Myeloperoxidase-specific anti-neutrophil cytoplasmic antibody (MPO–ANCA) is associated with rapidly progressive glomerulonephritis (RPGN) and glomerular crescent formation. Pathogenic factors in RPGN were analyzed by using SCG/Kj mice, which spontaneously develop MPO–ANCA-associated RPGN. The serum concentration of soluble IL-6R was determined by using ELISA and those of another 23 cytokines and chemokines by Bio-Plex analysis. Sections of frozen kidney tissue were examined by fluorescence microscopy and the CD3+B220+ T cell subset in the spleen determined by a flow cytometry. Concentrations of IL-6 and monocyte chemotactic protein-1 were significantly correlated with the percentages of crescent formation. Anti-IL-6R antibody, which has been effective in patients with rheumatoid arthritis, was administered to SCG/Kj mice to elucidate the role of IL-6 in the development of RPGN. MPO–ANCA titers decreased after administration of anti-IL-6R antibody, but not titers of mizoribine, which is effective in Kawasaki disease model mice. These results suggest that IL-6-mediated signaling is involved in the production of MPO–ANCA.
anti-neutrophil cytoplasmic autoantibodies
cluster of differentiation
granulocyte colony-stimulating factor
granulocyte-macrophage colony-stimulating factor
hematoxylin and eosin
monocyte chemotactic protein-1
macrophage inflammatory protein 2
anti-neutrophil cytoplasmic antibody against myeloperoxidase
rat monoclonal anti-mouse IL-6R antibody
regulated upon activation, normal T cell expressed and secreted
red blood cells
rapidly progressive glomerulonephritis
spontaneous crescentic glomerulonephritis forming/Kinjoh
systemic lupus erythematosus
tumor necrosis factor
Anti-neutrophil cytoplasmic antibody against myeloperoxidase is associated with the development of small vessel vasculitis [1, 2]. Patients with MPO–ANCA-positive RPGN must receive a combination of corticosteroids and immunosuppressants. However, side effects including infection, myelosuppression, and malignancy often occur with the well-known immunosuppressant, cyclophosphamide. Therefore, establishment of a new therapeutic agent with less side effects is desirable, particularly because MPO–ANCA -associated RPGN occurs in elderly people.
A mouse model is useful for development of treatment for MPO–ANCA-associated vasculitis. Several spontaneous mouse models of autoimmune crescentic glomerulonephritis such as MRL/lpr , New Zealand Black/White F1 hybrid , and SCG/Kj mouse have been used . Because SCG/Kj mice develop severe crescentic glomerulonephritis earlier and more rapidly and with higher MPO–ANCA positivity than do any other murine models, this strain is thought to be useful for investigating the pathogenic mechanisms and treatment of MPO–ANCA-associated RPGN.
It has been proposed that cytokines secreted by activated autoreactive T cells induce autoantibody production by plasma cells, thereby leading to glomerular damage caused mainly by neutrophils activated by autoantibodies. However, the factors related to T cell activation, autoantibody production and neutrophil activation are not yet clear. Therefore, we sought to identify the cytokines and/or chemokines responsible for the development of RPGN in SCG/Kj mice by using a microsphere-based multiplex protein assay, and to investigate their clinical potential as a target of monoclonal antibody therapy. A monoclonal antibody against IL-6R, tocilizumab, has been used to treat inflammatory diseases such as Castleman disease [6-8], systemic juvenile idiopathic arthritis [9, 10], and RA . Because the IL-6 signaling pathway plays an important role in autoantibody production, it should be studied in relation to MPO–ANCA production in SCG/Kj mice.
On the other hand, MZR, an immunosuppressive drug with few side effects, has been used for management of organ transplantation, lupus nephritis and RA. Some preclinical studies have shown an inhibitory effect of MZR on collagen-induced arthritis , anti-DNA antibody production in New Zealand Black/White F1 mice  and Kawasaki disease model mice . Some clinical studies on the efficacy of MZR in MPO–ANCA -associated vasculitis have also been reported [15-17].
In the present study, we examined the CD3+B220+ T cell subset in the spleen and the role of IL-6 in MPO–ANCA-associated vasculitis using RPGN model SCG/Kj mice. In addition, we compared the efficacy of the IL-6R antibody and MZR in RPGN model mice.
