The In-Vivo Use of Superparamagnetic Iron Oxide Nanoparticles to Detect Inflammation Elicits a Cytokine Response but Does Not Aggravate Experimental Arthritis

Background Superparamagnetic Iron Oxide Nanoparticles (SPION) are used in diagnostic imaging of a variety of different diseases. For such in-vivo application, an additional coating with a polymer, for example polyvinyl alcohol (PVA), is needed to stabilize the SPION and prevent aggregation. As the particles are foreign to the body, reaction against the SPION could occur. In this study we investigated the effects that SPION may have on experimental arthritis after intra-articular (i.a.) or intravenous (i.v.) injection. Methods PVA-coated SPION were injected either i.a. (6 or 24 mu g iron) or i.v. (100 mu g or 1 mg iron) into naive Toll-like receptor-4 deficient (TLR4-/-) or wild-type C57Bl/6 mice, or C57Bl/6 mice with antigen-induced arthritis. As control, some mice were injected with PVA or PBS. MR imaging was performed at 1 and 7 days after injection. Mice were sacrificed 2 hours and 1, 2, 7, 10 and 14 days after injection of the SPION, and RNA from synovium and liver was isolated for pro-inflammatory gene expression analysis. Serum cytokine measurements and whole knee joint histology were also performed. Results Injection of a high dose of SPION or PVA into naive knee joints resulted in an immediate upregulation of pro-inflammatory gene expression in the synovium. A similar gene expression profile was observed after SPION or PVA injection into knee joints of TLR4-/-mice, indicating that this effect is not due to LPS contamination. Histological analysis of the knee joints also revealed synovial inflammation after SPION injection. Two hours after i.v. injection of SPION or PVA into naive mice, an upregulation of pro-inflammatory gene expression was detected in the liver. Administration of SPION or PVA into arthritic mice via i.a. injection did not result in an upregulation in gene expression and also no additional effects were observed on histology. MR imaging and histology showed long-term retention of SPION in the inflamed joint. However, 14 days after the injections no long-term effects were evident for gene expression, histology or serum cytokine concentrations. Conclusions Injection of SPION, either locally or systemically, gives an acute inflammatory response. In the long term, up to 14 days after the injection, while the SPION reside in the joint, no further activating effects of SPION were observed. Hence, we conclude that SPION do not aggravate arthritis and can therefore be used safely to detect joint inflammation by MR imaging.