, p = 0.049, by IVW method). On the contrary, SAT was associated with IEAA (OR 0.933, 95% CI 0.884–0.984, p = 0.011, by IVW method; OR 0.938, 95% CI 0.881–0.998, p = 0.043, by weighted median method). Leave-one-out sensitivity analysis, heterogeneity test, pleiotropy test, and MR-PRESSO analysis provide good quality control.

Conclusion

The bidirectional MR analysis concluded that an increase in PhenoAA was correlated with a higher risk of SAT, indicating a potential causal relationship between PhenoAA and risk of SAT. Conversely, SAT was found to be closely associated with IEAA, suggesting that SAT may accelerate the aging process. Slowing down biological aging has emerged as a new research direction in curbing SAT.

Introduction

Subacute thyroiditis (SAT) is characterized by inflammation of the thyroid gland due to follicle damage, often accompanied by neck pain and tenderness in the affected area [1]. The majority of individuals affected by SAT are typically in the middle-aged bracket, with women exhibiting a significantly higher propensity, ranging from four to seven times more likely to develop the symptoms compared to men [2, 3]. Nonsteroidal anti-inflammatory medications (NSAIDs) and corticosteroids were utilized to alleviate patient symptoms and attenuate inflammatory reactions [4, 5]. Despite the utilization of medication in the treatment of SAT, the recurrence rate remains notably high, ranging from 1.6% to over 20%. Elevated recurrence rates and prolonged treatment durations have emerged as significant challenges in managing SAT [6, 7].

Aging is intricately linked to an increased risk of cancer, degenerative diseases, metabolic disorders, and cardiovascular conditions, influenced by a combination of environmental and genetic factors [8,9,10,11,12]. As is well known, physiological age measured over time does not precisely reflect the true aging status of the human body, including its tissues and organs. Individual aging is influenced by complex interactions of confounding factors and genetic element [13]. Epigenetic clocks emerge as the most promising predictors of biological age, capable of forecasting both chronological age and mortality [14]. DNA methylation (DNAm) at specific cytosine-phospho-guanine (CpG) sites was found to be strongly correlated with chronological age, so DNAm can be used to predict chronological age. Epigenetic clocks are heritable indicators of biological aging based on DNAm. Different epigenetic clocks based on different DNAm levels were used to assess the characteristics of epigenetic aging [15]. Epigenetic age acceleration (EAA), reflecting the actual variance between epigenetic age and chronological age is closely intertwined with numerous age-associated diseases and life expectancy across racial groups [13, 16]. At present, a series of methods were utilized to assess EAA. Intrinsic epigenetic age acceleration (IEAA) represents aging independent of blood cell type composition [17]. DNA methylation Hannum age acceleration (HannumAA) more accurately reflects extrinsic aging, which evaluated chronological age based on 71 age-related CpGs [18]. DNA methylation GrimAge acceleration (GrimAA) predicted chronological age by 1030 CpGs associated with smoking pack-years and 7 plasma proteins and DNA methylation PhenoAge acceleration (PhenoAA) are based on 513 CpGs associated with mortality and 9 clinical biomarkers. GrimAA and PhenoAA belong to the second generation and are designed to better predict age-associated diseases and mortality [10, 19]. Recent studies have unveiled correlations between epigenetic age acceleration and kidney function [20], as well as infertility [21], atrial fibrillation [11], and schizophrenia [22]. However, there is no study that has revealed correlations between epigenetic age acceleration and SAT.

Mendelian randomization (MR) is a causal inference method linking exposure and disease, employing genetic variants as instrumental variables (IVs) for exposure [23, 24]. To reduce systematic bias and provide results similar to RCTs, the MR approach uses genetic variations associated with exposure as IVs. By using this method, confounding variables within populations are lessened. A recent genome-wide association study (GWAS) meta-analysis disclosed 137 genetic loci related to EAAs, so can be applied to MR study [25]. To far, no research has used MR analysis to investigate the causal relationship between SAT and EAA. In this work, we investigate the causal relationship between the risk of SAT development and the epigenetic clock (IEAA, HunnumAA, GrimAA, PhenoAA) using two-sample MR analysis. For forward MR analysis, EAAs were applied with exposure, single-nucleotide polymorphisms (SNPs) closely associated with EAAs were considered as IVs, and SAT was outcome. For reverse MR analysis, SAT was used as exposure, SNPs associated with SAT were regarded as IVs, and EAAs were regarded as outcome. Our study was based on three assumptions: (1) the relevance assumption with genetic variants closely related to disease; (2) the independence assumption without confounding factors of disturbing casual association between exposure and outcome; and (3) the exclusion restriction assumption without ways for IVs influence outcome [26, 27]. Figure 1 displays MR assumptions and study design.

