Platelets are activated in ANCA-associated vasculitis via thrombin-PARs pathway and can activate the alternative complement pathway

We analyzed the platelet activation parameters in AAV patients by flow cytometry, including CD62P and platelet adhesion to neutrophils, lymphocytes, and monocytes.

Blood samples were freshly collected from 31 patients with active AAV before the initiation of immunosuppressive therapy. Blood samples of 27 patients with AAV, who achieved complete remission at least 3–6 months after the initiation of immunosuppressive therapy, were also collected at their regular ambulatory visits. All the AAV patients in the current study met the 2012 revised Chapel Hill Consensus Conference criteria for AAV [25]. Disease activity was evaluated by the Birmingham Vasculitis Activity Score (BVAS) [26]. “Remission” was defined as “absence of disease activity attributable to active disease qualified by the need for ongoing stable maintenance immunosuppressive therapy” (complete remission), or “at least 50% reduction of disease activity score and absence of new manifestations” (partial remission) [27]. The BVAS levels of all the AAV patients in remission stage in this study were 0.

Treatment protocols have been described previously [28, 29]. In brief, induction therapy included corticosteroids in combination with cyclophosphamide (CTX). Oral prednisone was prescribed at an initial dosage of 1 mg/kg/day for 4–6 weeks, with reducing doses over time to 12.5–15 mg by 3 months. CTX was administered by daily oral dose of 2 mg/kg/day or intravenously 0.5 g/m² every month. Patients with acute renal failure or pulmonary hemorrhage received three pulses of intravenous methylprednisolone (7–15 mg/kg/day) before the standard induction therapy. Patients with severe pulmonary hemorrhage or acute renal failure requiring dialysis at diagnosis received additional plasma exchanges. For maintenance therapy, daily oral azathioprine (AZA) was given (2 mg/kg/day) for at least 2 years.

Additionally, the blood samples of 40 healthy blood donors were also collected as normal controls. No VTE was observed by the time of blood collection, and none of the patients or healthy donors was receiving anticoagulant or antiplatelet therapies.

Citrate anticoagulated blood from patients and healthy donors were dispensed into an antibody cocktail containing CD41a APC (BD Pharmingen, San Jose, CA, USA), CD62P BV421 (BD Horizon™) and CD45-APC-H7 (BD Pharmingen) incubated for 15 min and analyzed by flow cytometry. CD41a-APC was used to gate platelets. As CD62P is a typical marker of platelet α-granule degranulation and lysosome degranulation, it serves as a marker of platelet activation [30‐32]. Neutrophils, lymphocytes and monocytes were identified based on their characteristic light scatter, and differential CD45 expression [33].

Platelets were isolated according to the methods described previously [30]. Briefly, blood was drawn from healthy donors, using a 21G or larger bore needle and a light tourniquet, into a sodium citrate vacutainer. Platelet-rich plasma (PRP) was formed by centrifugation at 200 × g for 15 min and then washed using a citrate wash buffer (11 mM glucose, 128 mM NaCl, 4.3 mM NaH₂PO₄, 7.5 mM Na₂HPO₄, 4.8 mM Na₃C₆H₅O₇, 2.4 mM C₆H₈O₇, 0.35% BSA, 50 ng/mL prostaglandin PGE1, pH 6.5) [30]. Platelets were pelleted by centrifugation at 1200 × g for 10 min and resuspended in modified Tyrode’s buffer (Leagene), as previously described [34].

Heparinized plasma samples from 14 active AAV patients and 14 patients with minimal change disease (MCD, as the disease control) were collected immediately after centrifugation at 2000 g for 15 minutes at 4 °C. Then, the plasma was aliquoted and stored at −80 °C until use. Repeated freeze/thaw cycles were avoided.

Platelets that were resuspended in modified Tyrode’s buffer were incubated at 37 °C for 15 min with plasma from patients at a final concentration of 50%. This concentration was chosen because it demonstrated the highest platelet activation in our pilot experiments. According to the previous study, 1 mM Gly-Pro-Arg-Pro (GPRP, Sigma-Aldrich, St. Louis, MO, USA) was added to prevent fibrin polymerization [35]. To investigate the potential role of the thrombin-PARs pathway in platelet activation by plasma from AAV patients, plasma was pretreated with 20 μM D-Phe-Pro-Arg-CH₂Cl (PPACK, EMD Millipore, Billerica, MS, USA), a thrombin inhibitor, and the cells were pretreated with 200 μM of a PAR1 inhibitor, SCH79797 (C₂₃H₂₅N₅.2HCl, Abcam, Cambridge, MA, USA), as previously described [36, 37]. Thrombin receptor-activating peptide (TRAP, Sigma-Aldrich), a platelet agonist, was used as a positive control for platelet activation [24].

