MAPKs signaling pathway

Free Radical Biology and Medicine 99 (2016) 214–224

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original article

SPAK plays a pathogenic role in IgA nephropathy through the activation of NF-κB/MAPKs signaling pathway Tsai-Jung Lin a, Sung-Sen Yang b,c,d, Kuo-Feng Hua e, Yu-Ling Tsai a, Shih-Hua Lin c, Shuk-Man Ka a,f,g,n a

Graduate Institute of Life Sciences, National Defense Medical Center, Taipei, Taiwan Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei, Taiwan c Division of Nephrology, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan d Institute of BioMedical Sciences, Academia Sinica, Taipei, Taiwan e Department of Biotechnology and Animal Science, National Ilan University, Ilan, Taiwan f Graduate Institute of Aerospace and Undersea Medicine, National Defense Medical Center, Taipei, Taiwan g Department of Pathology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan b

art ic l e i nf o

a b s t r a c t

Article history: Received 5 March 2016 Received in revised form 5 August 2016 Accepted 8 August 2016 Available online 9 August 2016

Sterile 20/SPS1-related proline/alanine-rich kinase (SPAK) can stimulate production of proinflammatory cytokines and interact with inflammation-related molecules. However, it has yet to be determined whether SPAK plays a pathophysiological role in the complicated pathological mechanisms of IgA nephropathy (IgAN), which is mainly characterized by mesangial cell (MC) proliferation and is the most common form of glomerulonephritis. In the present study, we examined the pathophysiological role of SPAK in IgAN using a mouse model and cell models. Our results clearly showed that (1) SPAK deficiency prevents the development of IgAN and inhibits production of immune/inflammatory mediators and T cell activation and proliferation; and (2) when primed with IgA immune complexes (IgA IC), both peritoneal macrophages and primary MCs from SPAK knockout mice show markedly reduced production of proinflammatory cytokines and inhibition of NF-κB/MAPKs activation. We proposed that activation of SPAK and the NF-κB/MAPKs signaling pathway in MCs, macrophages and T cells of the glomerulus may be a mechanism underlying the pathogenesis of IgAN. The activation of SPAK in renal tubuloepithelial cells either directly by IgA IC or an indirect action of the activated MCs or infiltrating mononuclear leukocytes seen in the kidney may further aggravate the disease process of IgAN. Our results suggest that SPAK is a potential therapeutic target for the glomerular disorder. & 2016 Elsevier Inc. All rights reserved.

Keywords: Sterile 20/SPS1-related proline/alanine-rich kinase IgA nephropathy IgA immune complexes Mesangial cells Knockout mice T cells Macrophages NF-κB/MAPKs signaling

1. Introduction Sterile 20/SPS1-related proline/alanine-rich kinase (SPAK), a member of the sterile 20 serine/threonine kinase family, plays an important role in blood pressure regulation by affecting renal sodium reabsorption in the renal tubules and vascular contractility [1–3]. Upregulation of SPAK can result in development of inflammatory bowel disease [4], while SPAK deficiency is protective [5]. IgA nephropathy (IgAN), the most common type of glomerular disorder, is characterized by glomerular mesangial cell (MC) proliferation and glomerular deposition of IgA immune complexes (IgA IC) [6,7] and is the most frequent type of renal disorder that occurs concomitantly with inflammatory bowel disease [8]. This n Correspondence to: Graduate Institute of Aerospace and Undersea Medicine, Academy of Medicine, National Defense Medical Center, No. 161, Sec. 6, Min-Quan E. Road, Taipei, Taiwan. E-mail address: [email protected] (S.-M. Ka).

http://dx.doi.org/10.1016/j.freeradbiomed.2016.08.008 0891-5849/& 2016 Elsevier Inc. All rights reserved.

clinical finding suggests that SPAK and related signaling pathways might contribute to the pathogenesis of IgAN, although the exact pathogenic mechanisms remain largely unknown. Additionally, dysregulated activation of T cells [7,9,10] and infiltration of macrophages [10–12] around the inflamed glomeruli (periglomerular region) in the interstitium of the kidney have been implicated in the development and the progression of IgAN [13–15]. Recently, we demonstrated that SPAK upregulation is involved in the pathogenesis of hyperoxia-induced acute lung injury and that deficiency of SPAK attenuates severity of injury [16]. NF-κB, an effector of TNF-α, has been reported to be a transcriptional regulator of SPAK [16,17], and ERK, JNK, and p38 MAPK act as the downstream effectors of SPAK [18,19]. However, it is unclear whether SPAK plays a role in the pathogenesis of IgAN. In this study, our aim was to determine whether SPAK and related signaling molecules play a role in the pathogenesis of IgAN using SPAK knockout (KO) mice and cell models. We found that increased expression of activated/phosphorylated SPAK was

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

observed in the renal tissues in a mouse model of IgAN and that SPAK deficiency led to improved clinical and renal histopathology in the mice through modulation of the NF-κB/MAPKs signaling pathway.