Fig. 1
figure 1

Study design of the correlation between EAAs and SAT. Assumption 1, the relevance assumption; Assumption 2, the independence assumption; Assumption 3, the exclusion restriction assumption. SAT, subacute thyroiditis; EAAs, epigenetic age accelerations; SNPs, single-nucleotide polymorphisms

Full size image

Materials and methods

Data sources

The MR study strictly adheres to the STROBE-MR guidelines (Table S1) [28]. To observe casual association between EAAs and subacute thyroiditis, we performed two-sample MR analyses. Two-sample MR analyses identified causal association between exposure and outcome by utilizing genetic variants of exposure as IVs, which remedied shortcoming of observational studies and utilized publicly GWAS datasets of exposure (as a risk factor) and outcome (as a disease).

To reduce bias from population stratification, GWAS meta-analyses of European ancestry were utilized to our study. GWAS data of epigenetic clocks were originated from GWAS database (https://gwas.mrcieu.ac.uk/). We selected most-recent largescale GWAS meta-analyses of EAAs, including IEAA, HannumAA, GrimAA and PhenoAA [25]. GWAS data of subacute thyroiditis were originated from FINNGEN database (https://www.finngen.fi/en/access_results). GWAS summary statistics of subacute thyroiditis were used to our study (finn-b-E4_THYROIDITSUBAC, Table S2) [29].

Selection of instrument variables

Appropriate IVs were used in order to guarantee the correctness of MR findings. First, based on the criteria (p < 5e−08), single-nucleotide polymorphisms (SNPs) connected to the exposure were selected as IVs. Nevertheless, we discovered that only four SNPs were chosen for GrimAA exposure in European ancestry, and six SNPs were chosen for exposure for subacute thyroiditis in FINNGEN. Therefore, in order to obtain a comparatively large number of IVs, we gently adjusted the threshold (p < 5e−06). Furthermore, the clump_data function is used to exclude the linkage disequilibrium (LD) and sets the kilobase pairs (kb) to 10,000. The r2 threshold is set to 0.001 (Tables S3 and S4). In addition, we calculated F value through evaluate the intensity of IVs by utilizing the formula beta2/SE2 with F value > 10 were selected [26, 30] (Tables S5 and S6). The formula for calculating the proportion of variation in traits explained by genetic instruments is R2 = 2 × MAF × (1-MAF) × beta, where MAF stands for minor allele frequency, beta for effect size, SE for standard error, N for sample size, and k for the number of IVs [31]. To verify whether selected IVs possessed the independence assumption, LDLink database (https://ldlink.nih.gov/?tab=ldtrait) was applied to exclude SNPs, which may be associated with outcome [32,33,34]. SNPs linked to inflammation and infection, such as immune cell counts (neutrophils, monocytes, white blood cells) and oral ulcers, are eliminated for EAA exposure based on the GWAS characteristic (Table S7). On the other hand, SNPs linked to age and smoking for SAT as exposure (Table S8). SNPs are covered in depth in Tables S9, S10, S11, S12 and S13, with GWAS trait and source citations.

Statistical analysis

In our study, we use five MR methods to perform study. The IVW method served as the primary analytical approach, while others were used as supplementary methods. If all SNPs are deemed valid IVs, then IVW can yield precise estimations [35]. MR Egger can detect and correct for pleiotropy but might have lower precision [36]. Weighted median can offer accurate estimates if over 50% of SNPs are deemed valid IVs [37]. Though simple mode might not wield the same statistical power as IVW, it retains the ability to detect pleiotropy [38]. The selection of the optimal bandwidth for mode estimation can influence weighted mode [39]. Therefore, we primarily rely on the IVW method.