Detection of complement activation was conducted according to the methods described previously [34, 38]. Briefly, non-activated or activated platelets were washed once with Tyrode/PGE buffer (Tyrode’s buffer with 1 μM PGE1 obtained from Sigma-Aldrich), centrifuged at 1200 g for 10 min at room temperature, and incubated with 50% plasma in 0.8 mM Mg-EGTA buffer (10 mM EGTA with 0.8 mM MgCl₂) to assess complement proteins activation. Instead of reaction buffers, 10 mM EDTA was added, which binds calcium and magnesium ions in the plasma, to restrain potential complement activation as the negative control. The samples were incubated for 60 min at 37 °C, and then the reactions were stopped by 3 μl EDTA/PGE buffer (EDTA buffer with 1 μM PGE1). After centrifugation at 1200 g for 10 min at room temperature, the supernatant was stored and quantitated with Quidel enzyme-linked immunosorbent assays (ELISAs) for human C3a, C5a, sC4d, and sC5b-9 (Quidel Corporation, San Diego, CA, USA). The remaining platelets were washed with Tyrode/PGE buffer to reduce nonspecific adhesion of complement molecules, then incubated with antibodies to CD41, C3c and C5b-9, and analyzed by fluorescent-activated cell sorting (FACS), as described above.

The study protocol complied with the Declaration of Helsinki and was approved by the ethics committee of Peking University First Hospital. Written informed consent was obtained from each participant.

Results were presented as the mean ± SD (for normally distributed data) or median and interquartile range (IQR, for non-normally distributed data), as appropriate. Difference in the paired samples was assessed using paired t test or Wilcoxon’s signed rank tests, as appropriate. Difference between unpaired groups was assessed using t tests or nonparametric tests, as appropriate. Difference between multiple groups was assessed using the one-way ANOVA. Difference between non-independent data was assessed using generalized estimating equations. Statistical analyses were performed with the SPSS statistics software, version 15.0 (Chicago, IL, USA). The p values less than 0.05 were considered significant.

Platelets from the patients and healthy donors were stained for CD62P. Among the 31 active AAV patient samples that were analyzed for CD62P expression, 19 were male and 12 were female, with an age of 59.4 ± 13.2 years at diagnosis. Twenty-seven and four patients were MPO-ANCA and PR3-ANCA positive, respectively. The initial serum creatinine (Scr) values were 456.8 ± 258.6 μmol/L. The initial BVAS levels were 19.5 ± 6.2 (Table 1).

The percentages of CD62P-expressing platelets in AAV patients in active stage were significantly higher than that in remission and the normal controls (15.8% [5.1%, 26.7%] vs. 8.9% [4.2%, 12%], p < 0.05; 15.8% [5.1%, 26.7%] vs. 5.7% [3.1%, 6.5%], p < 0.0001, respectively). Furthermore, among the 16 AAV patients with sequential blood samples from both active stage and remission, the percentages of CD62P-expressing platelets were significantly higher in AAV patients in active stage than those in remission (20.4% [12.7%, 28.9%] vs. 8.0% [2.5%,11.4%], p < 0.0001), and all these 16 patients had a decrease in the percentages of CD62P-expressing platelets in remission, compared with that in active stage (Fig. 1a), which was consistent with the above-mentioned results. There was no significant difference of the percentages of CD62P-expressing platelets among different induction therapy regimens in AAV patients.

Leukocytes from AAV patients and healthy blood donors were stained for CD45 and CD41a (the marker of platelets) in 22 of the above-mentioned 31 active AAV patients with a sufficient sample. Neutrophils, lymphocytes and monocytes were identified based on their characteristic light scatter and differential CD45 expression. The percentages of CD41a-positive neutrophils, lymphocytes and monocytes, i.e., neutrophil-, lymphocyte- and monocyte-platelet aggregates (abbreviated as NPA, LPA and MPA, respectively) were analyzed separately. The NPA, LPA and MPA percentages in the AAV patients in the active stage were significantly higher than those in the AAV patients in remission (19.3% [5.8%,27.2%] vs. 6.3% [4.6%, 7.2%], p < 0.0001; 31.8% [15.8%, 46.3%] vs. 14.2% [8.4%, 20.9%], p < 0.001; 8.8% [5.9%, 12.3%] vs. 5.9% [4.3%, 7%], p < 0.01; respectively) and in the normal controls (19.3% [5.8%,27.2%] vs. 7.7% [5.4%,9.5%], p < 0.0001; 31.8% [15.8%, 46.3%] vs. 14.5% [10.8%, 18.7%], p < 0.0001; 8.8% [5.9%, 12.3%] vs. 4.2% [2.1%, 6.1%], p < 0.0001; respectively) (Fig. 1b–d). Furthermore, the platelet-leukocyte aggregate levels in 11 AAV patients with sequential samples of both active stage and remission were compared. The neutrophil, lymphocyte and monocyte adhesion percentages were significantly higher in the active stage than in remission (26.4% [17.8%, 27.9%] vs. 6.6% [4.8%, 8.1%], p < 0.0001; 43.4% [31.2%, 58.9%] vs. 12% [7%,12.3%], p < 0.0001; 10.4% [7.6%, 12.3%] vs. 4.7% [4.1%,6%], p < 0.0001; respectively); all 11 patients had decreased NPA, LPA and MPA levels in remission compared with the active stage (Fig. 1b–d).

