Cell-free DNA Methylation and Transcriptomic Signature Prediction of Pregnancies with Adverse Outcomes - PubMed (original) (raw)
Cell-free DNA Methylation and Transcriptomic Signature Prediction of Pregnancies with Adverse Outcomes
Giorgia Del Vecchio et al. Epigenetics. 2021 Jun.
Abstract
Although analysis of maternal plasma cell-free content has been employed for screening of genetic abnormalities within a pregnancy, limited attention has been paid to its use for the detection of adverse pregnancy outcomes (APOs) based on placental function. Here we investigated cell-free DNA and RNA content of 102 maternal and 25 cord plasma samples. Employing a novel deconvolution methodology, we found that during the first trimester, placenta-specific DNA increased prior to the subsequent development of gestational diabetes with no change in patients with preeclampsia while decreasing with maternal obesity. Moreover, using cell-free RNA sequencing, APOs revealed 71 differentially expressed genes early in pregnancy. We noticed the upregulation of S100A8, MS4A3, and MMP8 that have been already associated with APOs but also the upregulation of BCL2L15 and the downregulation of ALPL that have never been associated with APOs. We constructed a classifier with a positive predictive ability (AUC) of 0.91 for APOs, 0.86 for preeclampsia alone and 0.64 for GDM. We conclude that placenta-specific cell-free nucleic acids during early gestation provide the possibility of predicting APOs prior to the emergence of characteristic clinical features.
Keywords: Cell-free DNA; cell-free RNA; gestational diabetes; gestational hypertension; high-risk pregnancy; preeclampsia.
Conflict of interest statement
No potential conflict of interest was reported by the authors.
Figures
Figure 1.
Illustration of tissue composition inference problem. Scheme identifying tissue-of-origin likelihood of cfDNA sequencing reads and their use for inferring normal tissue composition of plasma cfDNAs
Figure 2.
Changes in DNA methylation and cfDNA composition with advancing gestation. (a) The box plot shows the total CGs methylation in all the subjects (n = 26) recruited in the study across all pregnancies and in cord blood. (b) The box plot shows the total CpGs methylation over the whole genome for 12 normal tissue types and placenta. The normal tissues include four samples of heart, two of colon, and one of adipose tissue, adrenal gland, brain, liver, oesophagus, lung, pancreas and small intestine. The placenta includes 4 samples from normal individuals. The WGBS data for B-cells and neutrophils are from Hodges et al. 2009 [21] and the rest of the normal tissue are from Roadmap Epigenomics Mapping Consortium (ftp://ftp.genboree.org/EpigenomeAtlas/). The placenta data are from Jensen et al. 2015 [22] with mCG calculated as # of methylated CpGs/# of all CpGs from the sequencing data. (c) The plot shows the placenta relative percentage in maternal plasma during pregnancy. The plasma from 7 nulliparous young never pregnant women was used as controls and is represented with a grey bar. (D)The plot shows the deconvolution analysis result for seven tissues. Each bar is an average of the results we obtained from every single subject at each time point of the study. (e) The panel is a whole picture of the deconvolution results obtained from the cell-free DNA analyses. Each dot represents a single subject. (#) indicates statistically significant changes in comparison to the non-pregnant controls by the one-way ANOVA followed by Bonferroni correction. (#) p < 0.05; (##) p < 0.01. (*) indicates statistically significant changes when compared to 1st Trimester values by the one-way ANOVA followed by Bonferroni correction. (*) P < 0.05; (**) P < 0.01
Figure 3.
Maternal plasma cfDNA composition and total CG methylation analysis in APOs. Box plot of the placental contribution expressed in percentage within the GDM (n = 7) (a) group or PreX/gHTN (n = 8) group (b) compared to subjects that did not develop any adverse outcomes during pregnancy (Normal; n = 9). The black circle (A) is to highlight the 1st trimester result for the single obese subject included in the GDM group. (c) Bar plot representing placental weight and baby’s measurements (weight, length, and head circumference) in normal, GDM and PreX/gHTN. Percentage of total CGs methylation in GDM (d) and PreX/gHTN groups (e) compared to the Normal group. The P-values were obtained by the non-parametric Mann-Whitney U test. Each dot represents an individual subject. (#) indicates statistically significant changes with respect to the non-pregnant. (#) p < 0.05; (##) p < 0.01
Figure 4.