Furthermore, we performed a series of tests to obtain robust and stable results. Heterogeneity test assess heterogeneity of IVs through calculating Q statistic [40]. The MR-PRESSO analysis was employed to exclude SNPs with a p-value less than 0.05. Additionally, any outliers detected were removed from the analysis. In addition, pleiotropy test was used to evaluate horizontal pleiotropy through calculating MR Egger intercept values [41]. Leave-one-out sensitivity analysis was employed to analyze each SNPs, which verify total effect of SNPs after removing one of the SNPs.

All analyses in this study were performed using R software (version 4.3.3). We mainly used “TwoSampleMR” package. All statistical tests were conducted with a two-sided significance level set at p < 0.05.

Results

Selection of instrument variables

To enhance the feasibility of obtaining robust MR results, we strictly selected appropriate SNPs as IVs to examine the relationship between EAA and SAT. For EAAs as the exposure and subacute thyroiditis as the outcome, we selected 26 SNPs for IEAA, 14 SNPs for PhenoAA, 15 SNPs for GrimAA, and 21 SNPs for HannumAA. Conversely, for subacute thyroiditis as the exposure and EAAs as the outcome, 5 SNPs for subacute thyroiditis were chosen. Tables S7 and S8 provide detailed information on these IVs.

Causal analysis of epigenetic age accelerations on subacute thyroiditis

Through MR analyses based on five methods, we found that a partial causal relationship between EAAs and SAT. IVW method revealed a significant causal association between PhenoAA and SAT (OR 1.109, 95% CI 1.000 to 1.228, p = 0.049). However, there are no significant causal association based on other four methods (OR 1.122, 95% CI 0.905–1.390, p = 0.316 by MR Egger; OR 1.141, 95% CI 0.999–1.304, p = 0.052 by weighted median; OR 1.149, 95% CI 0.943–1.400, p = 0.190 by simple mode; OR 1.147, 95% CI 0.962–1.368, p = 0.141 by weighted mode) (Table 1, Table S14). Thus, we deduced that there was a strong correlation between PhenoAA and the elevated risk of SAT. Furthermore, based on five MR methodologies, our data indicate that there is not a significant relationship between SAT and other EAAs, such as GrimAA, HannumAA, and IEAA (Table 1, Table S14). Scatter plot displayed the significant causal association of EAAs on SAT (Fig. 2). Forest plots of SNPs are displayed in Fig. S2.

Table 1 The causal association of EAAs on SAT based on five MR methods
Full size table
Fig. 2
figure 2

Scatter plots of casual relationship evaluate for EAAs on SAT. A, GrimAA on SAT. B, HannumAA on SAT. C, IEAA on SAT. D, PhenoAA on SAT. Contains 15, 21, 26, and 14 IVs, respectively. SAT, subacute thyroiditis; EAAs, epigenetic age accelerations; IEAA, intrinsic epigenetic age acceleration; HunnumAA, DNA methylation Hannum age acceleration; GrimAA, DNA methylation GrimAge acceleration; PhenoAA, DNA methylation PhenoAge acceleration; IVs, instrumental variables

Full size image

There was no significant heterogeneity among SNPs by heterogeneity test based on IVW and MR Egger methods (GrimAA p = 0.868, and Q = 8.386 by IVW, p = 0.902 and Q = 6.995 by MR Egger; HannumAA p = 0.337, and Q = 22.073 by IVW, p = 0.295 and Q = 21.777 by MR Egger; PhenoAA p = 0.205, and Q = 30.536 by IVW, p = 0.171 and Q = 30.428 by MR Egger; IEAA p = 0.393, and Q = 13.732 by IVW, p = 0.319 and Q = 13.715 by MR Egger, Table 2). Leave-one-out sensitivity analysis revealed that after deleting SNP one by one, there are no changes of the total effect (Fig. S1). MR-PRESSO global test was utilized to check detect the presence of horizontal pleiotropy and MR-PRESSO outlier test was used to delete SNPs with horizontal pleiotropy. Results show that there is no horizontal pleiotropy in there SNPs (Table 2, Table S15). Funnel plots further revealed that there is no heterogeneity of SNPs (Fig. S3). MR Egger intercept of pleiotropy test shows the absence of horizontal pleiotropy (Table 2, Table S16). These results imply that our findings from the MR results exhibit substantial confidence, demonstrating robustness and consistency.

