Transcription Factor Binding and Individual Genetic Risk of Valproate Teratogenicity

  • Alison Anderson
  • , Piero Perucca
  • , Elena Vianca
  • , Daniel Sandvik
  • , Ana Antonic-Baker
  • , Roland Krause
  • , Dana Jazayeri
  • , Alison Hitchcock
  • , Janet Graham
  • , Marian Todaro
  • , Torbjörn Tomson
  • , Dina Battino
  • , Emilio Perucca
  • , Meritxell Martinez Ferri
  • , Anne Rochtus
  • , Lieven Lagae
  • , Maria Paola Canevini
  • , Elena Zambrelli
  • , Ellenr Campbell
  • , Aleksei Rakitin
  • Bobby Koeleman, Ingrid E Scheffer, Samuel F Berkovic, Patrick Kwan, Sanjay M Sisodiya, John J Craig, Frank J Vajda, Terence J O'Brien*,
*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

BACKGROUND AND OBJECTIVES: Valproate (VPA) use during pregnancy is associated with a wide range of structural birth defects, but not all exposed children are affected and there is evidence for a genetic predisposition. The development of a pharmacogenomic biomarker test that can be used for preconception counseling, allowing access to women unnecessarily denied VPA treatment because of concern about teratogenic risks if they were to become pregnant, is challenged by a poor understanding of how variation in maternal DNA could modify the risk. We hypothesized that genomic variants that affect the binding affinity of transcription factors (TFs), key regulators of gene expression, are integral to VPA-associated teratogenicity and a plausible explanation for both variance in interindividual risk and the wide range of birth defect types. METHODS: We interrogated genomic variants within maternal exomes from women recruited through international epilepsy pregnancy registries and genomics consortia. We applied a network-based approach that contextualized the variant spectra to genes associated with diverse birth defect types, gene burden tests, and evidence from multiple modalities to identify variant-sensitive TFs. RESULTS: Sixty-six pregnancies were exposed to VPA as monotherapy or polytherapy, leading to 28 cases with birth defects, and 184 were exposed to other antiseizure medications (ASMs), leading to 20 cases with birth defects. The variant burden within genes associated with 32 different birth defect types was higher for those exposed to VPA compared with those exposed to other ASMs (OR 1.73 [95% CI 1.39 to 2.13], p < 0.0001). Variants in a network comprising significant genes from VPA-exposed mothers were predicted to modify the binding affinity of 359 TFs. These variant-sensitive TFs formed a highly connected protein-protein interaction network, among which the acetyltransferase EP300 connected to 41% (147/359) of all proteins. Profiling of coexpression between EP300 and other TFs in an embryonic stem cell (hESC) model showed that VPA exposure alters EP300-TF interactions. DISCUSSION: These findings suggest that VPA-induced disruption of EP300-related gene regulation is a teratogenic mechanism that is common to heterogeneous birth defect types and sensitive to genetic variation. This has implications for the development of pharmacogenomic risk biomarkers and safer drugs for women of childbearing potential.

Original languageEnglish
Article numbere214570
JournalNeurology
Volume106
Issue number3
Early online date14 Jan 2026
DOIs
Publication statusPublished - 10 Feb 2026

Keywords

  • Abnormalities, Drug-Induced/genetics
  • Adult
  • Anticonvulsants/adverse effects
  • Epilepsy/drug therapy
  • Female
  • Genetic Predisposition to Disease
  • Humans
  • Pregnancy
  • Teratogens
  • Transcription Factors/metabolism
  • Valproic Acid/adverse effects

Fingerprint

Dive into the research topics of 'Transcription Factor Binding and Individual Genetic Risk of Valproate Teratogenicity'. Together they form a unique fingerprint.

Cite this