Among ANCA, those targeting PR3 (PR3-ANCA) have a strong and specific association with WG . Besides their significance as seromarkers, a pathogenic role has been proposed for these autoantibodies in relation to their capacity to activate leukocytes in vitro [9, 10, 12]. In the present study, an alternative approach was chosen to define the priming effect of ANCA against PR3 (PR3-ANCA) from WG patients on inflammatory leukocyte functions: human PBMCs were preincubated with substimulatory concentrations of PR3-ANCA and possible augmented cell activation by various microbial PAMPs was examined. As mentioned above, Hattar et al.  first reported that anti-PR3 Abs primed human leukocytes for the enhanced production of inflammatory cytokines upon stimulation with TLR4-agonistic LPS or TLR2-agonistic lipoteichoic acid in a CD14-dependent manner. In our previous study , we demonstrated that incubation with murine anti-PR3 Abs significantly up-regulated the expression of various TLRs (TLR2, 3, 4, 7, and 9, NOD1, and NOD2) in addition to CD14 in human monocytic THP-1 cells. In this context, however, Hattar et al.  found no increase in the expression of TLR2 and TLR4 by FACS analysis after treatment with anti-PR3 Abs, and their priming effects were CD14-dependent, but TLR2- and TLR4-independent. It is unclear why they did not find an increase in the expression of TLR. The difference in staining methods may be one reason; in their study, the cells were directly stained with PE-labeled Abs targeting TLR2 or TLR4, whereas in our study the cells were stained with various TLR or NOD Abs, followed by a FITC-conjugated secondary Ab.
In our previous study , we examined the priming effects of murine monoclonal anti-human PR3 Abs on human monocytic THP-1 cells. We were interested in whether similar effects would be observed with human PBMCs and with serum and purified Ig isolated from WG patients and control sera. In this study, we demonstrated that PR3-ANCA serum from WG patients up-regulated the expression of various TLRs and NODs in human PBMCs (Fig. 1) and primed PBMCs to produce proinflammatory cytokines such as IL-6, IL-8, MCP-1, and TNF-α via TLRs and NODs (Figs. 2 and 3). We solely used chemically synthesized PAMPs, because natural microbial cell surface preparations are inevitably contaminated with minor bioactive components that might have confused the results. Therefore, these results clearly indicated that PR3-ANCA primed TLR2-, 3-, 4-, 7-, and 9-, NOD1-, and 2-dependent cell activation in PBMCs. Although ANCA is autoantibodies characteristic of vasculitis diseases, PR3-ANCA and MPO-ANCA are known to be negative in some kinds of vasculitis such as polyarteritis nodosa and aortitis syndrome (Table 1). It must be noted here that four ANCA-negative vasculitis sera did not have any priming effects on PBMCs in PBMCs, contrary to two ANCA-positive sera (Fig. 3). Purified IgG from WG serum also significantly primed PBMCs to produce proinflammatory cytokines (Fig. 4). These findings certainly strengthen our conclusion and the relevance of our work to the human situation.
It is conceivable that microbial components exhibit powerful immuno-adjuvant activities through the activation of antigen-presenting cells via TLR and NOD pathways against various antigens, including autoantigens, which in turn might induce severe autoimmune diseases. In fact, in animal models, NOD2-agonistic MDP induces autoimmune diseases such as experimental encephalomyelitis, orchitis, uveoretinitis, thyroiditis, and polyarthritis . In contrast, our study showed a completely different model for the possible induction of autoimmune diseases by microbial components. Namely, microbial TLR- and NOD-ligands possibly induced tissue destruction through excessive inflammatory responses by triggering the activation of PR3-ANCA-primed cells. The background for the present experimental rationale is based on the fact that the disease activity in WG, which is paralleled by a rising PR3-ANCA titer [4, 8], appears to be triggered by microbial infections [9, 35–37]. As suggested by Condliffe et al.  and Hallett and Lloyds , in vivo, the response of leukocytes to microbial depends on the state of cellular activation, varying from "dormant" via "primed" to "fully activated". Priming is a key mechanism involved in the regulation of the leukocyte-dependent host defense. Although they do not directly activate neutrophil and monocyte functions, priming agents induce a "sensitization" of the leukocytes for subsequent stimulation with naturally occurring agonists, such as microbe-derived products. In addition to the important roles of ANCA in the regulation of inflammatory leukocyte functions, the phenomenon of leukocyte priming by PR3-ANCA may also be relevant to the pathogenesis of inflammation. Indeed, in active WG, neutrophils and monocytes display a phenotype attributable to a state of cellular preactivation with the enhanced surface expression of activation markers such as CD11b and CD64 [11, 40].
PR3-ANCA, being only weak direct activators of monocytes and neutrophils to release cytokines per se, exert a definite priming effect on these leukocytes, enhancing their responsiveness to secondary stimulation with microbial PAMPs. The up-regulation of various pattern recognition receptors, including TLRs and NODs, acting as respective PAMPs, was characterized as one mechanism underlying the ANCA-elicited priming response. Such cooperation between PR3-ANCA and microbial PAMPs may well trigger exacerbations of disease activity during infections and contribute to the persistence of inflammatory lesions, which might be a novel model for the pathogenesis of autoimmune diseases.