Table 2 Sensitivity analyses of EAAs on SAT
Full size table

Casual association of subacute thyroiditis on epigenetic age accelerations

To analyze reverse casual association, we performed a series of analyses by using subacute thyroiditis as exposure. Results show that SAT was associated with IEAA (OR, 0.933, 95% CI 0.884–0.984, p = 0.011, by IVW method; OR 0.938, 95% CI 0.881–0.998, p = 0.043, by weighted median) and the OR values of other methods are in the same direction (OR 0.967, 95% CI 0.877–1.066, p = 0.544 by MR Egger; OR 0.940, 95% CI 0.849–1.040, p = 0.296 by simple mode; OR 0.40, 95% CI 0.878–1.007, p = 0.151 by weighted mode) (Table 3, Table S17). Scatter plot displayed the significant causal association of SAT on EAAs (Fig. 3). Thus, we concluded that SAT was associated with decreased IEAA. Based on five methods, we discovered no causal relationship between SAT and other EAAs (including GrimAA, HannumAA and PhenoAA). Forest plots of SNPs are displayed in Fig. S5. Leave-one-out sensitivity analysis also presents higher the total effect after removing the SNPs one by one (Fig. S4). Heterogeneity test and pleiotropy test did not find heterogeneity and pleiotropy (Table 4, Tables S18–S19). MR-PRESSO analysis did not found SNPs to be excluded in relation to the association between SAT and IEAA (Table 4). Funnel plots further revealed that there is no heterogeneity of SNPs (Fig. S6).

Table 3 The causal association of SAT on EAAs based on five MR methods
Full size table
Fig. 3
figure 3

Scatter plots of casual relationship evaluate for SAT on EAAs. A, GrimAA on SAT. B, HannumAA on SAT. C, IEAA on SAT. D, PhenoAA on SAT. It contains 15, 21, 26, and 14 IVs, respectively. SAT, subacute thyroiditis; EAAs, epigenetic age accelerations; IEAA, intrinsic epigenetic age acceleration; HunnumAA, DNA methylation Hannum age acceleration; GrimAA, DNA methylation GrimAge acceleration; PhenoAA, DNA methylation PhenoAge acceleration; IVs, instrumental variables

Full size image
Table 4 Sensitivity analyses of SAT on EAAs
Full size table

Discussion

In the current study, we utilized GWAS datasets to investigate the association between EAAs and SAT using MR analysis. Our study revealed that there is relationship between EAAs and SAT, increase in PhenoAA was correlated with a higher risk of SAT and SAT was found to be closely associated with decreased IEAA. Health aging possessed obvious heterogeneity, so it is necessary that search higher predictor to cognize and measure aging [42]. Recently, epigenetic clock and telomere length were considered as the most remarkable predictor in biological aging [14]. Previous study shows telomere length was correlated with valve stenosis (AVS) but the predictive ability is lower [43, 44]. Comparing with chronological age and other age-associated biomarkers, epigenetic clock is best predictor of biological aging status, which reflects biological aging status by measuring DNA methylation levels [13, 14, 17, 18]. Pan et al. identified that increased levels of IEAA and GrimAA were associated with elevated risk of chronic kidney disease (CKD) and a decline in kidney function by MR analysis. Furthermore, it was observed that accelerated HannumAA, GrimAA, and PhenoAA may lead to a reduction in log-transformed estimated glomerular filtration rate (log-eGFR). Specifically, for each 1-unit rise in log-eGFR, HannumAA decreased by 3.14 years, GrimAA decreased by 1.99 years, and PhenoAA decreased by 2.88 years, respectively [20]. Roberts JD found that EAA measures of GrimAA and PhenoAA were closely related to higher risk of atrial fibrillation (AF) [11]. Pan et al. concluded that PhenoAA was link to the risk of aortic valve stenosis (AVS) [45].