2. Materials and methods 2.1. IgAN model We used 6- to 8-week-old male SPAK KO mice (C57BL/6 background), which were bred and maintained at the animal center of the National Defense Medical Center, Taipei, Taiwan as described previously [2], with age- and sex-matched C57BL/6 mice as wild type (WT) controls. In brief, the mice were given 36 consecutive daily intravenous injections of mouse purified IgA anti- phosphocholine (PC) antibodies (75 mg/kg body weight) and pneumococcal antigen (PnC) (5 mg/kg body weight), as described previously [6,7,10,20,21]. Mice treated with saline were used as normal controls. On day 36, all of the mice were sacrificed and urine and blood samples and the kidneys and spleen were collected and stored until use. All animal experiments were performed with the approval of the Institutional Animal Care and Use Committee of the National Defense Medical Center, Taipei, Taiwan, and in compliance with the NIH Guidelines for the Care and Use of Laboratory Animals. 2.2. Clinical and pathological evaluation Albumin levels in urine (determined by the albumin/creatinine [Cr] ratio) and blood urea nitrogen (BUN) and Cr levels in blood samples were measured as described previously [7,22]. For renal histopathology, renal tissues were fixed in 10% buffered formalin and embedded in paraffin, then 3 mm sections were prepared and stained with hematoxylin and eosin. Histological scoring of glomerular proliferation, glomerular sclerosis, or periglomerular mononuclear leukocyte infiltration was performed as described previously [7,22]. Briefly, 40 glomeruli were randomly selected from each section and positive signals within the selected glomerulus highlighted, measured, and quantified as a percentage of the total area of the glomerulus. To determine the percentage of glomeruli showing periglomerular mononuclear leukocyte infiltration, 40 randomly selected fields of the tubulointerstitial compartment in the cortical area were examined.

215

sections were single stained by incubation overnight at 4 °C with the anti-p-SPAK antibodies or antibodies against p38 MAPK or p-NF-κB p65 (both from Cell Signaling Technology, Beverly, MA), F4/80 þ (Serotec, Raleigh, NC), CD3 þ (pan T cell; Dako, Glostrup, Denmark) or CD4 þ (BioLegend, San Diego, CA) followed by incubation with horseradish peroxidase-conjugated secondary antibodies (DAKO) and DAB reagent (DAKO) and image analysis using Pax-it quantitative image analysis software (Paxcam, Villa Park, IL). The numbers of cells positive for p-SPAK, p-NF-κB p65, p38 MAPK, F4/80 þ , CD3 þ , or CD4 þ were counted at  400 magnification in either 40 consecutive glomeruli (expressed as cells/glomerular cross-section) or in 40 randomly selected fields of the tubulointerstitial compartment in the cortical area (expressed as cells per field). [24]. 2.5. T cell proliferation, T cell activation, and intracellular cytokine staining Mouse splenocytes were prepared as described previously [7,25]. We grind the whole spleen first to obtain a mixed cell populations, and then add anti-CD3 antibody (eBioscience, San Diego, CA) to the mixed of cells to induce T cell proliferation, and go on to perform each of the related experiments. Proliferation of T cells in the splenocytes was measured by measuring uptake of [3H]-thymidine (Amersham Pharmacia Biotech, Piscataway, New Jersey) using a TopCount (Packard/PerkinElmer, Boston, MA) as described previously [7,25]. For the T cell activation assay (CD69 þ cells), the cells were double-stained with FITC-conjugated antibodies against mouse CD3 þ (17A2; pan T cells), CD4 þ (GK1.5; helper T cells), or CD8 þ (53-6.7; cytotoxic T cells) and PE-conjugated anti-mouse CD69 þ antibodies (H1.2F3) (all from BD Biosciences, San Diego, CA) and the results expressed as a percentage of CD3 þ CD69 þ , CD4 þ CD69 þ or CD8 þ CD69 þ T cells, as described previously [7,25]. For intracellular cytokine staining, after stimulation of splenocytes for 5.5 h with phorbol 12-myristate 13-acetate, ionomycin, and monensin (all from Sigma, St. Louis, MO), the cells were stained with an FITC-conjugated antibody against mouse CD4 þ , then were incubated with saponin for 30 min before being incubated with PE-conjugated antibodies against mouse IL-2 (JES65H4), IFN-γ (XMG1.2), or IL-4 (11B11) (all from BD Biosciences) and the result expressed as the number of positive cells in T cells expressing IL-2, IFN-γ, or IL-4, as described previously [10,25]. Flow cytometry analysis was carried out using a FACSCalibur (BD Biosciences).

