In vivo effects of cyclic administration of 15-deoxyspergualin on leucocyte function in patients with Wegener's granulomatosis

Abstract

15-Deoxyspergualin (DSG) is an alternative treatment modality for Wegener’s granulomatosis (WG) patients refractory to conventional treatment. Nevertheless, it is unclear how DSG modulates disease activity in these patients. This study was conducted to investigate which parameters of adaptive and acquired immunity were influenced during two subsequent cycles of DSG treatment. Emphasis was put upon T cell and monocyte activation, neutrophil function and surface expression of proteinase-3 (PR-3). Anti-CD3/anti-CD28 and interleukin (IL)-15/IL-7-mediated T cell proliferation were assessed by fluorescence activated cell sorter (FACS) analysis using carboxyfluorescein succinimidyl ester (CSFE) labelling. Interferon (IFN)-γ and IL-10 production were determined in the supernatants of these cultures by enzyme-linked immunosorbent assay. Monocyte activation was assessed in lipopolysaccharide (LPS)-stimulated whole blood, using tumour necrosis factor (TNF)-α as read-out. Neutrophil function was determined by measuring oxidative burst, chemotaxis and phagocytosis. T cell activation markers and PR3 expression were measured by FACS. All parameters were determined directly before and after each DSG cycle. Anti-CD3/anti-CD28-mediated T cell proliferation was reduced directly after DSG treatment. Directly before a subsequent cycle of DSG was started, T cell proliferation was increased. Similar findings were observed for IFN-γ and IL-10 production by T cells. DSG did not influence IL-15/IL-7-mediated T cell proliferation. LPS-mediated TNF-α production was also impaired directly after DSG treatment. No influence on T cell activation markers, neutrophil function and surface PR-3 expression was observed in peripheral blood of these patients. Our data demonstrate that DSG influences T cell and monocyte activation in a reversible fashion. Although DSG causes neutropenia in these patients, it does not influence neutrophil function.

Introduction

Wegener’s granulomatosis (WG) belongs to the group of anti-neutrophil cytoplasmic autoantibody (ANCA)-associated small-vessel vasculitis (AASV) and is characterized by a necrotizing inflammation of small blood vessels, particularly those of the respiratory tract, skin and nerves. In addition, WG is characterized by the occurrence of pauci-immune crescentic glomerulonephritis.

In the early 1970s the introduction of cyclophosphamide and oral corticosteroids in the treatment of active WG changed a formerly almost fatal disease into a treatable, more chronic-relapsing condition associated with a 5-year survival rate of 80% [1]. Despite enormous clinical improvement, cyclophosphamide is associated with high treatment-related morbidity and mortality [2]. Therefore, new treatment strategies are warranted, particularly in frequent relapsing or refractory patients and in those suffering from side effects of cyclophosphamide. Previously we have reported that treatment with the immunosuppressive drug 15-deoxyspergualin is successful in patients with refractory WG [3] and may also have beneficial effects during long-term treatment [4].

15-Deoxyspergualin (DSG) is a synthetic analogue of spergualin isolated from the culture filtrate of Bacillus laterosporus. DSG binds to chaperone proteins required for nuclear translocation of the transcription factor NF kappa B, thereby preventing cell activation. Because T cell proliferation and B cell maturation [5,6] are highly dependent upon nuclear translocation of NF kappa B, DSG gained much interest as a new immunosuppressive agent for the treatment of transplant rejection [7] and autoimmunity [8].

In general, it is believed that binding of ANCA and subsequent activation of neutrophils are involved in the pathogenesis of AASV. Although the role of myeloperoxidase (MPO)–ANCA in disease onset has been demonstrated recently in a mouse model [9], formal proof for a pathogenetic role for proteinase 3 (PR3)–ANCA is still lacking. In addition to neutrophils and ANCA, T cells are likely to be involved in disease manifestation because T cell-depleting therapies such as anti-thymocyte globulin (ATG) have shown to be beneficial in the treatment of WG [10]. Moreover, T cell activation has been demonstrated during phases of active disease and convalescence in AASV patients [11,12].

In the present study, we investigated whether DSG influences parameters of acquired and adaptive immunity by testing whether T cell proliferation and cytokine production, monocyte activation and neutrophil function were impaired in WG patients during the course of treatment.