MATERIALS AND METHODS
Mice and treatment with anti-interleukin-6 receptor and mizoribine
Female SCG/Kj mice (Nippon Kayaku, Tokyo, Japan) and C57BL/6 mice for healthy controls (CLEA Japan, Tokyo, Japan) were maintained under specific pathogen-free conditions according to the guidelines for animal care at Chiba University. To test the effects of anti-IL-6R antibody, rat monoclonal anti-mouse IL-6R antibody (clone MR16-1; Chugai Pharmaceutical, Tokyo, Japan) was used. The antibody was injected i.v. into SCG/Kj mice (7–8 weeks old) at a dose of 2 mg/mouse in 160 µL of PBS. Seven days later a dose of 0.5 mg/mouse in 200 µL of PBS was injected i.p.; thereafter the same treatment was repeated once or twice a week. Control mice were injected with PBS in the same manner. To test the effects of MZR (Asahi Kasei Pharma, Tokyo, Japan), MZR was injected i.p. into SCG/Kj mice (9–11 weeks old) at a dose of 0.6 mg/mouse in 200 µL of PBS. Control mice were injected with PBS in the same manner. During these treatments, urinalysis was performed periodically. At the end of the treatments, the mice were killed, and their serum, kidneys and spleens removed for subsequent experiments.
Kidneys from SCG/Kj mice were fixed with buffered formalin and embedded in paraffin. Sections were stained with HE. The percentages of glomeruli involved in crescent formation were counted in cross-sections of kidneys containing over 50 glomeruli. For immunohistochemistry, kidney tissues frozen in Optimal Cutting Temperature compound were sectioned in a cryostat. These sections were fixed with cold acetone for 10 mins, blocked with 5% BSA (Wako, Osaka, Japan) for 30 mins and then incubated with primary antibodies (goat anti-mouse IgG (1:100, Cappel; MP Biochemical, Santa Ana, CA, USA) or goat anti-mouse IL-6R (1:100, R & D Systems, Minneapolis, MN, USA) for 1 hr. After washing with PBS, the sections were incubated with Alexa 488-labeled rabbit anti-goat IgG (1:200) for 1 hr, washed with PBS, mounted with PermaFluor Aqueous Mounting Medium (ThermoScientific, Waltham, MA, USA), and examined by fluorescence microscopy (Nikon, Tokyo, Japan).
Determination of cytokines and chemokines by enzyme-linked immunosorbent and Bio-Plex assays
Antibody titers to MPO were measured as described previously . Briefly, recombinant mouse MPO was coated onto a 96-well Maxisorp ELISA plate (ThermoScientific) overnight at 4°C. The plate was blocked with 1% BSA for 2 hrs and then incubated with serum diluted 1:50 with dilution buffer (1% BSA and 0.05% Tween 20 in PBS) for 1.5 hrs at room temperature. The bound mouse IgG was detected by 2 hrs incubation with alkaline phosphatase-labeled anti-mouse IgG antibody (Jackson ImmunoResearch, West Grove, PA, USA). The bound secondary antibody was subsequently quantified by changes in the absorbance at 405 nm after incubation with 1 mg/mL p-nitrophenyl phosphate (Sigma–Aldrich, St. Louis, MO, USA). MPO–ANCA titers in mouse serum were calculated using a standard curve obtained by rabbit standard anti-MPO antibody. Serum concentrations of sIL-6R were determined using an ELISA kit for mouse sIL-6R (R&D Systems) according to the manufacturer's instructions. Bio-Plex assay (Bio-Rad Laboratories, Hercules, CA, USA) was performed according to the manufacturer's instructions. Mouse Cytokine Group I (23-Plex) was used to quantify cytokine and chemokine concentrations in serum obtained from SCG/Kj mice.