2.3. Generation of phosphorylated SPAK (p-SPAK) antibodies 2.6. Enzyme-linked immunosorbent assay (ELISA) Antibodies against SPAK phosphorylated on Ser-380 were generated by immunizing rabbits with a keyhole limpet hemocyanin-conjugated synthetic phosphopeptide, RRVPGS (pS) GHLHKTE, corresponding to residues 374–387 of human SPAK, can be used for both mouse and human cells/tissues, followed by affinity purification on phosphopeptide- or non-phosphorylated peptide -conjugated cellulose. These antibodies also recognize OSR1 phosphorylated on Ser-325 [RRVPGS (pS) GRLHKTE], where H is the only residue that differs between the two sequences [2,3,23]. 2.4. Immunofluorescence and immunohistochemistry For immunofluorescence, frozen renal tissues were cut and incubated with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgA (Cappel, Durham, NC). Scoring of staining intensity was performed as described previously [7]. For immunohistochemistry, 10% formalin-fixed and paraffin-embedded tissue sections or frozen sections, as described previously. [24] The

IFN-γ, TNF-α, IL-6, MCP-1, and IL-1β levels were determined using ELISA kits (R&D Systems, Minneapolis, MN) following the manufacturer's instructions. 2.7. Cell culture The CRL-1927 mouse MC and CCL-139 mouse renal tubuloepithelial cell (TEC) lines were obtained from the American Type Culture Collection (Manassas, VA). CRL-1927 cells were grown in a 3:1 mixture of Dulbecco's modified Eagle's medium (DMEM) and Ham's F-12 medium, supplemented with 5% fetal bovine serum (FBS) and 14 mM HEPES (all from Gibco, Invitrogen, Carlsbad, CA), while CCL-139 cells were maintained in DMEM containing 10% FBS, both at 37 °C in an atmosphere of 5% carbon dioxide. For reactive oxygen species (ROS) production assay, the CRL 1927 mouse MC, the cells (3  103 cells) were grown in a phenol red-free DMEM medium for 24 h, then incubated for 0 min with 10 mM of N-acetyl-L-cysteine (NAC) (Sigma) at 37 °C in the dark in the

216

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

presence of 2 μM of 2′,7′-dichlorofluorescein diacetate (Sigma), then for 0–80 min with or without addition of IgA IC, and expressed relative to that at time 0 as described previously [7]. Peritoneal macrophages were collected from SPAK KO and WT mice that had been injected with 1 ml of 3% thioglycolate (Sigma) and maintained with free access to water and normal chow for 4 days as described previously. [26] Briefly, peritoneal macrophages were collected from the mouse abdomen by lavage with PBS. Cells were cultured for 12 h in RPMI 1640 medium containing 10% FBS (both from Gibco), then all floating cells were removed and the adherent cells cultured in RPMI 1640 medium containing 10% FBS for 2 days, then treated as indicated. For the generation of primary MCs, glomeruli were isolated from the renal cortex of untreated 6- to 8-week-old male SPAK KO and WT mice and primary MCs isolated as described previously [27]. Briefly, glomeruli were purified from minced renal cortex by serial sieving through meshes of different pore sizes, then the suspension was digested for 20 min at 37 °C with type IV collagenase and the dissociated glomerular cells cultured in RPMI 1640 medium containing 20% FBS, 10 mM HEPES, and penicillin/streptomycin (Gibco). The primary MCs had typical morphological characteristics, being positive for both smooth muscle α-actin and vimentin and negative for E-cadherin [27]. The cultured primary MCs were used for experiments between passages 6 and 10. A total of 1  106 cells were incubated 0–60 min or 24 h with preformed IgA IC of IgA anti-PC antibodies (5 ng/ml) and PnC antigen (100 ng/ml) for subsequent tests using ELISA or Western blot analysis as described above. 2.8. Western blot analysis Cytoplasmic and nuclear proteins were extracted from cells using a nuclear extract kit (Cayman, Ann Arbor, MI) according to the manufacturer's instructions. Target proteins were detected by SDSPAGE electrophoresis and immunoblotting using antibodies against mouse p-SPAK (prepared as described above) or p-NF-κB p65, p-p38 MAPK (Cell Signaling Technology), p-JNK (Sigma), p-ERK1/2, p-IKKα/β, p-IκB-α, lamin A, or β-actin (all from Santa Cruz Biotechnology, Dallas, TX), as described previously [2,24]. The results were expressed as the density of the target protein band normalized to that for the internal control band (laminin A or β-actin). 2.9. Cell line proliferation assay The MTT assay was used to measure the proliferation of primary MCs by determining numbers of viable IgA IC-treated cells and expressing this as a percentage of cell numbers in the salinetreated controls. MCs were cultured in 96-well culture plates (1000 cells/well) in 200 μl of RPMI 1640 medium containing 10 mM HEPES, 20% FBS, and antibiotics and incubated with IgA IC for 24 h, then MTT (5 mg/ml, Sigma) was added to each well for an additional 4 h, when the medium was removed and 100 μl of DMSO (Sigma) added to each well. The absorbance was then measured on a spectrophotometer at a wavelength of 570 nm. 2.10. Statistical analysis For the in vivo experiments, the results are presented as the mean 7SEM for the indicated number of mice. The significance of differences between two groups was determined using Student's t-test. Spearman's correlation test was applied to evaluate the correlation between the level of SPAK phosphorylation and total number of mononuclear leukocytes that infiltrated in the periglomerular region of the kidney. For ex vivo and in vitro experiments, all values are given as the mean 7SD and differences were analyzed using one-way ANOVA followed by a Scheffe’ test. A value of p o0.05 was considered statistically significant.