BMI effect on maternal plasma cfDNA. (a) The box plot shows placental tissue's relative contribution to the total cfDNA pool according to BMI sub-groups (lean, n = 9; overweight, n = 10; obese, n = 7). The non-pregnant group is represented in the graph as a baseline threshold. No information regarding their BMIs was available. (b) Bar plot of total CG methylation percentage in the different BMI groups during pregnancy and in cord blood. (c) Bar plot representing the placental weight and baby’s measurements (weight, length, and head circumference) in each BMI sub-group is shown. Each dot in the graph represents a single subject in the study. The P-values were obtained by the non-parametric Mann-Whitney U test. (*) indicates statistically significant differences with respect to lean subjects. (*) p < 0.05; (**) p < 0.01
Figure 5.
Deciphering cfRNA content with advancing gestation in normal pregnancy and in pregnancies with adverse outcomes. The box-plot shows the cfRNA placental (a) or bone marrow (e) signature in maternal plasma during pregnancy. The plasma from 7 nulliparous young never pregnant women was used as controls and is represented with a grey bar. (#) indicates statistically significant changes in comparison to non-pregnant controls by the one-way ANOVA followed by Bonferroni correction. (#) p < 0.05; (##) p < 0.01. (*) indicates statistically significant changes when compared to 1st Trimester values by the one-way ANOVA followed by Bonferroni correction. (*) P < 0.05; (**) P < 0.01
Figure 6.
1st and 2nd trimesters differentially expressed genes (DEG). (a) Heatmap showing in a blue-red gradient the DEG genes in the 1st and 2nd trimesters of pregnancy (n = 26) when compared to non-pregnant controls (n = 7). All the represented genes had an adjusted P value ≤ 0.05 and a log2 Fold-Change ≥ 1 or ≤ −1. The adjusted P value was calculated in DESeq that used the Benjamin-Hochberg correction [20]. The horizontal colour bars on the top of the heatmap represent the three different groups of subjects (the lower one) and different time points during pregnancy (the upper one). Each row of the heatmap refers to a single gene and each column is an individual subject. The red boxes highlight the genes that were selected for further analysis. (b) Venn diagram showing the numbers of differentially expressed genes for each group when compared to non-pregnant controls. Yellow represents Normal pregnancy, red GDM, and light blue PreX/gHTN. In addition, the Venn diagram shows the number of overlapping genes between the different groups during the 1st (top) and 2nd (bottom) trimesters of pregnancy
Figure 7.
qRT-PCR reflects gene expression changes observed with RNA sequencing. (A) Comparison of the log2 fold change in expression of six genes quantitated by RNA-seq and qRT-PCR assays. Statistically significant Pearson’s correlation is shown. Quantitative real-time PCR results for CSH2 (B), S100A8 (C), MMP8 (D), ALPL (E), MS4A3 (F), and BCL2L15 (G) are represented. For all genes, the absolute expression was quantified using GAPDH as the internal control employing the ΔCt method. Data are represented as the mean of 2^(-ΔCt) ± SEM. Three replicates were performed at each time point for each subject. The P-values shown in the figure were based on the unpaired Student’s t-test. (*) indicates statistically significant changes when compared to the normal pregnancy group (*) p < 0.05; (**) p < 0.01. (#) indicates statistically significant changes when compared to the non-pregnant group (#) p < 0.05; (##) p < 0.01
Figure 8.
Receiver operator curves (ROC) of classification models
Similar articles
- Integrative single-cell and cell-free plasma RNA transcriptomics elucidates placental cellular dynamics.
Tsang JCH, Vong JSL, Ji L, Poon LCY, Jiang P, Lui KO, Ni YB, To KF, Cheng YKY, Chiu RWK, Lo YMD. Tsang JCH, et al. Proc Natl Acad Sci U S A. 2017 Sep 12;114(37):E7786-E7795. doi: 10.1073/pnas.1710470114. Epub 2017 Aug 22. Proc Natl Acad Sci U S A. 2017. PMID: 28830992 Free PMC article. - Noninvasive preeclampsia prediction using plasma cell-free RNA signatures.
Zhou S, Li J, Yang W, Xue P, Yin Y, Wang Y, Tian P, Peng H, Jiang H, Xu W, Huang S, Zhang R, Wei F, Sun HX, Zhang J, Zhao L. Zhou S, et al. Am J Obstet Gynecol. 2023 Nov;229(5):553.e1-553.e16. doi: 10.1016/j.ajog.2023.05.015. Epub 2023 May 19. Am J Obstet Gynecol. 2023. PMID: 37211139 - Global methylation in the placenta and umbilical cord blood from pregnancies with maternal gestational diabetes, preeclampsia, and obesity.