Determination of T-cell subpopulation by flow cytometry
The spleens were dissociated in FACS buffer (2% FBS [Gibco, Langley, OK, USA] in PBS) between two frosted slides. The cell suspensions were filtered through a stainless steel mesh (75 µm) and centrifuged for 5 mins at 300 g. The RBCs were then lysed in RBC lysis buffer (0.83% NH4Cl, 20 mM Tris–HCl, pH 7.6) for 1 min, washed twice with FACS buffer and suspended in FACS buffer. Minced kidneys of SCG/Kj mice were filtered through a stainless steel mesh (106 µm), washed with RPMI-1640 medium (Sigma–Aldrich) and treated with 1 mg/mL collagenase (Wako) diluted in PBS for 30 mins at 37°C with agitation. After centrifugation (450 g, 5 mins, 4°C), the cell pellets were resuspended in 5 mL of RPMI-1640 medium, mixed with 5 mL of 100% Percoll PLUS (GE Healthcare, Uppsala, Sweden), and centrifuged for 30 mins at 800 g. One hundred percent Percoll stock solution was made by mixing nine volumes of Percoll PLUS with one volume of 10 × PBS. The cell pellets were treated with RBC lysis buffer for 1 min, washed twice with FACS buffer, and suspended in FACS buffer. Cells from spleens or kidneys were counted and resuspended in FACS buffer at 2 × 106 cells/100 µL. To block Fc receptor-mediated binding of antibodies, anti-mouse CD16/CD32 (2.4G2; BD Pharmingen; BD Biosciences, San Jose, CA, USA) was added. The cells were labeled with FITC-conjugated anti-CD3 antibody (145-2C11, 1:30 dilution) and phycoerythrin-conjugated anti-B220 antibody (RA3-6B2, 1:200 dilution) for 30 mins on ice. After antibody labeling, the cells were washed, resuspended in 300 µL of FACS buffer and then analyzed with FACSCalibur (BD Biosciences).
All data are expressed as mean ± SD. For individual comparisons, Student t-test or Welch corrected t-test was used where appropriate and differences with P < 0.05 were considered significant.
Onset of glomerulonephritis in spontaneous crescentic glomerulonephritis forming/Kinjoh mice
It has been reported that SCG/Kj mice have urinary abnormalities, crescent formation and increase in MPO–ANCA titers [5, 18]. Therefore, we first monitored urinary protein and hemoglobin in SCG/Kj mice from 5 to 14 weeks of age. As shown in Figure 1a,b, SCG/Kj mice began to show increases in the urinary concentrations of protein and hemoglobin between 8 and 9 weeks of age. In addition, the mice began to die from 11 weeks of age (Fig. 1c). These findings, in addition to data reported elsewhere, indicate that SCG/Kj mice develop RPGN from 8 to 10 weeks of age. Therefore, we divided SCG/Kj mice into two groups based on age (initial phase 6–7 weeks and active phase 14–15 weeks) and analyzed differences between the two groups. As shown in Figure 1d, the group of SCG/Kj mice in active phase had higher MPO–ANCA titers than did those in the initial phase, which had MPO–ANCA titers similar to those of healthy control C57BL/6 mice. Likewise, the severity of inflammation in the active phase group was greater than that in the initial phase group, as judged by the degree of glomerular crescent formation and inflammatory cell infiltration in perivascular and periglomerular areas (Fig. 1e). The percentage of crescent formation was also higher in the active phase group (Fig. 1f).
Determination of population of CD3+B220+ T cell subset
Because it has been reported that a spleen T cell subset expressing both CD3 and B220 in SCG/Kj [18, 20] and MRL/lpr mice  is caused by Fas mutation and leads to a defect in Fas-mediated apoptosis, we further examined changes in the population of the CD3+B220+ T cell subset with aging. FACS analysis of splenocytes showed that CD3+B220+ T cells were rarely present in control C57BL/6 mice but their numbers were increased in SCG/Kj mice (Fig. 2a). As shown in Figure 2b, the CD3+B220+ T cell population was significantly increased in the active phase group, suggesting that this T cell subset increases with age in SCG/Kj mice. Furthermore, the percentage of CD3+B220+ T cells correlated significantly with MPO–ANCA titers (Fig. 2c). In addition to their occurrence in the spleen, the presence of CD3+B220 T cells in kidneys is of particular interest because vascular damage generally occurs there. Therefore, we further assessed the presence of CD3+B220+ T cells among the mononuclear cells infiltrating the kidneys in SCG/Kj mice and found increasing infiltration with age of CD3+B220+ T cells into the kidneys of SCG/Kj mice (Fig. 2d).
Correlation of serum concentrations of interleukin-6 and monocyte chemotactic protein-1 with development of glomerulonephritis
Having shown substantial differences between initial and active phases in urine, MPO–ANCA titers, histology and abnormal T cell populations, we next sought to identify the molecules responsible for development of RPGN in SCG/Kj mice by comparing the serum cytokine and chemokine profiles of the two phases. Serum cytokine and chemokine profiles in the initial and active phases obtained by using the Bio-Plex Mouse Cytokine 23-Plex Panel revealed that nine molecules (IL-4, IL-6, IL-10, IL-12p40, G-CSF, GM-CSF, IFN-γ, MCP-1 and RANTES) were in significantly higher concentrations in the sera of active phase mice than in those of initial phase mice (Fig. 3a). Among these molecules, IL-6 and MCP-1 were significantly correlated with the percentages of crescent formation (Fig. 3b,c).
