Integrating network pharmacology and in vitro verification of the anti-breast cancer potential of Lepidium sativum seed extract: Insights into DNA damage and cellular signaling pathways

Background Cancer comprises diverse diseases characterized by uncontrolled cell growth, genetic instability, and accumulated DNA damage. This study investigated the anticancer potential and mechanisms of Lepidium sativum seed extract in breast cancer using network pharmacology, molecular docking, and experimental validation. Targets of active compounds in L. sativum seeds were predicted using SwissTargetPrediction, while breast cancer-associated targets were obtained from GeneCards. Their intersection was used to construct a protein–protein interaction network and identify hub targets. AKT1 emerged as a prominent predicted target and was selected for molecular docking. The effects of the whole seed extract were evaluated by a 24-h MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) in MCF-7 and MDA-MB-231 cells. DNA damage response and repair genes were assessed using an 84-gene PCR array and validated by RT-qPCR. Results The PPI network included 82 overlapping targets. GO analysis identified affected molecular functions, biological processes, and cellular components, while KEGG analysis revealed several cancer-related pathways. The extract caused a dose-dependent reduction in cell viability in both cell lines. PCR-array analysis showed modulation of DDR genes, and RT-qPCR confirmed significant changes with fold changes >2 or <−2. PPP1R15A and GADD45G were markedly upregulated ( p < 0.0001), whereas MRE11A was significantly downregulated ( p = 0.008). Conclusions These findings indicate that L. sativum seed extract exerts moderate, cell line-dependent antiproliferative effects and modulates DDR-related gene expression, supporting its potential as a source of complementary anticancer compounds. Further studies using purified constituents, normal breast epithelial cells, protein-level validation, longer treatment periods, and in vivo models are needed.