Bingbing Sun, Xiang Wang, Yu-Pei Liao, Zhaoxia Ji, Chong Hyun Chang, Suman Pokhrel, Justine Ku, Xiangsheng Liu, Meiying Wang, Darren R. Dunphy, Ruibin Li, Huan Meng, Lutz Mädler, C. Jeffrey Brinker, André E. Nel, and Tian Xia.

ACS Nano (2016) 10, 8054-8066

http://dx.doi.org/10.1021/acsnano.6b04143

Contrary to the notion that the use of fumed silica in consumer products can “generally (be) regarded as safe” (GRAS), the high surface reactivity of pyrogenic silica differs from other forms of synthetic amorphous
silica (SAS), including the capacity to induce membrane damage and acute proinflammatory changes in the murine lung. In addition, the chain-likestructure and reactive surface silanols also allow fumed silica to activate the NLRP3 inflammasome, leading to IL-1β production. This pathway is known to be associated with subchronic inflammation and profibrogenic effects in the lung by α-quartz and carbon nanotubes. However, different from the latter materials, bolus dose instillation of 21 mg/kg fumed silica did not induce sustained IL-1β production or subchronic pulmonary effects. In contrast, the NLRP3 inflammasome pathway was  continuously activated by repetitive-dose administration of 3 × 7 mg/kg fumed silica, 1 week apart.
We also found that while single-dose exposure failed to induce profibrotic effects in the lung, repetitive dosing can trigger increased collagen production, even at 3 × 3 mg/kg. The change between bolus and repetitive dosing was due to a change in lung clearance, with recurrent dosing leading to fumed silica  biopersistence, sustained macrophage recruitment, and activation of the NLRP3 pathway. These subchronic proinflammatory effects disappeared when less surface-reactive titanium-doped fumed silica was used for recurrent administration. All considered, these data indicate that while fumed silica may be regarded as safe for some applications, we should reconsider the GRAS label during repetitive or chronic inhalation exposure conditions.