Nevertheless, no research has looked at the connection between SAT and EAAs. Our research is the first to show a causal association between EAAs and SAT as well as a genetic correlation. There is a strong causal relationship between SAT and EAAs. Thus, a new line of inquiry for SAT prevention is to slow down biological aging. Furthermore, our research revealed an extremely interesting finding that SAT is associated with lower IEAA. Reduced EAA was generally thought to be a beneficial protective factor. Decreased HannumAA, GrimAA, and PhenoAA are connected with increased log-eGFR, which indicated a progressive improvement in kidney function [20]. A recent MR study about obesity and EAAs revealed that EAAs was related to lower risk of obesity and obesity increased EAAs (including GrimAA and PhenoAA) [46], increasing age, and decreasing immunity [47]. The incidence of SAT is closely related to the low immunity [48], which interpreted our result that increased EAAs increase the risk of SAT. SAT is characterized by inflammation of the thyroid gland [1]. Previous study concluded that inflammation is not necessarily harmful to the body, triggering an inflammatory response by releasing inflammatory mediators to destroy pathogens and repair damaged tissues [49]. Therefore, a moderate inflammatory response is necessary to maintain human health. Our study uncovered that SAT related to decreased IEAA. We speculated that SAT may trigger inflammatory response to resist aging, which is a good aspect for SAT patients.

Our study explored causal association between EAAs and SAT. To build reliable causal inferences and overcome influences of confounding and reverse causality biases, we employed basic theory of random allocation of alleles by MR analysis [50]. In order to reduce the potential bias caused by population differences, we only used GWAS data of European ancestry in this study. In addition, rigorous quality evaluation shows that our results are reliable. Our study first reported that there are association between EAAs and SAT. The epigenetic clock may become surveillance indicator to evaluate risk of SAT for clinicians and preventive medicine practitioners, sowing down biological aging has become a new research direction for SAT.

Certainly, our study also has some limitations. Firstly, we only selected European ancestry, which led to concerns that our results would not apply to other populations. Secondly, when selecting SNPs, we used a standard bioinformatic threshold of p < 5 × 10−8, only 4 SNPs were selected for exposure for GrimAA in European-ancestry and 6 SNPs were selected for exposure for subacute thyroiditis in FINNGEN. Consequently, we chose to select SNPs under the threshold of p < 5 × 10−6, as previously employed in other studies [51,52,53,54]. Of course, this approach may introduce bias from weak instrumental variables. To mitigate this limitation, we used the F-test to seek strong evidence. Lastly, epigenetic aging is essential associated with environmental exposures rather than genetic factors, so it is significance to interpret results.

Conclusion

In conclusion, the bidirectional MR analysis revealed that an increase in PhenoAA was correlated with a higher risk of SAT, indicating a potential causal relationship between PhenoAA and risk of SAT. Conversely, SAT was found to be closely associated with IEAA, suggesting that SAT may accelerate the aging process. Slowing down biological aging has emerged as a new research direction in curbing SAT.

Availability of data and materials

No datasets were generated or analyzed during the current study.

References

  1. Desailloud R, Hober D. Viruses and thyroiditis: an update. Virol J. 2009;6:5.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Samuels MH. Subacute, silent, and postpartum thyroiditis. Med Clin North Am. 2012;96(2):223–33.

    Article  PubMed  CAS  Google Scholar 

  3. Stasiak M, Michalak R, Stasiak B, Lewinski A. Clinical characteristics of subacute thyroiditis is different than it used to be-current state based on 15 years own material. Neuro Endocrinol Lett. 2019;39(7):489–95.

    PubMed  Google Scholar 

  4. Ross DS, Burch HB, Cooper DS, et al. 2016 American thyroid association guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis. Thyroid. 2016;26(10):1343–421.

