Macrophage migration inhibitory factor is involved in antineutrophil cytoplasmic antibody-mediated activation of C5a-primed neutrophils

Background C5a is important for antineutrophil cytoplasmic antibody (ANCA)-mediated activation of neutrophils. The present study aimed to assess the role of macrophage migration inhibitory factor (MIF) in ANCA-mediated activation of C5a-primed neutrophils. The effects of MIF on ANCA-mediated neutrophil respiratory burst and degranulation were determined. In addition, the effect of a MIF antagonist on the activation of C5a-primed neutrophils was assessed. Results MIF treatment resulted in increased membrane proteinase-3 (mPR3) expression on neutrophils and enhanced myeloperoxidase (MPO) amounts in neutrophil culture supernatants. The concentration of MIF was significantly higher in the neutrophils supernatant primed with C5a (negative control: 14.2 ± 1.16 ng/ml; C5a: 45.8 ± 2.8 ng/ml, P < 0.001 vs. negative control; C5a + IgG: 44.8 ± 1.93 ng/ml, P < 0.001 vs. negative control; C5a + MPO-ANCA: 73.0 ± 5.5 ng/ml, P < 0.001 vs. C5a; and C5a + PR3-ANCA: 69.4 ± 5.35 ng/ml, P < 0.001 vs. C5a). MIF primed neutrophils to undergo respiratory burst and degranulation in response to ANCA. Indeed, mean fluorescence intensity (a measure of respiratory burst) was significantly higher in MIF-primed neutrophils activated with MPO-ANCA-positive IgG or PR3-ANCA-positive IgG compared with non-primed neutrophils. Meanwhile, a MIF antagonist reduced oxygen radical production in C5a-primed neutrophils treated with patient-derived ANCA-positive IgG. Conclusions MIF can prime neutrophils to undergo ANCA-mediated respiratory burst and degranulation. Blocking MIF resulted in reduced ANCA-mediated activation of C5a-primed neutrophils. These findings indicated that the interaction between MIF and C5a may contribute to ANCA-mediated neutrophil activation.

The current hypothesis is that AAV pathogenesis involves neutrophils. Indeed, these cells constitute the first line of defense against bacteria and fungi. Neutrophils kill target cells using a combination of phagocytosis, oxygen radicals produced through respiratory burst, and the release of intracellular granules containing antimicrobial and inflammatory factors (degranulation) [4]. Lactoferrin is one of the iron-binding glycoproteins found in neutrophil granules, and can be used as a marker of neutrophil degranulation [5,6]. Previous preclinical studies showed that ANCAs are able to stimulate respiratory burst and degranulation in primed neutrophils. The above effects are considered important contributors to the development of AAVs [7][8][9][10][11]. Indeed, the pathogenic potential of ANCAs has been demonstrated in a mouse model of vasculitis induced by anti-MPO antibodies [12]. Furthermore, animal studies demonstrated that neutrophils are the main cellular effectors in AAVs [12,13]. Nevertheless, previous clinical and in vivo studies have implicated different pathways in the pathogenesis of AAVs [14][15][16][17][18]. It was shown that priming of neutrophils by C5a is necessary for an ANCAinduced respiratory burst to occur [19]. Although therapies targeting C5a have shown some initial promise [20], our knowledge of the key molecules and physiological events activating C5a-primed neutrophils by ANCAs is still limited at best.
Macrophage migration inhibitory factor (MIF) represents a proinflammatory cytokine with important roles in the innate and adaptive immune systems [21]. The cytoplasm of human nucleated cells contains MIF, which is released in response to cellular stress, leading to the activation of the CD74/CD44 receptor complex [21][22][23][24]. Hence, it was hypothesized that MIF plays important roles in a number of diseases, including chronic renal disease [25], asthma [26], rheumatoid arthritis [27], inflammatory bowel diseases [28,29], systemic lupus erythematosus [30,31] and crescentic glomerulonephritis [30]. MIF enhances the chemotactic responses and survival of neutrophils [32,33], and is therefore involved in AAVs [34,35]. C5a stimulates neutrophils to produce MIF during sepsis, and C5a receptor inhibition decreases the release of MIF during the initial stages of sepsis [36].
It was previously reported that active AAV cases have high plasma levels of MIF, while MIF-primed neutrophils could be stimulated by MPO-ANCA-positive IgG or PR3-ANCA-positive IgG to undergo respiratory burst and degranulation [37]. Nevertheless, MIF's function in C5a-primed neutrophils remains unknown. Based on the above findings, we hypothesized that MIF released from C5a-primed neutrophils would boost neutrophil activation and contribute to ANCA-induced respiratory burst and degranulation in these cells. Therefore, the present study assessed the effect of MIF on ANCA-associated activation of C5a-primed neutrophils.

