Inhibition of inflammatory arthritis using fullerene nanomaterials

Abstract

Inflammatory arthritis (e.g. rheumatoid arthritis; RA) is a complex disease driven by the interplay of multiple cellular lineages. Fullerene derivatives have previously been shown to have anti-inflammatory capabilities mediated, in part, by their ability to prevent inflammatory mediator release by mast cells (MC). Recognizing that MC can serve as a cellular link between autoantibodies, soluble mediators, and other effector populations in inflammatory arthritis, it was hypothesized that fullerene derivatives might be used to target this inflammatory disease. A panel of fullerene derivatives was tested for their ability to affect the function of human skin-derived MC as well as other lineages implicated in arthritis, synovial fibroblasts and osteoclasts. It is shown that certain fullerene derivatives blocked FcγR- and TNF-α-induced mediator release from MC; TNF-α-induced mediator release from RA synovial fibroblasts; and maturation of human osteoclasts. MC inhibition by fullerene derivatives was mediated through the reduction of mitochondrial membrane potential and FcγR-mediated increases in cellular reactive oxygen species and NF-κB activation. Based on these in vitro data, two fullerene derivatives (ALM and TGA) were selected for in vivo studies using K/BxN serum transfer arthritis in C57BL/6 mice and collagen-induced arthritis (CIA) in DBA/1 mice. Dye-conjugated fullerenes confirmed localization to affected joints in arthritic animals but not in healthy controls. In the K/BxN moldel, fullerenes attenuated arthritis, an effect accompanied by reduced histologic inflammation, cartilage/bone erosion, and serum levels of TNF-α. Fullerenes remained capable of attenuating K/BxN arthritis in mast cell-deficient mice Cre-Master mice, suggesting that lineages beyond the MC represent relevant targets in this system. These studies suggest that fullerene derivatives may hold promise both as an assessment tool and as anti-inflammatory therapy of arthritis.

Fig 1. Fullerene derivatives reduce degranulation and cytokine production from synovial fibroblast from RA patients, mouse BMMC, and human MC (hMC), and osteoclast formation from human PBMC.

Fig 1A shows FcγRII-null BMMC incubated with fullerene derivatives overnight (10 μg/ml). The next day anti-FcγRII/III antibody 2.4G2 or isotype control was added followed by cross-linking donkey anti-rat (DAR) F(ab)₂. Cells were centrifuged and β-hexosaminidase release or IL-1 production determined in supernatants or lysates, respectively. Data shown are means ± SE of triplicate samples that is representative of three experiments. All data was statistically significant with P values < 0.05. In Fig 1B tissue MC were incubated with fullerene derivatives (10 μg/ml) overnight, washed and preformed IgG anti-NP–NP-BSA immune complexes [8.8 μg/ml anti-NP Ab with 0.13 μg/ml NP-BSA [35]], were incubated with MC for 30 minutes or four hours. Supernatants and cell lysates were prepared for mediator release analysis as described. Data is expressed as mean ± SE from three individual experiments. P values < 0.05 by ANOVA when experimental values are compared with the Ab-only control (not shown). Fig 1C shows fullerene derivatives can inhibit cytokine production from rheumatoid arthritis-derived synovial fibroblasts. Synovial fibroblasts from RA patients were preincubated with or without various fullerene derivatives (10 μg/ml) overnight, washed, and incubated with TNF-α (10 ng/ml for 12 hours). Supernatants were saved and cytokines measured in the supernatants. The percent inhibition of the treated cells was calculated based on the release of cytokines from non- fullerene derivative treated cells. Fig 1D shows the ability for fullerene derivatives to inhibit osteoclast formation. Human PBMC were incubated without (negative) or with RANK ligand (30 ng/ml) and GMCSF (25 ng/ml). After one hour fullerene derivatives were added (10 μg/ml) and remained throughout. In order to verify the differentiation of mononuclear cells to osteoclasts, after eight days of culture, cells were analyzed for tartrate resistant acid phosphatase (TRAP) activity by cytochemistry. The cells with the reddish color represent osteoclast formation and are quantified in the graph (bottom). Results are representative of two separate experiments. Magnification 40X.

Fig 2. Mitochondrial membrane potential correlates with MC degranulation through FcγR receptors and is inhibited by fullerene derivatives.