    Article  PubMed  Google Scholar 

  5. Sencar ME, Calapkulu M, Sakiz D, et al. An evaluation of the results of the steroid and non-steroidal anti-inflammatory drug treatments in subacute thyroiditis in relation to persistent hypothyroidism and recurrence. Sci Rep. 2019;9(1):16899.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Stasiak M, Tymoniuk B, Stasiak B, Lewiński A. The risk of recurrence of subacute thyroiditis is HLA-dependent. Int J Mol Sci. 2019;20(5):1089.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  7. Zhang J, Ding G, Li J, et al. Risk factors for subacute thyroiditis recurrence: a systematic review and meta-analysis of cohort studies. Front Endocrinol (Lausanne). 2021;12:783439.

    Article  PubMed  Google Scholar 

  8. Niccoli T, Partridge L. Ageing as a risk factor for disease. Curr Biol. 2012;22(17):R741-752.

    Article  PubMed  CAS  Google Scholar 

  9. Dartigues JF, Le Bourdonnec K, Tabue-Teguo M, et al. Co-Occurrence of geriatric syndromes and diseases in the general population: assessment of the dimensions of aging. J Nutr Health Aging. 2022;26(1):37–45.

    Article  PubMed  Google Scholar 

  10. Levine ME, Lu AT, Quach A, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 2018;10(4):573–91.

    Article  PubMed  Google Scholar 

  11. Roberts JD, Vittinghoff E, Lu AT, et al. Epigenetic age and the risk of incident atrial fibrillation. Circulation. 2021;144(24):1899–911.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  12. Zhang F, Deng S, Zhang J, et al. Causality between heart failure and epigenetic age: a bidirectional Mendelian randomization study. ESC Heart Fail. 2023;10(5):2903–13.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Horvath S, Raj K. DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nat Rev Genet. 2018;19(6):371–84.

    Article  PubMed  CAS  Google Scholar 

  14. Jylhävä J, Pedersen NL, Hägg S. Biological age predictors. EBioMedicine. 2017;21:29–36.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Liu Z, Leung D, Thrush K, et al. Underlying features of epigenetic aging clocks in vivo and in vitro. Aging Cell. 2020;19(10):e13229.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  16. Chen BH, Marioni RE, Colicino E, et al. DNA methylation-based measures of biological age: meta-analysis predicting time to death. Aging (Albany NY). 2016;8(9):1844–65.

    Article  PubMed  CAS  Google Scholar 

  17. Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013;14(10):R115.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Hannum G, Guinney J, Zhao L, et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell. 2013;49(2):359–67.

    Article  PubMed  CAS  Google Scholar 

  19. Lu AT, Quach A, Wilson JG, et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging (Albany NY). 2019;11(2):303–27.

    Article  PubMed  CAS  Google Scholar 

  20. Pan Y, Sun X, Huang Z, et al. Effects of epigenetic age acceleration on kidney function: a Mendelian randomization study. Clin Epigenetics. 2023;15(1):61.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Lee Y, Bohlin J, Page CM, et al. Associations between epigenetic age acceleration and infertility. Hum Reprod. 2022;37(9):2063–74.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  22. Wu X, Ye J, Wang Z, Zhao C. Epigenetic age acceleration was delayed in schizophrenia. Schizophr Bull. 2021;47(3):803–11.

    Article  PubMed  Google Scholar 

  23. Smith GD, Ebrahim S. “Mendelian randomization”: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol. 2003;32(1):1–22.

    Article  PubMed  Google Scholar 

  24. Boef AG, Dekkers OM, le Cessie S. Mendelian randomization studies: a review of the approaches used and the quality of reporting. Int J Epidemiol. 2015;44(2):496–511.

    Article  PubMed  Google Scholar 

  25. McCartney DL, Min JL, Richmond RC, et al. Genome-wide association studies identify 137 genetic loci for DNA methylation biomarkers of aging. Genome Biol. 2021;22(1):194.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  26. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23(R1):R89-98.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  27. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44(2):512–25.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Skrivankova VW, Richmond RC, Woolf BAR, et al. Strengthening the reporting of observational studies in epidemiology using mendelian randomisation (STROBE-MR): explanation and elaboration. BMJ. 2021;375:n2233.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Kurki MI, Karjalainen J, Palta P, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023;613(7944):508–18.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  30. Palmer TM, Lawlor DA, Harbord RM, et al. Using multiple genetic variants as instrumental variables for modifiable risk factors. Stat Methods Med Res. 2012;21(3):223–42.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Pierce BL, Ahsan H, Vanderweele TJ. Power and instrument strength requirements for Mendelian randomization studies using multiple genetic variants. Int J Epidemiol. 2011;40(3):740–52.

