We thank Dr. Xu and colleagues for their interest in our recent study and for positioning our findings in the context of current knowledge of the function of TLR-2 in RA. In full agreement with our observations, Huang et al recently reported that K/BxN serum–induced arthritis and structural joint damage are exacerbated in TLR-2–deficient mice ([1]). The 2 studies agree on the lack of transcriptional regulation of FcγRs by TLR-2 in murine macrophages. In our hands, quantitative real-time polymerase chain reaction also confirmed unaltered messenger RNA expression of various FcγRs in synovial tissue as well as splenocytes from naive TLR-2−/− mice (Abdollahi-Roodsaz S, et al: unpublished observations).

For determination of cell surface protein expression, we used specific monoclonal antibodies differentially detecting the 4 distinct FcγR isoforms, and very importantly, discriminating between FcγRIIB and FcγRIII (clone K9.361 anti-Ly17.2 for FcγRIIB [[2, 3]] and clone 275003 for FcγRIII [[4, 5]]). This differs from the approach used in the study by Huang et al, in which antibodies reacting to both CD32 and CD16 (FcγRII and FcγRIII, respectively) were used. By applying isoform-specific antibodies, we identified a significant reduction in cell surface expression of the inhibitory FcγRIIB in TLR-2−/− peritoneal as well as bone marrow–derived primary macrophages (Figure 1A).

Figure 1.

A, Fcγ receptor IIB (FcγRIIB) expression on bone marrow–derived macrophages. B, Up-regulation of FcγRIIB expression by the Toll-like receptor 2 (TLR-2) agonist Pam3Cys. WT = wild-type.

The suggested link between TLR-2 and FcγRIIB expression is also clearly supported by the fact that stimulation of murine bone marrow–derived macrophages with the TLR-2 agonist Pam3Cys increases both the percentage of FcγRIIB-positive cells and the intensity of FcγRIIB expression per cell (Figure 1B).

The macrophage production by cytokines observed in these studies is presumably affected by the distinct types of FcγR triggers used, i.e., bovine serum albumin (BSA)/anti-BSA immune complexes in our study as opposed to plate-coated IgG in the Huang study. Although side-by-side comparisons have not been reported, we presume that the FcγR isoform to be triggered and the ultimate cellular response are affected by the type of the stimulation used; plate-coated IgG may prefer the high-avidity FcγRI, whereas immune complexes may mainly interact with FcγRII, which serves for their phagocytosis ([6]).

Thus, 2 potential mutually nonexclusive mechanisms are suggested for TLR-2 to down-modulate immune complex–induced arthritis: induction of the inhibitory FcγRIIB and induction of the antiinflammatory cytokine IL-10. The notably identical arthritis phenotypes observed in the 2 studies increase our appreciation of the protective function of TLR-2 during the autoantibody-driven effector phase of the disease. In this context, we envision that it is the cell-specific and timely stimulation of TLR-2, rather than its neutralization, that should be considered as a potential therapeutic approach in RA. We look forward to future research elucidating the relevance of each of the mechanisms to the observed disease phenotype.


Supported by The Netherlands Organization for Scientific Research (Veni grant 916.12.039) and the European Innovative Medicines Initiative (BTCure project grant 115142-2).