Infertility and treatments utilized have minimal effects on first trimester human placenta DNA methylation and gene expression.

OBJECTIVE: To determine whether deoxyribonucleic acid (DNA) methylation alterations exist in the first trimester human placenta between conceptions using fertility treatments and those that do not and, if so, whether differences are the result of underlying infertility or fertility treatments. We also assessed whether significant alterations led to changes in gene expression using bulk total RNA-seq from the same cohort (see Lee et al 2019, PMID: 30611556, NCBI GEO GSE215421). DESIGN: We compared DNA methylation of the first trimester placenta from singleton pregnancies that resulted in live births from unassisted, in vitro fertilization (IVF), and non-IVF fertility treatment (NIFT) conceptions using the Infinium MethylationEPIC BeadChip array. Differentially methylated probes (DMPs) were identified using a generalized linear model adjusted for fetal sex. Associated genes for significant CpG sites were cross-references in RNA-seq results to determine whether methylation alterations lead to differences in gene expression. SETTING: Academic medical center. PATIENTS: A total of 138 singleton pregnancies undergoing chorionic villus sampling resulting in a live birth were recruited for methylation analysis (56 unassisted, 38 NIFT, and 44 IVF conceptions). INTERVENTIONS AND TREATMENTS: In vitro fertilization-conceived pregnancy or pregnancy conceived via NIFT, such as ovulation induction and intrauterine insemination. RESULTS: Of the 741,145 CpG probes analyzed in the placenta, few were significant at Bonferroni <0.05: 185 CpG sites (0.025%) significant in pregnancies conceived with the fertility treatments (NIFT + IVF) versus unassisted conceptions; 28 in NIFT versus unassisted; 195 in IVF versus unassisted; and only 13 (0.0018%) in IVF versus NIFT conceptions. CONCLUSION: Underlying infertility is the most significant contributor to the minimal differences in first trimester placental methylation, and not the specific fertility treatment used, such as IVF. Patients were recruited with informed consent on the day of their appointment for chorionic villus sampling, a diagnostic prenatal test performed at late first trimester. Leftover chorionic villi (placental tissue) discarded by the clinical lab was collected for research and stored at -80C until normal karyotype was confirmed and birth outcomes were known. Only singleton pregnancies with normal karyotypes that resulted in live births were used. We measured DNA methylation of human first trimester placenta using the human Infinium MethylationEPIC BeadChip array. We compared pregnancies conceived unassisted (n=56 without fertility treatment or "spontaneous"), with in vitro fertilization (n=44 IVF), or with non-IVF fertility treatment (n=38 NIFT), see original publication (PMID: 36379261). The unassisted group, lacking an infertility diagnosis, was a control for infertility phenotype. The NIFT group was another control to help distinguish DNA methylation profiles due to infertility phenotype (affecting both NIFT and IVF groups) versus IVF technology (IVF group only). DNA methylation sex differences in the unassisted group were also analyzed (PMID: 39152463).