    Article  PubMed  Google Scholar 

  32. Machiela MJ, Chanock SJ. LDlink: a web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variants. Bioinformatics. 2015;31(21):3555–7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  33. Lin SH, Brown DW, Machiela MJ. LDtrait: an online tool for identifying published phenotype associations in linkage disequilibrium. Cancer Res. 2020;80(16):3443–6.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  34. Lin SH, Thakur R, Machiela MJ. LDexpress: an online tool for integrating population-specific linkage disequilibrium patterns with tissue-specific expression data. BMC Bioinform. 2021;22(1):608.

    Article  Google Scholar 

  35. Yuan S, Larsson S. Causal associations of iron status with gout and rheumatoid arthritis, but not with inflammatory bowel disease. Clin Nutr. 2020;39(10):3119–24.

    Article  PubMed  CAS  Google Scholar 

  36. Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32(5):377–89.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40(4):304–14.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Milne RL, Kuchenbaecker KB, Michailidou K, et al. Identification of ten variants associated with risk of estrogen-receptor-negative breast cancer. Nat Genet. 2017;49(12):1767–78.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  39. Hartwig FP, Davey Smith G, Bowden J. Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption. Int J Epidemiol. 2017;46(6):1985–98.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Greco MF, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in Mendelian randomisation studies with summary data and a continuous outcome. Stat Med. 2015;34(21):2926–40.

    Article  Google Scholar 

  41. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–8.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  42. Lowsky DJ, Olshansky SJ, Bhattacharya J, Goldman DP. Heterogeneity in healthy aging. J Gerontol A Biol Sci Med Sci. 2014;69(6):640–9.

    Article  PubMed  Google Scholar 

  43. Kurz DJ, Kloeckener-Gruissem B, Akhmedov A, et al. Degenerative aortic valve stenosis, but not coronary disease, is associated with shorter telomere length in the elderly. Arterioscler Thromb Vasc Biol. 2006;26(6):e114-117.

    Article  PubMed  CAS  Google Scholar 

  44. Saraieva I, Benetos A, Labat C, Franco-Cereceda A, Bäck M, Toupance S. Telomere length in valve tissue is shorter in individuals with aortic stenosis and in calcified valve areas. Front Cell Dev Biol. 2021;9:618335.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Pan W, Huang Q, Zhou L, et al. Epigenetic age acceleration and risk of aortic valve stenosis: a bidirectional Mendelian randomization study. Clin Epigenetics. 2024;16(1):41.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Li J, Wang W, Yang Z, et al. Causal association of obesity with epigenetic aging and telomere length: a bidirectional Mendelian randomization study. Lipids Health Dis. 2024;23(1):78.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  47. Linterman MA. Age-dependent changes in T follicular helper cells shape the humoral immune response to vaccination. Semin Immunol. 2023;69:101801.

    Article  PubMed  CAS  Google Scholar 

  48. Bindra A, Braunstein GD. Thyroiditis. Am Fam Physician. 2006;73(10):1769–76.

    PubMed  Google Scholar 

  49. Hotamisligil GS. Inflammation, metaflammation and immunometabolic disorders. Nature. 2017;542(7640):177–85.

    Article  PubMed  CAS  Google Scholar 

  50. Nazarzadeh M, Pinho-Gomes AC, Bidel Z, et al. Plasma lipids and risk of aortic valve stenosis: a Mendelian randomization study. Eur Heart J. 2020;41(40):3913–20.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  51. Zheng J, Baird D, Borges MC, et al. Recent developments in Mendelian randomization studies. Curr Epidemiol Rep. 2017;4(4):330–45.