IgG preparation
Cases of active MPO-ANCA-or PR3-ANCA-positive primary AAV (none had both ailments) were recruited, and ANCA-positive IgG molecules were extracted from their plasma [38,39]. Healthy volunteers were recruited, and normal IgG was obtained [40,41]. We controlled the amount of ANCA-positive IgG (total IgG), and the amount of ANCA-IgG was kept consistent for each individual patient in the experiment. LPS amounts in ANCA-positive IgG samples were < 0.1 ng/ml.
This study followed the Declaration of Helsinki, and had approval from the Clinical Research Ethics Committee of the Affiliated Hospital of Inner Mongolia Medical College. Each participant or donor provided written informed consent.

Membrane expression of PR3 and MPO in neutrophils
PR3 and MPO levels on the neutrophil surface were examined by flow cytometry. C5a (100 ng/ml for 15 min at 37°C; Biovision, USA) primed and control neutrophils were administered MIF (50 ng/ml; Sigma-Aldrich, USA) or not (buffer without MIF) for 30 min at 37°C. All subsequent procedures were carried out on ice; all washes were performed with HBSS+/+ containing 1% bovine serum albumin (BSA). Heat-denatured goat IgG (0.5 mg/ ml) was incubated with the cells for 15 min for Fcγ receptor saturation. The neutrophils underwent staining for 30 min with mouse monoclonal IgG1 antibodies targeting human PR3 (WGM2; Abcam, UK) and MPO (2C7; Abcam), respectively, or control IgG1 antibodies (BioLegend, USA). This was followed by incubation with phycoerythrin (PE)-conjugated goat anti-mouse antibodies (Abcam) and 0.5 mg/ml heat-treated goat IgG. The expression levels of ANCA antigens were evaluated flow-cytometrically based on PE fluorescence on a BD FACScan (Becton Dickinson, USA). A total of 10,000 cells were analyzed per specimen. PR3 and MPO levels were assessed as mean fluorescence intensity (MFI) values of specific signals. In certain experiments, neutrophils underwent pre-treatment with the small-molecule MIF antagonist (S,R)3-(4-hydroxyphenyl)-4,5-dihydro-5isoxazole acetic acid methyl ester (ISO-1; Tocris Bioscience, USA) [42,43] or vehicle (dimethyl sulfoxide, DMSO) prior to other interventions [44].

Measurement of MPO levels in cell culture supernatants
A commercial enzyme-linked immunosorbent assay (ELISA) (USCNK, China) kit was used for measuring MPO levels in the supernatants of neutrophils treated with MIF and/or C5a, as directed by the manufacturer. Absorbance was measured at 450 nm on a spectrophotometer.

Measurement of MIF levels in cell culture supernatants
A commercial kit (Abcam, UK) was used for measuring MIF levels in the supernatant of neutrophil cultures treated with C5a (100 ng/ml) and ANCA-positive IgG or buffer, according to the manufacturer's instructions. The absorbance was measured at 450 nm.

Respiratory burst assessment
The production of reactive oxygen species was estimated based on dihydrorhodamine (DHR, non-fluorescent) oxidation into rhodamine (green fluorescence). Neutrophils were administered 0.05 mM DHR123 (Sigma-Aldrich) for 10 min at 37°C, with 2 mM NaN 3 supplemented for preventing intracellular degradation of H 2 O 2 . Then, the cells were sequentially administered MIF (50 ng/ml) or C5a (100 ng/ml) [45], and normal or patient-derived ANCApositive IgG (200 μg/ml) for 1 h at 37°C. Reactions were terminated by addition of 1 ml of ice-cold HBSS containing 1% BSA. Data (10,000 cells per sample) were collected by flow cytometry on a BD FACScan. MFI values determined for all experimental conditions represented the amounts of reactive oxygen species produced [40,41].

ANCA-induced degranulation
Neutrophils treated with MIF and/or C5a were administered MPO-ANCA-positive, PR3-ANCA-positive, or normal IgG for 1 h, with control cells treated with the buffer. Some groups of cells also underwent pretreatment with ISO-1 or DMSO (15 min on ice). Supernatant lactoferrin content was detected with an ELISA kit (USCNK, China), as directed by the manufacturer.

