Enhanced formation and disordered regulation of NETs in myeloperoxidase-ANCA-associated microscopic polyangiitis

Abstract

Microscopic polyangiitis (MPA) is an ANCA-associated vasculitis that affects small vessels, especially renal glomeruli. We recently demonstrated that the abnormal formation and impaired degradation of neutrophil extracellular traps (NETs) may be crucially involved in the generation of myeloperoxidase (MPO)-ANCA and subsequent development of MPA. This study assessed the formation and regulation of NETs in patients with MPO-ANCA-associated MPA. Peripheral blood samples were obtained from 38 patients with MPO-ANCA-associated MPA, 23 patients with systemic lupus erythematosus (SLE), and 8 healthy controls. IgG eluted from MPO-ANCA-associated MPA sera demonstrated the highest ability to induce NETs, and this ability correlated with disease activity and paralleled ANCA affinity for MPO. Moreover, addition of recombinant human MPO to these IgG samples reduced NET induction. Additionally, MPO-ANCA-associated MPA sera exhibited lower rates of NET degradation that recovered partially upon depletion of IgG. The activity of DNase I, an important regulator of NETs, was also lower in MPO-ANCA-associated MPA and SLE sera. IgG depletion from MPO-ANCA-associated MPA sera partially restored the rate of NET degradation, and addition of DNase I synergistically enhanced this restoration. Addition of anti-MPO antibodies did not inhibit DNase I activity, and some MPO-ANCA-associated MPA sera contained anti-NET antibodies at levels not correlated with MPO-ANCA titers, suggesting the involvement of unidentified autoantibodies as well. The collective evidence suggests a vicious cycle involving MPO-ANCA and the regulation of NETs could be critically involved in the pathogenesis of MPO-ANCA-associated MPA.

Keywords: ANCA; immunology; pathology; vasculitis.

Figure 1.

IgG eluted from MPO-ANCA–associated MPA sera demonstrates the highest ability to induce NETs. Neutrophils (1×10⁶) from healthy volunteers were pretreated with 5 ng/ml TNF-α for 15 minutes at 37°C. The TNF-α–primed neutrophils were incubated with 250 μg/ml MPO-ANCA–associated MPA IgG (A–C) and control IgG (D–F) for 3 hours at 37°C. Representative figures are shown. (A and D) Cit H3. (B and E) DNA. (C and F) Overlay. Original magnification, ×400. (G) NET induction rate was evaluated by calculating the proportion of Cit H3–positive neutrophils on IF. (H) MPO-DNA complexes in the supernatants were measured by ELISA. Healthy, n=8; MPO-ANCA–associated MPA, n=38; SLE, n=23. **P<0.01; ***P<0.001.

Figure 2.

NET induction ability of MPO-ANCA–associated MPA IgG is absorbed by MPO. Neutrophils (1×10⁶) from healthy volunteers were pretreated with 5 ng/ml TNF-α for 15 minutes at 37°C. The TNF-α–primed neutrophils were incubated with 250 μg/ml mouse antihuman MPO monoclonal antibody (A–C) and MPO-ANCA–associated MPA IgG (D–F) with or without recombinant human MPO for 3 hours at 37°C. The concentrations of recombinant human MPO were 0 µg/ml (A and D), 1 µg/ml (B and E), and 100 µg/ml (C and F). NET formation was evaluated as DAPI-positive extracellular chromatin fibers. Representative figures are shown. Original magnification, ×400. (G) The NET induction ability of MPO-ANCA–associated MPA serum (n=5) was significantly decreased by addition of 100 µg/ml recombinant human MPO. **P<0.01.

Figure 3.

NET induction ability of MPO-ANCA–associated MPA IgG correlates with BVAS and ANCA affinity for MPO. (A) Correlation between NET induction ability and BVAS. (B) No correlation between NET induction ability and MPO-ANCA titer. (C) Comparison of NET induction ability between patients with low-affinity MPO-ANCA (n=5) and with high-affinity MPO-ANCA (n=4). *P<0.05. NS, not significant.

Figure 4.

MPO-ANCA–associated MPA sera exhibit lower rates of NET degradation than healthy controls. (A) Neutrophils (1×10⁶) from healthy volunteers were treated with 100 nM PMA for 3 hours at 37°C. The PMA-induced NET were incubated with 10% serum from healthy controls, patients with MPO-ANCA–associated MPA, and patients with SLE, or with PBS for 6 hours at 37°C. The residual NETs were recognized as DAPI-positive extracellular chromatin fibers. Representative figures are shown. Original magnification, ×40. (B) Comparison of NET degradation ability of serum among healthy controls (n=8), patients with MPO-ANCA–associated MPA (n=38), and patients with SLE (n=23). ***P<0.001.

Figure 5.

IgG depletion from MPO-ANCA–associated MPA sera partially restores the rate of NET degradation, and addition of DNase I synergistically enhances the restoration. (A) In this study, 15 MPO-ANCA–associated MPA serum samples that showed low ability for NET degradation (NET degradation rates<65%) were included. IgG was removed from the sera using the immunoabsorbent column. Thereafter, 10 U/ml DNase I was added to the IgG-depleted serum. After these preparations, the NET degradation ability of each sample was evaluated. Fifteen samples were classified into two groups: group 1, ability for NET degradation was recovered by IgG depletion; group 2, ability for NET degradation was not recovered by IgG depletion. (B) The NET degradation ability of MPO-ANCA–associated MPA serum (n=15) was significantly increased by the combination of IgG depletion and DNase I addition but not by the IgG depletion alone. **P<0.01. NS, not significant.

Figure 6.

DNase I activity in MPO-ANCA–associated MPA sera is lower than in healthy controls. (A) Comparison of serum DNase I activity among healthy controls (n=8), patients with MPO-ANCA–associated MPA (n=38), and patients with SLE (n=23). (B) No correlation between DNase I activity and BVAS in MPO-ANCA–associated MPA. ***P<0.001. NS, not significant.

Figure 7.

Some MPO-ANCA–associated MPA sera contain anti-NET antibodies at levels not correlated with MPO-ANCA titers. (A) Neutrophils from healthy volunteers were seeded on chamber slides (1×10⁶/ml) and then treated with 20 nM PMA for 2 hours at 37°C. After fixation with 4% PFA, the samples were washed with PBS and then incubated with 250 µg/ml IgG eluted from MPO-ANCA–associated MPA sera or healthy control sera for 1 hour at 37°C. After removal of unbound IgG, the samples were next allowed to react with 1:5000 dilution of FITC-conjugated antihuman IgG antibodies (Santa Cruz Biotechnology, Dallas, TX) for 1 hour at 37°C. In some patients in group 1, but not in group 2 or healthy controls, human IgG bound to the NETs was seen. Representative figures are shown. Original magnification, ×100. (B) The amount of anti-NET antibodies was represented as the rate of IgG-binding area in NETs. No significant correlation between the amount of anti-NET antibodies and the MPO-ANCA titer. NS, not significant. (C) The PMA-induced NETs were treated by 1 U/ml DNase I with 10 µg/ml or 100 µg/ml antihuman MPO antibody (Ab) at room temperature. Before and 10, 20, and 30 minutes after the treatment, the residual NETs were determined using ImageJ software. No inhibitory effect of the anti-MPO antibody on the DNase I activity was seen.