    Article  PubMed  PubMed Central  Google Scholar 

  52. Ference BA, Majeed F, Penumetcha R, Flack JM, Brook RD. Effect of naturally random allocation to lower low-density lipoprotein cholesterol on the risk of coronary heart disease mediated by polymorphisms in NPC1L1, HMGCR, or both: a 2 × 2 factorial Mendelian randomization study. J Am Coll Cardiol. 2015;65(15):1552–61.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  53. Zou XL, Wang S, Wang LY, et al. Childhood obesity and risk of stroke: a Mendelian randomisation analysis. Front Genet. 2021;12:727475.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Cao Z, Wu Y, Li Q, Li Y, Wu J. A causal relationship between childhood obesity and risk of osteoarthritis: results from a two-sample Mendelian randomization analysis. Ann Med. 2022;54(1):1636–45.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We want to acknowledge the participants and investigators of GWAS and FinnGen study.

Funding

This research is supported by the fellowship of China Postdoctoral Science Foundation (2021M702340), the Science and Technology Department of Sichuan Province (2021ZYCD016, 2022NSFSC1441) and the Postdoctoral research grant of Sichuan University (2023SCU12047).

Author information

Author notes

  1. Bingbing Shen and Yusheng Pu contributed equally to this article.

Authors and Affiliations

  1. Division of Thyroid Surgery, Department of General Surgery, Laboratory of Thyroid and Parathyroid Diseases, Frontiers Science Center for Disease-Related Molecular Network, West China Hospital, Sichuan University, No 37. Guoxue Alley, Chengdu, 610000, China

    Bingbing Shen, Jiaye Liu & Zhihui Li

  2. Department of Respiratory and Critical Care Medicine, Frontiers Science Center for Disease-Related Molecular Network, Center of Precision Medicine, Precision Medicine Key Laboratory of Sichuan Province, West China Hospital, Sichuan University, Chengdu, 610000, China

    Bingbing Shen, Jiaye Liu & Zhihui Li

  3. Center of Gerontology and Geriatrics, West China Hospital, Sichuan University, Chengdu, 610000, China

    Yusheng Pu

  4. West China School of Nursing, Sichuan University, Chengdu, 610000, China

    Yusheng Pu

  5. Department of Endocrinology and Metabolism, West China Hospital, Sichuan University, Chengdu, China

    Xiaofeng Zheng

  6. Center for Diabetes and Metabolism Research, Division of Endocrinology and Metabolism, West China Hospital, Sichuan University, Chengdu, 610000, China

    Xiaofeng Zheng

  7. Department of Obstetrics and Gynecology, The Second Affiliated Hospital, Chongqing Medical University, Chongqing, 400010, China

    Yang Liu

  8. Department of Gynecology, The First Affiliated Hospital, Chongqing Medical University, Chongqing, 400042, China

    Lin Yang

  9. Frontiers Medical Center, Tianfu Jincheng Laboratory, Sichuan University, Chengdu, 610000, China

    Jiaye Liu

Authors
  1. Bingbing Shen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yusheng Pu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Xiaofeng Zheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yang Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Lin Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Jiaye Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Zhihui Li
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

ZHL, BBS, YSP and JYL conceptualized and designed the study. BBS and JYL provided the “TwoSampleMR” package codes in R software and analyzed the data in the study. BBS drafted the manuscript. YL and LY download data of GWAS. JYL and XFZ gave constructive suggestions when writing the manuscript. All authors have read and agreed to the published version of the manuscript.

Corresponding authors

Correspondence to Jiaye Liu or Zhihui Li.

Ethics declarations

Ethics approval and consent to participate

All data for the study were derived from the GWAS database, which has been reviewed and approved by the relevant ethics committee.

Consent for publication

Not applicable.

Conflict of interest

The authors declare 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

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, 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 you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. 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-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Shen, B., Pu, Y., Zheng, X. et al. Causal association of epigenetic age acceleration and risk of subacute thyroiditis: a bidirectional Mendelian randomization study. Clin Epigenet 16, 133 (2024). https://doi.org/10.1186/s13148-024-01743-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13148-024-01743-6

Keywords

Clinical Epigenetics

ISSN: 1868-7083

Contact us