Methods

Patients

The clinical characteristics of each patient are given in Table 1. All patients [two male, three female; mean age 48 (range 31–71) years] had histologically proven disease and were classified according to the Chapel Hill nomenclature as Wegeners’s granulomatosis. All patients were anti-PR3–ANCA and C–ANCA positive by enzyme-linked immunosorbent assay (ELISA) and indirect immunofluorescence as specified in Table 1. The activity of disease was evaluated according to clinical and radiological parameters and assessed according to the Birmingham Vasculitis Activity Score (BVAS). The clinical outcome of the patients has been described previously [4].

Table 1.

Clinical characteristics of all patients studied.

	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5
Age (years)	31	43	37	58	71
Gender	Male	Male	Female	Female	Female
Diagnosis	WG	WG	WG	WG	WG
Onset of disease	December 1999	August 1994	January 1997	August 1995	July 2000
ANCA-/antibody type	C-ANCA/PR-3	C-ANCA/PR-3	C-ANCA/PR-3	C-ANCA/PR-3	C-ANCA/PR-3
Cumulative organ involvement	Lung, kidney, ENT, joints	Lung, kidney, ENT, eye	Lung, ENT, joints, eye, PNS	Kidney, ENT, eye, bowel, joints, PNS	Lung, kidney, ENT, joints, PNS
BVAS*	0	0	0	0	0
Additional immunosuppression	GC	GC	GC	GC	GC
DSG since	03/03–05/04	05/98–09/99 07/00–08/01 10/01–present	12/02–present	10/98–03/00 04/02–04/04	11/2001–present
Immunosuppression before therapy with DSG	CYC 68 g, AZA	CYC 4,5 g, AZA, MMF, MTX, PE, IVIG	CYC 54 g, MTX, MMF, anti-TNF	CYC 15 g, AZA, MMF, anti-TNF, ATG	CYC 49 g, AZA

ANCA: anti-neutrophil cytoplasmic autoantibody; DSG: deoxyspergualin;

^*at time-point of study; WG: Wegener’s granulomatosis; PR-3: proteinase-3; BVAS: Birmingham Vasculitis Activity Score; ENT: ear, nose and throat; PNS: peripheral nervous system; GC: glucocorticosteroids; CYC: cyclophosphamide; AZA: azathioprine; MMF: mycophenolate mofetil; MTX: methotrexate; PE: plasma exchange; ATG: anti-thymocyte globulin; IVIG: intravenous immunoglobulin.

The study was approved by the local ethics committee and all patients gave informed consent.

Treatment protocol

DSG was given for 2–3 weeks [0·5 mg/kg daily by self-administered subcutaneous (s.c.) injection] with the aim of reaching a leucocyte [white blood cell (WBC)] nadir of 3000/µl. WBC were monitored at least twice weekly. At the nadir, treatment was discontinued for at least 2 weeks to allow leucocyte count to recover (WBC at least 4000/µl). Treatment was then resumed in the same manner [3].

Besides steroids, no other immunosuppressants were given (Table 1). Blood samples were drawn from each patient during two subsequent cycles, directly before DSG administration and on the last day of treatment. Each patient served as its own control. A standard differential blood count was performed at the time of the experiments to monitor neutrophil and monocyte counts.

Isolation of peripheral blood mononuclear cells (PBMC) and CD4⁺ T cells

PBMCs were prepared by gradient centrifugation using Ficoll-Hypaque (Amersham Biosciences, Freiburg, Germany). CD4⁺ T cells were isolated from PBMC by negative selection (Miltenyi Biotec, Bergisch-Gladbach, Germany). Overall purity of the isolated CD4⁺ T cells was above 95%.

Isolation of neutrophils from heparinized whole blood and staining of PR-3

Neutrophils were isolated from heparinized whole blood by Ficoll-Hypaque density gradient centrifugation followed by red blood cell sedimentation with dextrane 1% (Fluka, Seelze, Germany) and hypotonic erythrocyte lysis. Directly after isolation, neutrophils were stained with 10 µl of unlabelled monoclonal antibody Pelicluster ANCA (Sanquin, Amsterdam, the Netherlands) directed against PR-3. Then, 10 µl of fluorescein isothiocyanate (FITC)-labelled polyclonal goat anti-mouse immunglobulin antibody (Dako, Hamburg, Germany) was used as secondary antibody. PR-3 expression on unstimulated neutrophils was assessed by fluorescence activated cell sorter (FACS) analysis.

