Integrated Transcriptomic and Structural Prioritization of Sulforaphane Targets in Neurodegeneration: A Multi-Dataset In Silico Study
Year: 2026
Type: preprint
Abstract: Sulforaphane (SFN), a natural isothiocyanate from cruciferous vegetables, exerts neuroprotective effects through mechanisms extending beyond canonical Nrf2/ARE activation. Here, we developed a multi-dataset computational framework integrating transcriptomic analysis, multi-criteria target prioritization, and structural docking evaluation. R- and S-SFN were docked against 693 prioritized targets using AutoDock Vina, followed by cross-validation, brain expression profiling, and molecular dynamics simulations. Four genes (IGSF9, KCNQ5, KCTD12, and NFIX) were dysregulated across all four datasets. Among 693 targets, 88 showed favorable predicted docking scores (ΔG ≤ −4.0 kcal/mol), with 65 (73.9%) exhibiting expression in at least one GTEx brain region. Independent cross-validation showed significant enrichment of prioritized targets in an SFN-responsive Alzheimer’s disease transcriptomic dataset (OR = 1.44, 95% CI: 1.08–1.89, p = 0.0104). Complementary disease gene set enrichment confirmed that 33% of strong hits (29/88) overlap established neurodegeneration gene sets across AD, PD, and ALS — including ACHE,target of FDA-approved anti-AD drugs — with highly significant enrichment across all three disease categories (OR range: 67–127, all p < 0.001). R-SFN showed a modest but statistically detectable preference over S-SFN across the 88 strong hits (Wilcoxon p = 0.0081; paired t-test p = 0.0047). TMEM45A, GRIK2, and FABP3 emerged as putative neurodegeneration-relevant targets. Molecular dynamics simulations supported the structural stability of R-SFN poses for TMEM45A, GRIK2, and FABP3, while MM-GBSA estimated favorable binding energies for GRIK2 (ΔG_bind = −19.78 kcal/mol) and FABP3 (ΔG_bind = −14.49 kcal/mol). These findings generate testable hypotheses regarding previously underexplored molecular targets that may contribute to the neuroprotective actions of SFN.
Authors: Semra Ugur, Tugba Onal Suzek
Institution: Muğla University
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7223569774