In Fig 2A, change in mitochondrial membrane potential as a function of the concentration of IC stimulus was assessed. Human MCs were stimulated with graded concentrations of preformed IgG anti-NP/NP-BSA immune complexes as indicated for 10 minutes. As a control, cells without JC-1 and cells with JC-1 plus NP IgG only (no antigen) were incubated in parallel. The above experiment is representative of two separate samples. The percent of degranulation from these cells was 23%, 32%, 39%, and 45% respectively. Mitochondrial membrane polarization was quantified by cytofluorimetry (FL2 channel) using FACs analysis as described above. As seen in Fig 2B, change in mitochondrial membrane potential as a function of time with fixed concentration of IC stimulus was assessed. Human MCs were stimulated with 8.8 μg/ml anti-NP Ab with 0.13 μg/ml NP-BSA of preformed IgG anti-NP/NP-BSA IC for the indicated times. As a control, cells without JC-1 were incubated in parallel. In Fig 2C, Fullerene derivatives inhibit IC-induced increases in mitochondrial membrane potential. Mast cells were incubated overnight with ALM or TGA (10 μg/ml) or media only. The next day cells were challenged with media containing JC-1 probe for 10 minutes at 37°C with or without IC (as in A). After 10 minutes cells were washed with cold PBS, centrifuged and the JC-1 aggregates detected using the FL2. The above experiment is representative of three separate samples. As shown in Fig 2D fullerene derivatives inhibit IC-induced elevations in intracellular ROS levels. Mast cells were incubated overnight with fullerene derivatives, washed and DCF-DA added to cells for 30 minutes at 37°C. After washing cells were activated with optimal concentrations of IC and the fluorescence intensity measured at 525nm after establishing baseline. Figs. show representative numbers from duplicate samples for each condition and are representative of three separate MC cultures. Fig 2E shows that fullerene derivatives can block Fcγ receptor mediated activation of the MC transcription factor NF-κB. Mast cells were incubated with or without fullerene derivatives (10 μg/ml) overnight, washed, and challenged with IC for 24 hours. After washing, in-cell Westerns were performed using the manufacturers protocol. Control wells (those without primary antibodies) were reserved as a source for background well intensity. Further controls were cells incubated without fullerene derivatives or IC. Results represent results from two separate experiments.

Fig 3. Fullerenes targets joints in inflammatory arthritis.

In Fig 3A, non-arthritic control (left) and arthritic (right) mice were injected intravenously with 50 μg/300μl of IR800 conjugated fullernes and imaged six hours later using the Odyssey imaging system. Control mice (left) without inflammatory arthritis received the same concentration of fullerene-dye. Note the joint localization of the Dye-fullerene conjugate in the arthritic mouse. Fig 3B shows whole mouse imaging and Fig 3C shows imaging of externalized organs performed 24 hours after fullerene-dye injection (50 μg/300 μl). Fluorescence intensity is equally portrayed in all and represent a typical mouse out of three treated in parallel. All of the images have undergone background noise subtraction. Fig 3D shows the quantification of fullerene dye concentration in representative organs from the mouse portrayed in Fig 3B–3C.

Fig 4. Fullerene derivatives attenuate inflammatory arthritis in the K/BxN but not CIA model.

As shown in Fig 4A, C57Bl/6 (n = 5 mice/group) mice were injected with K/BxN serum as described in Methods. Two fullerene derivatives, TGA or ALM (40 μg/100 μl), were injected i.p. on Day 0, 2, and every second day. As a control 100 μl of PBS was injected in the control group. Measurements were taken every second day by a blinded observer. Error bars, ±SEM. The * indicates significant differences observed on that day in fullerene derivatives compared to non-fullerene-treated mice (see text). Fig 4B shows representative ankle sections from K/BxN treated C57Bl/6 mice without TGA (left) or with TGA (middle). Control mice not serum challenged are shown on the right. (Scale bars, 50 μm). Fig 4C shows disease pathogenesis in Cre-Master mice (n = 10 mice/group) with and without fullerene derivative, TGA, therapy as above. Fig 4D. Fullerene derivatives inhibit serum TNF-α levels in the K/BxN model and prevent the joint erosion induced by inflammatory arthritis. Serum levels were obtained at peak symptoms from K/BxN-induced C57Bl/6 mice and TNF-α measured as described (CIA model revealed no significant reductions) [36] (n = 5 mice per group).