Proliferation assay

T cell proliferation assays were performed in microlon 96-well plates coated with anti-CD3 (1 µg/ml, UCHT-1) and anti-CD28 (1 µg/ml, clone 37407·11) or by stimulation with interleukin (IL)-15/IL-7 (25 ng/ml each). The optimal concentrations of anti-CD3 and anti-CD28 antibodies were determined by dose–response experiments. Carboxyfluorescein succinimidyl ester (CSFE)-labelled T cells (1 × 10⁶ per well) were cultured in 200 µl of Iscove’s modified Dulbecco’s medium (IMDM) containing 10% fetal calf serum (FCS) and penicillin/streptomycin (10 000 U/ml/10 mg/ml) at 37°C/5% CO₂. At the end of the culture period (7 days), cells and supernatant were collected and analysed by FACS and ELISA, respectively. Assessment of interferon (IFN)-γ and IL-10 production by ELISA was performed according to the manufacturer’s instructions (R&D Systems, Wiesbaden, Germany).

Whole blood assay

To assess monocyte activation, whole blood was diluted (1 : 4) in IMDM and stimulated with lipopolysaccharide (1 μg/ml) for 24 h. Supernatants were harvested and assessed for tumour necrosis factor (TNF)-α production by ELISA (R&D Systems).

Flow cytometry

Antigen expression on T lymphocyte subsets was determined by triple immunofluorescence staining using directly conjugated antibodies. To this end, PBMC were incubated for 30 min at 4°C with saturating amounts (10 µl) of conjugated monoclonal antibodies directed against CD4, CD45RO, CD25 and CCR7 (all from BD Biosciences, Heidelberg, Germany). The antibodies were either conjugated to FITC, R-phycoerythrin (RPE), peridinin chlorophyll (PercP) or allophycocyanin (APC), depending on the combination of specific antibodies used. The cells were washed twice to remove unbound antibodies and were finally resuspended in 300 µl of Cell Wash (BD Biosciences, Heidelberg, Germany). Analysis of all cells was performed on a FACSCalibur flowcytometer (BD Biosciences, Heidelberg, Germany) and the data were analysed using WinMDI version 2·8 software.

Oxidative burst, chemotaxis and phagocytosis

Whole blood (100 µl) was stimulated either with phorbol ester [phorbol myristate acetate (PMA)], bacteria (Escherichia coli), the chemotactic peptide formylmethionylleucylphenylalanine (fMLP) or phosphate-buffered saline (PBS) as a control (20 µl each). After 10 min of stimulation, 20 µl of dihydrorhodamine 123 (DHR) was added for another 10 min of incubation. The fluorogenic substrate DHR became oxidized to its fluorescent product rhodamine 123 within the cell. Adding 2 ml of a fixation reagent stops the oxidative burst followed by a washing step with PBS. Finally, the red fluorescent DNA stain propidium iodide was used as a counterstain to trigger the measurement on nucleated cells (all reagents supplied as a kit, BURSTTest, Orpegen Pharma, Heidelberg, Germany). Samples were analysed subsequently by flow cytometry at 488 nm (FACSCalibur, BD Biosciences, San Jose, CA, USA). Light scatter parameters were used to gate the neutrophilic granulocytes. The mean intensity of their log green fluorescence (530 nm) is a measure of the formation of oxidative radicals. To test phagocytosis, 100 µl heparinized whole blood was cooled to 0°C and mixed with 20 µl cooled E. coli. Phagocytosis and chemotaxis tests were performed according to the manufacturer’s instructions (Phagotest/Migratest; Orpegen Pharma, Heidelberg, Germany).

Statistical analysis

All data are given as means ± s.d. Differences in continuous variables pre- and post-DSG were compared by means of paired Wilcoxon test. A two-sided P < 0·05 was considered to indicate statistical significance.

Results

Influence of DSG on T cells

To test whether DSG influences the expression of T cell activation markers in vivo, CD25 and CCR7 on naive and memory CD4⁺ T cells were measured by FACS analysis. We observed no significant differences in CD25 expression during DSG treatment (in naive CD4 40·78 ± 29·57 before DSG, 46·32 ± 28·69 after DSG (P = 0·314); in memory CD4 56·83 ± 28·70% before DSG and 78·78 ± 11·21% after DSG (P = 0·066). Similarly, the expression of CCR7 was unaffected during therapy [in naive CD4 77·39 ± 26·16% before DSG, 77·35 ± 25·0% after DSG (P = 0·767); in memory CD4 52·02 ± 15·69% before DSG, 54·09 ± 15·55% after DSG (P = 0·214)].