Statistical analysis
The Shapiro-Wilk test was employed to assess data normality. Quantitative data are mean ± standard deviation (SD) (normal distribution) or median and range (skewed distribution). Student's t test and the Mann-Whitney U test were used to compare data with normal and nonnormal distributions, respectively. P < 0.05 indicated statistical significance. All analyses were carried out with SPSS 16.0 (SPSS, USA).
Degranulation levels were evaluated by determination of lactoferrin amounts in supernatants from neutrophils treated with MIF and ANCA-positive IgGs. In comparison with unstimulated cells, MIF-treated neutrophils exposed to MPO-ANCA-positive IgG (1515.0 ± 44.94 ng/ ml vs. 499.2 ± 18.64 ng/ml, P < 0.001) or PR3-ANCApositive IgG (1613.0 ± 49.52 ng/ml vs. 499.2 ± 18.64 ng/ ml, P < 0.001) showed markedly increased lactoferrin amounts in supernatants. Although degranulation was somewhat stimulated in neutrophils administered ANCA-IgG only, normal IgG only or MIF combined with normal IgG, further significant increases were observed following exposure to MPO-ANCA-or MPO-ANCA-positive IgG. In contrast, degranulation levels did not differ between neutrophils treated with normal IgG alone and cells administered MIF combined with normal IgG (Fig. 2c and d).

Discussion
Neutrophils are an important cell type involved in AAV [12,13]. ANCAs probably have a direct role in the pathogenesis of AAVs by stimulating respiratory burst and degranulation in activated neutrophils [7][8][9][10][11]. Clinical and in vivo data indicate complement activation through the alternative pathway is also critical in AAV development [14-19, 44, 46, 47]. Among others, C5a can prime neutrophils for respiratory burst induced by ANCA [19]. Consistently, we previously demonstrated that translocation of ANCA antigens and ANCA-related respiratory burst in neutrophils are enhanced by recombinant C5a [45]. MIF is an immunomodulatory factor that regulates chemotaxis and survival in neutrophils [32,33], and is involved in AAV [34,35] and other autoimmune diseases [26][27][28][29][30][31]. Notably, elevated levels of plasma MIF were reported in patients with active AAV [37]. Furthermore, respiratory burst and degranulation by MIF-treated neutrophils are stimulated by MPO-ANCA-and PR3-ANCA-positive IgGs [37]. A previous study showed that C5a induces MIF production by neutrophils in sepsis [36], although the exact interaction between C5a and MIF as well as their roles in AAV remain mostly unknown. Therefore, the current study aimed to assess MIF's role in the activation of C5a-primed neutrophils by ANCAs.
The present study indicated that neutrophils are an important source of MIF, and therefore may have a critical function in MIF release in pathologic conditions such as AAV, corroborating previous findings. MIF contributes to the pathogenesis of LPS-induced acute lung injury (ALI), and could be a therapeutic target in ALI [48]. Of note, MIF could also be a therapeutic target in a number of immune-related pathologies such as asthma [26], rheumatoid arthritis [27], chronic colitis [28], ulcerative colitis [29], glomerulonephritis [30], and lupus erythematosus [31], highlighting the possible clinical significance of the present study.
A previous report showed that patients with GPA or MPA have high MIF levels compared with those in remission [35]. Furthermore, a recent study by our group revealed that almost all patients with active MPA have high plasma levels of MIF [37]. Increased translocation of ANCA antigens is an important characteristic of neutrophils primed with MIF. ANCA-positive IgGs from patients with MPA can be used to further prime neutrophils and induce respiratory burst and degranulation. Therefore, we hypothesized that MIF promotes inflammation and disease activity in AAV.
MIF production after stimulation with recombinant human C5a and ANCA in neutrophils could be an important step in the initial stages of AAV. As shown above, ISO-1, a MIF antagonist, could significantly attenuate the effects of ANCA on neutrophils. Interestingly, C5a and MIF are proinflammatory factors that interact during ANCA-induced neutrophil activation. The above results are supported by a study of sepsis demonstrating that blockade or absence of the C5aR leads to a significant reduction in MIF biosynthesis [36].
Limitations of this study should be noted. First, direct measurements of MIF in cell culture supernatants were not performed. Secondly, residual C5a in cell supernatants may have contributed to the effects observed in supernatant-treated neutrophils. Thirdly, only one antagonist was used to inhibit MIF's function, and the results should be confirmed using another antagonist or the siRNA technology. Fourthly, the molecular pathways underlying the interactions among C5a, MIF and ANCAs were not explored. Therefore, additional studies are required to confirm the current findings.