Bactericidal effects of NP-TiO2 under dark condition are still the subject of debate. As an example, some authors found that NP-TiO2 have no effect on the laboratory model strain E. coli while others found for this strain EC50 values differing from 10 magnitude orders. Such contradictory results may be explained by multiple causal factors, including differences in the intrinsic characteristics of NP-TiO2 and bacterial cells used in these studies, exposure conditions for toxicity assessment and even the method used to assess the toxicity, the latter being often inappropriate for nanoparticle toxicity assessment. Increasing knowledge about exposition parameters which trigger NP toxicity and associated molecular mechanisms would contribute to properly evaluate and predict NP effects/fate in the environment. In our study, analysis of the interactions between nanoparticles (NPs) and bacteria highlighted the paramount role played by interfacial electrostatic interactions (NP-cell vs. NP-NP interactions) in determining the extent of NP toxicity and the importance of physico-chemical parameters such as pH and ionic strength in controlling these interactions. In condition of interaction, we showed that nanoparticles causes E.coli membrane depolarization and loss of membrane integrity leading to higher cell permeability. A transcriptomic analysis highlighted that deregulated genes are involved in the response to osmotic stress, metabolism of various cell envelope components and the uptake/metabolism of other endogenous and exogenous compounds. In addition, a significant number of deregulated genes encode proteins localized in the membrane and periplasmic space. All in all, results indicate that the primary effect of NP-TiO2 is initiated at the cell envelope level (membrane depolarization, loss of integrity) triggering an osmotic stress response in bacteria. These results are supported by the observed massive leakage of intracellular K+/Mg2