A Lead-Anchored, Mixture-Aware Multi-Layer Bioinformatics Framework for Prenatal Metal Exposure at the Placental Interface
International Journal of Molecular Sciences · 20.11.2026 · 27 (22) · 10475
DOI: 10.3390/ijms272210475 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Molecular Sciences, online erschienen am 20.11.2026. Verfasst von Muhammet Kesin und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Prenatal lead (Pb) exposure may perturb the maternal–fetal interface, while real-world exposure frequently involves multiple metals. We integrated a 308-record Pb-centered bibliometric corpus; 25 placental-interface studies; a 29-gene literature-derived candidate panel whose source attributions were verified against the primary publications; and human placental epigenomic, transcriptomic, miRNA–mRNA, toxicogenomic and protein-interaction resources. In the Rhode Island Child Health Study (RICHS) cohort, higher maternal Pb exposure, assessed as toenail Pb concentration, was associated with reduced placental 5-methylcytosine (5mC) at GLI2 cg05403744 (estimate = −0.004067 per log2 exposure; source epigenome-wide association study [EWAS] q = 0.02765). A candidate-probe enrichment sensitivity analysis did not show enrichment of the 29-gene probe set over the source Illumina Infinium MethylationEPIC (EPIC) background (P ≈ 0.57; the result was insensitive to the choice of EPIC denominator). In the Extremely Low Gestational Age Newborns (ELGAN) cohort, 16 of the 29 candidates were expressed within the source 11,402-transcript analysis universe and were therefore testable for differential expression. No candidate overlapped the 191 Pb-associated differentially expressed genes (DEGs), an outcome compatible with chance (expected 0.27; lower-tail P = 0.763). RAB5A was the sole candidate among 657 principal component 2 (PC2)/multi-metal-associated DEGs, also compatible with chance (expected 0.92; P[X ≥ 1] = 0.613). Expression-matched permutation against the primary count data reproduced both results (empirical P = 0.83–0.88 and 0.54–0.64), so RAB5A is treated as hypothesis-generating mixture-level evidence rather than statistical convergence. Three experimentally supported miRNA–mRNA pairs showed modest inverse placental correlations (ρ = −0.145 to −0.203) that persisted after adjustment for clinical and histological covariates and in singleton pregnancies, but were abolished by removal of the dominant axis of expression variation (37.5% of variance) and were not distinguishable from randomly paired miRNA–mRNA controls (empirical P = 0.22–0.38). PIK3R1 and WNT3A were therefore removed from the prioritized exposure-linked follow-up set, and the miRNA strand of evidence for RAB5A was downgraded to exploratory context. The prespecified 12-protein core physical protein–protein interaction (PPI) network (CORE12) retained only CASP1–NLRP3 (STRING score = 0.998; enrichment p = 0.103). Overall, the integrated evidence supports a branched, Pb-anchored and mixture-aware hypothesis framed within the adverse outcome pathway (AOP) concept, rather than a demonstrated causal sequence or a statistically enriched gene set. The study contributes a systematic placental-interface map of the Pb-centred developmental literature and an explicit calibration of a literature-derived candidate set against chance expectation, showing how readily overlaps of the size often reported as cross-layer convergence arise under the null.
Zitieren
APA
Kesin, M., Öztan, G., İşsever, T., & İşsever, H. (2026). A Lead-Anchored, Mixture-Aware Multi-Layer Bioinformatics Framework for Prenatal Metal Exposure at the Placental Interface. International Journal of Molecular Sciences, 27(22), 10475. https://doi.org/10.3390/ijms272210475
Vancouver
Kesin M, Öztan G, İşsever T, İşsever H. A Lead-Anchored, Mixture-Aware Multi-Layer Bioinformatics Framework for Prenatal Metal Exposure at the Placental Interface. International Journal of Molecular Sciences. 2026;27(22):10475. doi:10.3390/ijms272210475
BibTeX
@article{Kesin2026,
author = {Kesin, Muhammet and Öztan, Gözde and İşsever, Tuğçe and İşsever, Halim},
title = {A Lead-Anchored, Mixture-Aware Multi-Layer Bioinformatics Framework for Prenatal Metal Exposure at the Placental Interface},
journal = {International Journal of Molecular Sciences},
year = {2026},
volume = {27},
number = {22},
pages = {10475},
doi = {10.3390/ijms272210475}
}
EUVIMED