Associations between epigenetic age acceleration and infertility

Abstract

Study question: Is the use of ART, a proxy for infertility, associated with epigenetic age acceleration?

Summary answer: The epigenetic age acceleration measured by Dunedin Pace of Aging methylation (DunedinPoAm) differed significantly between non-ART and ART mothers.

What is known already: Among mothers who used ART, epigenetic age acceleration may be associated with low oocyte yield and poor ovarian response. However, the difference in epigenetic age acceleration between non-ART and ART mothers (or even fathers) has not been examined.

Study design, size, duration: The Norwegian Mother, Father and Child Cohort Study (MoBa) recruited pregnant women and their partners across Norway at around 18 gestational weeks between 1999 and 2008. Approximately 95 000 mothers, 75 000 fathers and 114 000 children were included. Peripheral blood samples were taken from mothers and fathers at ultrasound appointments or from mothers at childbirth, and umbilical cord blood samples were collected from the newborns at birth.

Participants/materials, setting, methods: Among the MoBa participants, we selected 1000 couples who conceived by coitus and 894 couples who conceived by IVF (n = 525) or ICSI (n = 369). We measured their DNA methylation (DNAm) levels using the Illumina MethylationEPIC array and calculated epigenetic age acceleration. A linear mixed model was used to examine the differences in five different epigenetic age accelerations between non-ART and ART parents.

Main results and the role of chance: We found a significant difference in the epigenetic age acceleration calculated by DunedinPoAm between IVF and non-ART mothers (0.021 years, P-value = 2.89E-06) after adjustment for potential confounders. Further, we detected elevated DunedinPoAm in mothers with tubal factor infertility (0.030 years, P-value = 1.34E-05), ovulation factor (0.023 years, P-value = 0.0018) and unexplained infertility (0.023 years, P-value = 1.39E-04) compared with non-ART mothers. No differences in epigenetic age accelerations between non-ART and ICSI fathers were found. DunedinPoAm also showed stronger associations with smoking, education and parity than the other four epigenetic age accelerations.

Limitations, reasons for caution: We were not able to determine the directionality of the causal pathway between the epigenetic age accelerations and infertility. Since parents' peripheral blood samples were collected after conception, we cannot rule out the possibility that the epigenetic profile of ART mothers was influenced by the ART treatment. Hence, the results should be interpreted with caution, and our results might not be generalizable to non-pregnant women.

Wider implications of the findings: A plausible biological mechanism behind the reported association is that IVF mothers could be closer to menopause than non-ART mothers. The pace of decline of the ovarian reserve that eventually leads to menopause varies between females yet, in general, accelerates after the age of 30, and some studies show an increased risk of infertility in females with low ovarian reserve.

Study funding/competing interest(s): This study was partly funded by the Research Council of Norway (Women's fertility, project no. 320656) and through its Centres of Excellence Funding Scheme (project no. 262700). M.C.M. has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement number 947684). The authors declare no conflict of interest.

Trial registration number: N/A.

Keywords: in vitro fertilization; Norwegian Mother; assisted reproductive technology; ather and Child Cohort Study; epigenetic aging; infertility; intracytoplasmic sperm injection.

Figure 1.

Selection of study participants. The numbers in red refer to the mother–father pairs that were the focus of this study. The samples used in subsequent analyses could be slightly fewer because of missing data in the adjusting variables. IDAT, Intensity Data.

Figure 2.

Scatter plots of maternal age at blood sampling against maternal biomarkers of aging. (A) DNAm-estimated ‘Pace of Aging’ (referred to as DunedinPoAm) by Belsky et al. (2020), (B) PhenoAge by Levine et al. (2018), (C) DNAm-estimated telomere length (DNAmTL) by Lu et al. (2019), (D) DNAmAge by Horvath (2013) and (E) DNAmAge by Hannum et al. (2013). The red dots refer to the mothers with 95% percentile of DunedinPoAm, while the blue dots refer to the mothers with 5% percentile of DunedimPoAm. The dotted line indicates the linear regression of each biomarker of aging on maternal age at blood sampling. DNAm, DNA methylation; PhenoAge, DNAm-estimated phenotypic age; DNAmTL, DNAm-estimated telomere length; DNAmAge, DNA methylation age.

Figure 3.

Scatter plots of paternal age at blood sampling against paternal biomarkers of aging. (A) DNAm-estimated ‘Pace of Aging’ (referred to as DunedinPoAm) by Belsky et al. (2020), (B) PhenoAge by Levine et al. (2018), (C) DNAm-estimated telomere length (DNAmTL) by Lu et al. (2019), (D) DNAmAge by Horvath (2013) and (E) DNAmAge by Hannum et al. (2013). The red dots refer to the fathers with 95% percentile of DunedinPoAm, while the blue dots refer to the fathers with 5% percentile of DunedimPoAm. The dotted line indicates the linear regression of each biomarker of aging on paternal age at blood sampling. DNAm, DNA methylation; PhenoAge, DNAm-estimated phenotypic age; DNAmTL, DNAm-estimated telomere length; DNAmAge, DNA methylation age.

Figure 4.

Differences in epigenetic age accelerations between non-ART and IVF/ICSI parents. The plot on the right side was generated based on the beta coefficient estimates with the standardized outcomes. DNAm, DNA methylation; DunedinPoAm, DNAm-estimated ‘Pace of Aging’ by Belsky et al. (2020); PhenoAgeAccel, DNAm-estimated phenotypic age acceleration; DNAmTL, DNAm-estimated telomere length; DNAmAgeAccel, DNA methylation age acceleration. ¹In mothers, the associations were adjusted for smoking (pre-pregnancy), alcohol intake (pre-pregnancy), BMI (pre-pregnancy), education, parity and plate. ²In fathers, the associations were adjusted for smoking, alcohol intake, BMI, education, partner’s parity and plate. ³Age-adjusted DNAm-estimated telomere length (DNAmTL) by Lu et al. was multiplied by −1. DNAmTL declines with advanced chronological age. ⁴The epigenetic age accelerations as outcome variables were not standardized. ⁵The epigenetic age accelerations as outcome variables were standardized.