Nomura Y, Lambertini L, Rialdi A, Lee M, Mystal EY, Grabie M, Manaster I, Huynh N, Finik J, Davey M, Davey K, Ly J, Stone J, Loudon H, Eglinton G, Hurd Y, Newcorn JH, Chen J. Nomura Y, et al. Reprod Sci. 2014 Jan;21(1):131-7. doi: 10.1177/1933719113492206. Epub 2013 Jun 13. Reprod Sci. 2014. PMID: 23765376 Free PMC article. - Cell-Free Nucleic Acids for Early Prediction of Preeclampsia.
Moufarrej MN, Winn VD, Quake SR. Moufarrej MN, et al. Curr Hypertens Rep. 2024 Apr;26(4):175-182. doi: 10.1007/s11906-023-01291-z. Epub 2023 Dec 26. Curr Hypertens Rep. 2024. PMID: 38147201 Review. - Cell-free DNA methylation-based preeclampsia prediction: A journey to improve maternal health.
Aerden M, De Borre M, Thienpont B. Aerden M, et al. Prenat Diagn. 2024 Apr;44(4):418-421. doi: 10.1002/pd.6478. Epub 2023 Dec 4. Prenat Diagn. 2024. PMID: 38047711 Review.
Cited by
- Deciphering DNA Methylation in Gestational Diabetes Mellitus: Epigenetic Regulation and Potential Clinical Applications.
Li N, Liu HY, Liu SM. Li N, et al. Int J Mol Sci. 2024 Aug 29;25(17):9361. doi: 10.3390/ijms25179361. Int J Mol Sci. 2024. PMID: 39273309 Free PMC article. Review. - Longitudinal integrative cell-free DNA analysis in gestational diabetes mellitus.
Tang Z, Wang S, Li X, Hu C, Zhai Q, Wang J, Ye Q, Liu J, Zhang G, Guo Y, Su F, Liu H, Guan L, Jiang C, Chen J, Li M, Ren F, Zhang Y, Huang M, Li L, Zhang H, Hou G, Jin X, Chen F, Zhu H, Li L, Zeng J, Xiao H, Zhou A, Feng L, Gao Y, Liu G. Tang Z, et al. Cell Rep Med. 2024 Aug 20;5(8):101660. doi: 10.1016/j.xcrm.2024.101660. Epub 2024 Jul 25. Cell Rep Med. 2024. PMID: 39059385 Free PMC article. - DNA methylation landscape in pregnancy-induced hypertension: progress and challenges.
Deng F, Lei J, Qiu J, Zhao C, Wang X, Li M, Sun M, Zhang M, Gao Q. Deng F, et al. Reprod Biol Endocrinol. 2024 Jul 8;22(1):77. doi: 10.1186/s12958-024-01248-0. Reprod Biol Endocrinol. 2024. PMID: 38978060 Free PMC article. Review. - A review on cell-free RNA profiling: Insights into metabolic diseases and predictive value for bariatric surgery outcomes.
Mareboina M, Deng E, Mouratidis I, Yee NS, Pitteloud N, Georgakopoulos-Soares I, Chartoumpekis DV. Mareboina M, et al. Mol Metab. 2024 Sep;87:101987. doi: 10.1016/j.molmet.2024.101987. Epub 2024 Jul 6. Mol Metab. 2024. PMID: 38977131 Free PMC article. Review. - Prediction of preeclampsia using maternal circulating mRNAs in early pregnancy.
Chen J, Xu X, Xu X, Yang S, Wang X, Ye A, Yu B. Chen J, et al. Arch Gynecol Obstet. 2024 Jul;310(1):327-335. doi: 10.1007/s00404-024-07486-2. Epub 2024 Apr 3. Arch Gynecol Obstet. 2024. PMID: 38568284
References
- Norton ME, Wapner RJ.. Cell-free DNA Analysis for Noninvasive Examination of Trisomy. N Engl J Med. 2015;373:2581–2582. - PubMed
Publication types
MeSH terms
Substances
Grants and funding
- R01 CA246329/CA/NCI NIH HHS/United States
- U01 HD087221/HD/NICHD NIH HHS/United States
- R01 HD100015/HD/NICHD NIH HHS/United States
- U01 CA237711/CA/NCI NIH HHS/United States
- R01 HD089714/HD/NICHD NIH HHS/United States
LinkOut - more resources
Full Text Sources
Other Literature Sources
Miscellaneous