In contrast, anti-CD3 and anti-CD28-mediated proliferation of CD4⁺ T cells was impaired under DSG therapy (Fig. 1a, P = 0·086). This was partially restored 2–3 weeks after cessation of DSG, i.e. before initiation of the next treatment cycle. Interestingly, cytokine (IL-7, IL-15)-mediated proliferation remained unchanged during DSG therapy (Fig. 1b, P = 0·374).

Fig. 1.

T cell proliferation before and after deoxyspergualin (DSG) was determined by carboxyfluorescein succinimidyl ester (CSFE)-labelling of peripheral blood mononuclear cells (PBMC). The percentage of non-proliferating cells before and after DSG treatment of each patient is shown (each patient is measured twice). (a) DSG increases the percentage of non-proliferating T cells after T cell receptor-mediated proliferation (stimulation with anti-CD3 and anti-CD28). (b) The percentage of non-proliferating T cells after cytokine-mediated proliferation by interleukin (IL)-7 and IL-15 is not changed after DSG treatment.

In addition to T cell receptor-mediated proliferation, DSG strongly inhibited cytokine production in activated CD4⁺ T cells. There was no indication for a specific inhibition of Th1 or Th2 cells as both IFN-γ and IL-10 production were strongly impaired at the end of each of the DSG cycles (Fig. 2, P < 0·005). Two weeks after cessation of DSG, i.e. directly before the next cycle was started, IFN-γ and IL-10 production in activated CD4⁺ T cells were restored (Fig. 2).

Fig. 2.

Interferon (IFN)-γ and interleukin (IL)-10 production of stimulated CD4⁺ T cells. Cytokine concentration was measured in supernatant of stimulated CD4⁺ T cells by enzyme-linked immunosorbent assay. (a) Significant decrease in IFN-γ production of each patient after treatment with deoxyspergualin (DSG) (P < 0·005), which was reversible. (b) Significant decrease of IL-10 production of each patient after treatment with DSG (P < 0·005), which was reversible (paired Wilcoxon’s test).

Effect of DSG on monocytes

To analyse in vivo effects of DSG on monocyte activation, lipopolysaccharide (LPS)-responsiveness was measured in a whole blood assay. TNF-α production was significantly impaired after DSG treatment (Fig. 3, P < 0·005). TNF-α production was increased after treatment pause compared to TNF-α production before the first DSG cycle (Fig. 3, P = 0·043). The reduction in TNF-α production after DSG could not be explained by a decrease in monocyte count, because monocyte counts did not change significantly (monocyte count before DSG 389 ± 269 µl, after DSG 162 ± 151 µl, P = 0·123, Table 2).

Fig. 3.

Tumour necrosis factor (TNF)-α production measured in whole blood after incubation of 24 h with 1 ng/ml lipopolysaccharide (LPS). To express the ability of monocytes to respond to stimulation with LPS, TNF-α production after LPS challenge minus TNF-α production without LPS is expressed. There was a significant decrease in TNF-α production upon LPS challenge after treatment with deoxyspergualin (DSG), which was reversible in all patients (P < 0·005) (paired Wilcoxon’s test).

Table 2.

Monocyte count in cell/µl of each patient before and after deoxyspergualin (DSG) therapy.

	Monocyte count (cells/µl)
	Before DSG	After DSG	Before DSG	After DSG
Patient 1	330	63	240	310
Patient 2	126	320	481	n.d.
Patient 3	n.d.	148	100	70
Patient 4	750	160	850	20
Patient 5	455	60	170	130

n.d.: Not done.

Effect of DSG on granulocytes

In all patients, a significant decrease in neutrophil counts was observed at the end of each treatment cycle (neutrophils before DSG 8114 ± 5977 µl, neutrophils after DSG 2798 ± 3250 µl, P = 0·012, Table 3), which was restored after treatment pause (Table 3). Neutropenia, i.e. a neutrophil count below 2200 µl, was seen in three of five patients during therapy.

Table 3.

Neutrophil count in cell/µl of each patient before and after deoxyspergualin (DSG) therapy.

	Neutrophil count (cells/µl)
	Before DSG	After DSG	Before DSG	After DSG
Patient 1	21 000	2420	3 820	3540
Patient 2	3 950	2990	8 040	n.d.
Patient 3	n.d.	1030	11 500	1520
Patient 4	5 280	782	12 500	1100
Patient 5	4 510	1460	2 430	1360

n.d.: Not done.

