Background NF-κB activation serves as a master regulatory switch for an inflammatory response in macrophages. Chondroitin sulphate (CS), a natural glycosaminoglycan of the cartilage extracellular matrix, has been shown to have clinical benefit in osteoarthritis (OA) producing anti-inflammatory and positive effects on the joint (1). We have previously shown that hyaluronan (HA) fragments induce an inflammatory response, in the form of IL-1β release, in macrophages that have been primed with the Toll-Like Receptor 4 (TLR4) agonist, lipopolysaccharide (LPS) and that CS can attenuate this response. We further demonstrated that this attenuation by CS occurred upstream of the inflammasome (2).
Objectives Determine if this inhibition is acting at the level of NF-κB activation.
Methods The THP1-Lucia NF-κB reporter monocyte cell line (InvivoGen) was grown as recommended. 105 cells/well were plated along with varying concentrations of CS (0, 10–200 μg/ml) and either 1 μg/ml LPS or 107 heat killed listeria monocytogenes (HKLM). In a 2nd experiment, cells were set up similarly but with the addition of 10 μg/ml ultralow molecular weight (ULMW) HA. In a 3rd experiment cells were set up as above with ULMW HA but substituting a caspase-1 inhibitor (EMD Millipore,10 μM) for CS. After 24 hours, media were tested for the NF-κB activity using QUANTI-Luc reagent (Invivogen) and for cell viability using PrestoBlue reagent (Invitrogen).
After normalizing for cell viability, all results were expressed as fold change from the media only control. One-way ANOVA with Dunnett's post-hoc test and inhibition dose response curves were performed.
Results NF-κB activity increased significantly (p<0.0001) in THP1-Lucia NF-κB reporter monocytes treated with either TLR4 or TLR2 agonists (i.e. 1 μg/ml of LPS or 107 HKLM bacteria). CS (50–200 μg/ml) produced a dose dependent (p=0.02 to p=0.0005) reduction in NF-κB activity in LPS treated cells. In HKLM treated cells, CS (100–200 μg/ml) also produced a similar dose dependent (p=0.004 to p=0.01) reduction in NF-κB activity. Dose response curves indicated a 50% inhibition concentration (IC50) of 146 μg/ml CS for LPS (●) stimulated cells and 64 μg/ml for HKLM (▲) stimulated cells (Fig. 1A). HA fragments alone produced no increase in NF-κB activity but were found to have an additive effect through either the TLR4 or TLR2 pathway (i.e. LPS or HKLM). The combination of TLR activation and HA produced greater NF-κB activity than either LPS or HKLM alone (p<0.0001**** and p=0.0017***, Fig. 1A). CS (200 μg/ml) was still an effective inhibitor under these conditions (p=0.045* and p=0.02**, Fig. 1B). Inhibiting caspase-1 activity, and thus blocking IL-1β production, completely blocked the additive effect of HA fragments (↓, Fig. 1C). Caspase inhibition had no effect on LPS or HKLM alone.
Conclusions The inflammatory response induced by HA fragments in the presence of TLR2 or TLR4 activation can be attenuated by CS at physiologically achievable concentrations via inhibition of NF-κB activation. HA's synergistic effect on NF-κB is due to positive feedback through inflammasome mediated active IL-1β production. These data provide a plausible explanation for the clinical effects of CS in OA.
du Souich P, Pharmacol Ther 2014, 142:362.
Stabler T, Osteoarthritis Cartilage 2015, 23:A263.
Disclosure of Interest T. Stabler Grant/research support from: Bioiberica, E. Montell Employee of: Bioiberica, J. Vergés Employee of: Bioiberica, V. Kraus Grant/research support from: Bioiberica