Up-regulation of interleukin-6 receptor in both serum and glomeruli with development of glomerulonephritis
Because the serum concentrations of IL-6 and MCP-1 were significantly correlated with the degree of crescent formation, both molecules are probably involved in the development of RPGN in SCG/Kj mice. Increased serum IL-6 has been previously shown in patients with MPO–ANCA-associated renal vasculitis, whereas MCP-1 reportedly increase mainly in the urine of patients. To develop a therapeutic strategy, we focused on further investigating the role of IL-6 in the development of vasculitis in SCG/Kj mice. Recent findings have shown that the pathophysiological effects of IL-6 may depend strongly on a soluble form of its receptor . Therefore, we measured serum concentrations of sIL-6R in SCG/Kj mice. As shown in Figure 4a, serum sIL-6R concentrations in SCG/Kj mice in active phase were significantly higher than those in initial phase, these findings being similar to those of control C57BL/6 mice. When IL-6R expression in kidneys from SCG/Kj mice was assessed by immunohistochemistry (Fig. 4b), the glomeruli in the active phase group showed positive immunofluorescence for IL-6R, whereas those in the initial phase group did not. However, we did not identify which particular category of cells was positive for expression of IL-6R.
Decrease in titers of anti-neutrophil cytoplasmic antibody against myeloperoxidase induced by treatment with anti-interleukin-6 receptor antibody
Comparison of the concentrations of IL-6 and sIL-6R in the active phase with those in the initial phase group suggested that IL-6 and sIL-6R may be involved in development of RPGN in SCG/Kj mice. Moreover, it has been shown that the recombinant humanized anti-IL-6R antibody tocilizumab is clinically effective for treatment of Castleman disease [6, 7], systemic juvenile idiopathic arthritis [9, 10] and RA . Therefore, we treated mice with MR16-1 to explore the effects of IL-6R antibody on RPGN in SCG/Kj mice. In addition, we examined MZR, an immunosuppressive drug expected to be used for maintenance of remission of ANCA-associated vasculitis.
First, we examined a treatment protocol in which we gave SCG/Kj mice one shot of MR16-1 per week for 3 weeks. Urinalysis during this treatment revealed that SCG/Kj mice treated with MR16-1 had a tendency, albeit not statistically significant, to have decreased scores for proteinuria and hematuria (Fig. 5a). We attempted to improve the efficacy of MR16-1 by increasing the frequency of injections to two per week. However, the second protocol of treatment unexpectedly aggravated proteinuria and hematuria (Fig. 5b). However, MZR treatment tended to decrease the hematuria but did not affect the proteinuria (Fig. 5c).
We also measured serum MPO–ANCA titers in SCG/Kj mice treated with MR16-1 or MZR. Interestingly, treatment with two shots per week of MR16-1 significantly decreased MPO–ANCA titers although one shot per week did not (Fig. 6a,b). MZR slightly decreased MPO–ANCA titers (Fig. 6c). We also examined the percentage of crescent formation, splenic CD3+B220+ T cells, and serum cytokine and chemokine profiles: but neither MR16-1 nor MZR induced significant change in these variables (data not shown).
Increase in glomerular immunoglobulin G deposition induced by administration of anti-interleukin-6 receptor antibody or mizoribine
We were interested in why renal functions were not significantly improved by two shots per week of anti-IL-6R antibody, given that MPO–ANCA titers were decreased by this treatment. Therefore, we investigated IgG deposition as another criterion for glomerular damage. As shown in Figure 7, the glomeruli of SCG/Kj mice treated with anti-IL-6R antibody had more intense IgG deposition than those of control mice. This observation indicates that injected anti-Il-6R antibody may be deposited in glomeruli. Because repeated injections of rat antibody (MR16-1) may have induced production of antibody against rat IgG, we measured anti-rat IgG titers in SCG/Kj mice treated with MR16-1. Surprisingly, SCG/Kj mice possessed antibodies against rat IgG regardless of whether MR16-1 had been injected or not (data not shown), indicating that injected MR16-1 and endogenous anti-rat IgG may form immunocomplexes that are deposited in glomeruli.