Although neutrophil counts decreased, DSG had no effect on neutrophil oxidative burst, chemotaxis and phagocytosis. These parameters were within the normal range before and after DSG (Table 4). Similarly, DSG did not influence surface PR-3 expression on neutrophils [83·38 ± 20·76% before DSG and 84·44 ± 14·19% after DSG (P = 0·635)].

Table 4.

Oxidative burst, phagocytosis and chemotaxis before and after deoxyspergualin (DSG) therapy.

Patients before and after DSG	Oxidative burst (FI)			Phagocytosis (FI)		Chemotaxis (counts)
	Neutrophils			Neutrophils	Monocytes	Neutrophils
	fMLP	E. coli	PMA	E. coli		fMLP
1
Before	50	205	1297	739	508	6 711
After	111	356	1151	1034	624	282
2
Before	105	285	912	1121	1010	941
After	207	438	553	1149	967	526
3
Before	38	329	520	1122	640	7 573
After	80	295	1005	842	696	11 554
4
Before	110	286	305	2092	985	1 552
After	61	385	554	1368	903	2 471
P-value	0·273	0·144	0·715	0·715	0·715	1·0

FI: fluorescence intensity; fMLP: formylmethionylleucylphenylalanine; PMA: phorbol myristate acetate. P-value: before versus after treatment.

Discussion

In the present study, we investigated the influence of cyclic DSG administration on leucocyte function and phenotype in WG patients. We showed that DSG inhibited T cell receptor-mediated T cell proliferation, cytokine production in both Th1 and Th2 T cells and monocyte activation in a reversible manner. In contrast, no effects were observed during DSG treatment on T cell activation surface markers, cytokine-mediated T cell proliferation, granulocyte function (i.e. oxidative burst, phagocytosis and chemotaxis) and the surface expression of PR-3 on granulocytes.

The results of the present study show possible sites where DSG might have beneficial effects in the treatment of refractory WG patients. First, DSG blocks monocyte activation, hence reducing TNF-α production required for neutrophil priming. Secondly, DSG causes neutropenia and thereby reduces the number of effector cells responsible for endothelial cell damage. However, the investigated functions of the remaining neutrophils were not affected, i.e. the oxidative burst, chemotaxis and phagocytosis of neutrophils were unchanged before and after DSG administration. Also, the PR-3 concentration on the surface of neutrophils remained unchanged before and after application of DSG. In previously published clinical trials, we have shown that DSG treatment resulted partly in a decline of ANCA titres [3,4]. The effect of DSG on B cell function was not part of the present study.

It is believed that monocyte activation and subsequent TNF-α production might play a pivotal role in neutrophil priming in AASV patients. DSG clearly diminished monocyte activation, as evidenced by inhibition of LPS-mediated TNF-α production. It must be mentioned, however, that TNF-α production was higher at the beginning of the second cycle than at the beginning of the first cycle. This might have been caused due to the mode of action of DSG [6]. Activation of NF-kappa B is initiated by proteolytic breakdown of its natural inhibitor I-kappa B. Subsequently translocation of NF-kappa B occurs via so called molecular chaperones, e.g. heat shock protein 70 (HSP70). DSG inhibits translocation of NF-kappa B by competing for binding with HSP70. DSG does not interfere in the proteolytic breakdown of I-kappa B. If breakdown of I-kappa B occurs during DSG therapy this will not result in TNF-α production, as translocation is blocked. However, free NF-kappa B might be higher in some patients depending on the extent of I-kappa B degradation that occurred during DSG treatment. Because we did not investigate I-kappa B degradation in this study, other possibilities for an increased TNF-α production after discontinuation of DSG cannot be excluded.

Neutrophil oxidative burst was not impaired during DSG treatment. This demonstrates that NF-kappa B nuclear translocation is not strictly required for activation of neutrophil nicotinamide adenine dinucleotide phosphate hydrogen (NADPH) oxidase. It has been suggested that this enzyme is located in two different pools in neutrophils, i.e. the plasma membrane and in the membrane of granules and that NADPH oxidase in these pools is activated by different signal transduction pathways [13]. Activation of protein kinase C and to a lesser extent IP3 kinase seems to be involved more critically in activation of NADPH oxidase. However, a recent report also claims that expression of gp91 (phox) requires functional p65/RelA dimers [14]. Alternatively, the lack of influence of DSG on neutrophil oxidative burst is that NADPH oxidase is already preformed in these pools.

