Blue light from dental photopolymerization devices has significant biological effects on cells. These effects may alter normal cell function of tissues exposed during placement of oral restorations, but recent data suggest that some light-induced effects may also be therapeutically useful, for example in the treatment of epithelial cancers. Reactive oxygen species (ROS) appear to mediate blue light effects in cells, but the sources of ROS (intra- versus extracellular) and their respective roles in the cellular response to blue light are not known. In the current study, we tested the hypothesis that intra- and extracellular sources of blue light-generated ROS synergize to depress mitochondrial function. Normal human epidermal keratinocytes (NHEK) and oral squamous cell carcinoma (OSC2) cells were exposed to blue light (380-500 nm; 5-60 J/cm2) from a dental photopolymerization source (quartz-tungsten-halogen, 550 mW/cm2). Light was applied in cell-culture media or balanced salt solutions with or without cells present. Intracellular ROS levels were estimated using the dihydrofluorescein diacetate (DFDA) assay; extracellular ROS levels were estimated using the leucocrystal violet assay. Cell response was estimated using the MTT mitochondrial activity assay. Blue light increased intracellular ROS equally in both NHEK and OSC2. Blue light also increased ROS levels in cell-free MEM or salt solutions, and riboflavin supplements increased ROS formation. Extracellularly applied ROS rapidly (50-400 μM, <1 min) increased intracellular ROS levels, which were higher and longer-lived in NHEK than OSC2. The type of cell-culture medium significantly affected the ability of blue light to suppress cellular mitochondrial activity; the greatest suppression was observed in DMEM-containing or NHEK media. Collectively, the data support our hypothesis that intra- and extracellularly generated ROS synergize to affect cellular mitochondrial suppression of tumor cells in response to blue light. However, the identity of blue light targets that mediate these changes remain unclear. These data support additional investigations into the risks of coincident exposure of tissues to blue light during material polymerization of restorative materials, and possible therapeutic benefits.