BACKGROUND: Precisely how silver nanoparticles (AgNPs) kill mammalian cells still is not fully understood. It is not clear if AgNP-induced damage differs from silver cation (Ag), nor is it known how AgNP damage is transmitted from cell membranes, including endosomes, to other organelles. Cells can differ in relative sensitivity to AgNPs or Ag, which adds another layer of complexity to identifying specific mechanisms of action. Therefore, we determined if there were specific effects of AgNPs that differed from Ag in cells with high or low sensitivity to either toxicant. METHODS: Cells were exposed to intact AgNPs, Ag, or defined mixtures of AgNPs with Ag, and viability was assessed. The level of dissolved Ag in AgNP suspensions was determined using inductively coupled plasma mass spectrometry. Changes in reactive oxygen species following AgNP or Ag exposure were quantified, and treatment with catalase, an enzyme that catalyzes the decomposition of HO to water and oxygen, was used to determine selectively the contribution of HO to AgNP and Ag induced cell death. Lipid peroxides, formation of 4-hydroxynonenol protein adducts, protein thiol oxidation, protein aggregation, and activation of the integrated stress response after AgNP or Ag exposure were quantified. Lastly, cell membrane integrity and indications of apoptosis or necrosis in AgNP and Ag treated cells were examined by flow cytometry. RESULTS: We identified AgNPs with negligible Ag contamination. We found that SUM159 cells, which are a triple-negative breast cancer cell line, were more sensitive to AgNP exposure less sensitive to Ag compared to iMECs, an immortalized, breast epithelial cell line. This indicates that high sensitivity to AgNPs was not predictive of similar sensitivity to Ag. Exposure to AgNPs increased protein thiol oxidation, misfolded proteins, and activation of the integrated stress response in AgNP sensitive SUM159 cells but not in iMEC cells. In contrast, Ag cause similar damage in Ag sensitive iMEC cells but not in SUM159 cells. Both Ag and AgNP exposure increased HO levels; however, treatment with catalase rescued cells from Ag cytotoxicity but not from AgNPs. Instead, our data support a mechanism by which damage from AgNP exposure propagates through cells by generation of lipid peroxides, subsequent lipid peroxide mediated oxidation of proteins, and via generation of 4-hydroxynonenal (4-HNE) protein adducts. CONCLUSIONS: There are distinct differences in the responses of cells to AgNPs and Ag. Specifically, AgNPs drive cell death through lipid peroxidation leading to proteotoxicity and necrotic cell death, whereas Ag increases HO, which drives oxidative stress and apoptotic cell death. This work identifies a previously unknown mechanism by which AgNPs kill mammalian cells that is not dependent upon the contribution of Ag released in extracellular media. Understanding precisely which factors drive the toxicity of AgNPs is essential for biomedical applications such as cancer therapy, and of importance to identifying consequences of unintended exposures.