In the present study, we tested only the effect of DSG on surface PR3 expression and neutrophil oxidative burst, chemotaxis and phagocytosis. These data suggest that translocation of NF kappa B is not required for these functional activities of neutrophils. We would like to emphasize, however, that in the present paper other potentially important neutrophil functions, e.g. cytokine production, might well be dependent upon NF kappa B activation. This was not investigated in the present study. Because DSG inhibits NF kappa B-dependent cell functions, this might also be the case in neutrophils and therefore we cannot state firmly that DSG does not impair neutrophil function per se. Similarly, this study does not exclude that translocation of PR-3 to the membrane is independent of NF kappa B activation and that translocation is not influenced by DSG.

In AASV patients several abnormalities in T lymphocytes have been reported, which suggests that T cells may also contribute to disease onset and/or perpetuation. These abnormalities include lymphopenia [15], a high number of activated T lymphocytes in peripheral blood [16], skewing of Th1 to Th2 responses [17] and senescence of CD28^– lymphocytes in the CD8⁺ subpopulation [18].

Our results also demonstrate beneficial effects of DSG on T cells by inhibiting T cell receptor-mediated T cell proliferation and cytokine production in both Th1 and Th2 T cells. Our results are in agreement with the mode of action of DSG, i.e. blocking the nuclear translocation of the transcription factor NF-kappa B [6] which is involved in T cell receptor-mediated proliferation and cytokine production [19]. In contrast, DSG did not inhibit cytokine-mediated homeostatic proliferation, suggesting that nuclear translocation of NF-kappa B is not strictly required for this type of T cell proliferation. Homeostatic proliferation is mediated by IL-7 and IL-15 and both factors activate transcription factors belonging to the stat-family [20]. To our knowledge, the involvement of NF-kappa B in homeostatic proliferation has not been studied thus far.

DSG did not influence the expression of the T cell activation marker CD25 in vivo. This seems to be in contradiction to the inhibitory effect of DSG on T cell receptor-mediated activation we observed in vitro. This discrepancy might be explained as follows: first, the involvement of NF-kappa B in the regulation of CD25 on T cells has not been established beyond doubt [21,22]. Secondly, the cause of a persistent increase in CD25 expression in AASV patients [11] is unknown and therefore not necessarily related to T cell receptor-mediated activation.

Based on our results, we cannot distinguish whether the reduced IFN-γ production after DSG treatment is mediated directly via inhibition of IFN-γ gene expression or whether this reflects diminished T cell activation. To our knowledge, there are no data available that show NF kappa B involvement in IFN-γ gene expression.

Our findings are compatible with previous studies investigating the effects of DSG on T cell activation: Holcombe et al. also described an inhibition of IFN-γ production by CD4⁺ T cells in vitro [23] and Odaka et al. reported an inhibition in proliferation of murine T cell hybridomas resulting in cell death after treatment with the DSG analogue methyl-DSG [24]. However, it must be noted that our results conflict, to a certain extent, with the data of Borg et al. describing no influence of DSG on the production of IFN-γ in vitro and a reduced proportion of CD25⁺ cells. Nevertheless, they also found that DSG inhibits T cell proliferation in human peripheral blood leucocytes when stimulated with Staphylococcus aureus antigen [25].

In the present study we did not measure T cell responses against PR-3. Autoreactive T cells against PR-3 are present in both healthy individuals and WG patients, but there are conflicting data as to whether T cell responses to PR3 are generally increased in vasculitis patients [26,27]. Nevertheless, it is believed that ANCA production is, at least partly, T cell-dependent. Because DSG treatment decreases ANCA titre, as we have shown previously [3,4], this could indicate that activation of autoreactive T cells is also impaired by DSG, either directly or by influencing antigen presentation. However, direct proof for this assumption has to be established.

In conclusion, our findings might explain the beneficial effects of DSG for the treatment of refractory patients with WG. Taking the current understanding of the pathogenesis of AASV into account, activated monocytes produce TNF-α which primes neutrophils. In the presence of ANCA, primed neutrophils degranulate and produce reactive oxygen species which lead subsequently to endothelial damage [28]. In an animal model of ANCA-associated disease, systemic administration of LPS dose-dependently increased renal injury induced by anti-MPO IgG which could be attenuated by anti-TNF-α treatment [29]. Neutropenia caused by DSG will result in a reduction of effector cells. In addition, diminished monocyte activation influences TNF-α production and thus might decrease neutrophil priming. Moreover, DSG therapy decreases ANCA titre and might reduce ANCA-mediated neutrophil activation. Although DSG did not influence the sustained expression of T cell activation markers in vivo, ex vivo T cell receptor-mediated activation was clearly impaired under DSG treatment.