In this study, we attempted to identify factors responsible for the development of RPGN by comparing various variables in the initial and active phases of this condition. Increases in proteinuria, hematuria, MPO–ANCA titers, inflammatory cell infiltration and crescent formation clearly correlated with increasing age. We also identifed CD3+B220+ T cells in the kidneys and spleens of SCG/Kj mice. So-called double-negative T cells (CD3+CD4-CD8-), which are more numerous in the peripheral blood of patients with SLE than in that of healthy controls, facilitate production of pathogenic anti-DNA autoantibody . It has been recently found that the DNA of T cells infiltrates the kidneys of SLE patients, causing production of significant amounts of IL-17 . In the present study, there was a significant positive correlation between numbers of CD3+B220+ T cells in the spleen and MPO–ANCA titers, suggesting that CD3+B220+ T cells may be involved in the production of autoantibodies, including MPO–ANCA, in SCG/Kj mice.
In this study, there was suggestive evidence that IL-6 and MCP-1 contribute to an increase of the crescent formation in SCG/Kj mice. Several reports have demonstrated increased plasma concentrations of IL-6 and IL-6R in patients with ANCA-associated systemic vasculitis [25, 26]. Upon binding of a complex of IL-6 and sIL-6R, human umbilical vein endothelial cells are activated by a trans-signaling pathway via gp130 expressed on their cell surfaces and up-regulate expression of MCP-1, MCP-3, IL-8, IL-6, vascular cell adhesion molecule 1, intercellular adhesion molecule 1, and protein S, thus contributing to leukocyte recruitment [27, 28]. In addition, sIL-6R released from neutrophils by shedding is one of the major sources of sIL-6R, leading to activation of endothelial cells by forming IL-6–sIL-6R complexes . Therefore, the strong correlation between MCP-1 concentrations and crescent formation found in the present study is reasonable. We propose that neutrophils activated by MPO–ANCA release sIL-6R, forming a complex with the increased plasma IL-6, leading to endothelial activation and subsequent crescent formation.
In addition to IL-6 and MCP-1, we found that serum concentrations of several cytokines and chemokines such as IL-4, IL-10, IL-12p40, G-CSF, GM-CSF, IFN-γ, and RANTES are significantly increased with aging in SCG/Kj mice. We have previously shown that concentrations of IL-10 and IL-12p40 are decreased by treatment with 15-deoxyspergualin , further indicating that these two cytokines are important in the development of RPGN and would be potential therapeutic targets.
Based on known facts about IL-6 and sIL-6R, we chose IL-6 as a potential target for treatment of ANCA-associated RPGN in the present study. One shot per week of MR16-1 did not significantly improve any indicator of renal function in SCG/Kj mice. Increasing the frequency of injections to two per week cuase deterioration of renal functions and increased glomerular IgG deposition. These adverse effects of anti-IL-6R antibody may be attributable to endogenous anti-rat IgG antibody in SCG/Kj mice. The anti-rat IgG antibody would reduce the therapeutic efficacy of MR16-1 by binding to rat anti-IL-6R antibody. Furthermore, the immunocomplexes formed in blood would deposit in the glomeruli of SCG/Kj mice, resulting in deterioration of renal function. In contrast, anti-IL-6R treatment was beneficial in reducing MPO–ANCA titers. IL-6 was originally identified as a B cell differentiation factor: thus one of the major functions of IL-6 is antibody induction. Indeed, IL-6 is reportedly required for autoantibody production [30, 31]. Unfortunately, it was difficult to show that anti-IL-6R antibody is effective in SCG/Kj mice; probably because endogenous antibodies against MR16-1 potentially decrease autoantibody concentrations during anti-IL-6R antibody therapy. Therefore, future research assessing the clinical usefulness of this therapy would be worthwhile.
On the other hand, the effect of MZR on MPO–ANCA-associated RPGN was not marked; our findings raise the possibility that MZR will only be helpful in reducing hematuria. A basic study reported the efficacy of MZR in MRL/lpr mice, an autoimmune disease-prone mouse that has low MPO–ANCA positivity . However, no reported studies have investigated the effects of MZR, which is effective against a model of childhood vasculitis Kawasaki disease , on the MPO–ANCA-associated RPGN model.
In summary, by further studying the involvement of IL-6 in the pathogenic mechanism of MPO–ANCA-associated RPGN, we found that MPO–ANCA production can be reduced by inhibition of IL-6-mediated signaling.
We thank laboratory members in the Department of Immunology, Graduate School of Medicine, Chiba University, Chiba, for their valuable discussions. Financial support for this study was from a Research-in-Aid Grant from the Ministry of Health, Labour and Welfare of Japan (H21-S-I-004).
The authors have no financial conflicts of interest.