Collectively, these data suggest that platelets are activated in patients with active AAV.

We collected plasma samples from 14 active AAV patients before the initiation of immunosuppressive therapy for platelets stimulation. Among these 14 patients, 7 were male and 7 were female, with an average age of 52.0 ± 13.0 years at diagnosis. Seven and seven patients were MPO-ANCA and PR3-ANCA positive, respectively. The initial BVAS level was 20.6 ± 7.1. When the platelets were incubated with plasma from the active AAV patients, the CD62P-expressing platelet levels increased significantly compared with those incubated with plasma from the healthy donors (97.7 ± 3% vs. 1 ± 0.2%, p < 0.0001) (Fig. 2).

To evaluate the potential role of the thrombin-PARs pathway in platelet activation, thrombin inhibition and PAR1 blockage were performed. The plasma from active AAV patient was pretreated with 20 μM PPACK to inhibit its potential thrombin activity. PPACK pretreatment led to a significant decrease in the CD62P-expressing platelets after stimulation with plasma of AAV patients (97.7 ± 3.0% vs. 2.7 ± 1.0%, p < 0.0001). Furthermore, pre-incubation with 200 μM SCH79797 induced a significant decrease in the CD62P expression level after stimulation with plasma of AAV patients (97.7 ± 3.0% vs. 5 ± 1.4%, p < 0.0001) (Fig. 2).

Collectively, these results suggest that thrombin-PARs pathway contributes at least partially to platelet activation in AAV patients.

After the platelets were activated by the plasma of AAV patients, Ca/Mg-Tyrode buffer was used to measure complement activation via all three pathways, i.e., the classical, mannose-binding lectin (MBL) and alternative pathways. In the liquid phase, the complement fragments C3a, C5a and soluble membrane attack complex (MAC) sC5b-9 levels were significantly higher than those in the vehicle treatment (2,921.5 ± 295.7 ng/ml vs. 618.1 ± 87.6 ng/ml, p < 0.0001; 56.3 ± 1.8 ng/ml vs. 14.8 ± 1.0 ng/ml, p < 0.0001; 4425.3 ± 243.8 ng/ml vs. 1445 ± 67.2 ng/ml, p < 0.0001; respectively), whereas sC4d, a common factor of the classical and mannose-binding lectin pathways, was comparable to that in the vehicle group (7.0 ± 0.3 ng/ml vs. 7.4 ± 0.4 ng/ml, p = 0.629). On the platelets, levels of adhesion complement fragment C3c, a cleavage product of C3b and the C5b-9 complex, were significantly higher than in the negative control samples (23.1 ± 1.7% vs. 5.9 ± 0.8%, p < 0.0001; 22.7 ± 1.6% vs. 5.9 ± 0.8%, p < 0.0001; respectively) (Fig. 3).

Additionally, after platelets were activated by ANCA-positive plasma, Mg-EGTA buffer was used to selectively measure activation of the complement alternative pathway and not the classic or MBL pathways. In the fluid phase, the concentrations of the complement fragments, C3a, C5a and sC5b-9, were significantly higher than those in the vehicle samples (1420.5 ± 150.0 ng/ml vs. 618.1 ± 87.6 ng/ml, p < 0.0001; 33.8 ± 2.3 ng/ml vs. 14.8 ± 1 ng/ml, p < 0.0001; 3035.5 ± 267.6 ng/ml vs. 1445 ± 67.2 ng/ml, p < 0.0001; respectively). On the platelet surfaces, the C3c and C5b-9 complex levels were significantly higher than in the negative control samples (23.5 ± 1.6% vs. 5.5 ± 0.3%, p < 0.0001; 23.7 ± 1.7% vs. 5.5 ± 0.3%, p < 0.0001; respectively) (Fig. 4).

These results show that the alternative complement pathway is activated by activated platelets in AAV.