3. Results 3.1. Decreased proteinuria, renal function impairment, and renal lesions in IgAN in SPAK KO mice To investigate a potential pathogenic role of SPAK in IgAN, male SPAK KO mice and their WT littermates were injected intravenously daily for 36 days with purified IgA anti-PC antibodies and PnC to attempt to induce IgAN, then clinical, pathological, and molecular mechanism analyses were performed. As shown in Fig. 1A, the SPAK KO mice induced IgAN (SPAK KO-IgAN mice) showed significantly less albuminuria than the WT mice induced IgAN (WT-IgAN mice). In addition, they had significantly lower serum levels of BUN (Fig. 1B) and Cr (Fig. 1C). Using light microscopy, significantly attenuated renal histopathology was seen in the SPAK KO-IgAN mice compared to WTIgAN mice, which showed wide-spread proliferation (Fig. 1D–H) and sclerosis (Fig. 1D–G and I) in the glomerulus and intense periglomerular mononuclear leukocyte infiltration (Fig. 1D–G and J), as well as scattered tubular atrophy in the tubulointerstitial compartment. The glomerular deposits of IgA was seen in SPAK KOIgAN (total intensity score 120714) and WT-IgAN (total intensity score 125727) mice, although there were no significant differences in the total intensity of glomerular IgA between the both groups. 3.2. Splenic T cell activation/proliferation is inhibited in SPAK KO-IgAN mice SPAK is expressed in CD4 þ T cells [28] and has been implicated in the regulation of T cell activation [29]. In addition, binding of CD28 to the T cell receptor increases the binding of SPAK to phosphokinase C [30]. We therefore examined the impact of SPAK on the biological functions of T cells in the mouse IgAN model. As shown in Fig. 2A, splenocytes from SPAK KO-IgAN mice showed significantly decreased proliferation of T cells compared to those from WT-IgAN mice, as demonstrated by thymidine uptake analysis. When the activation status of T cells was examined using splenocytes and flow cytometry, the results showed that activation of pan T cells (Fig. 2B; CD3 þ CD69 þ ) or helper T cells (Fig. 2C; CD4 þ CD69 þ ) was significantly reduced in SPAK KO-IgAN mice compared to WT-IgAN mice. However, there was no detectable difference in the percentage of activated splenic cytotoxic T cells between SPAK KO-IgAN and WT-IgAN mice (Fig. 2D; CD8 þ CD69 þ ). Examination of possible bias in the type of helper T cell present in SPAK KO-IgAN mice showed that, compared to WTIgAN mice, SPAK KO-IgAN mice had a significantly decreased number of positive cells of CD4 þ IL-2 þ cells (Fig. 2E) and CD4 þ IFN-γ þ cells (Fig. 2F), but not CD4 þ IL-4 þ cells (Fig. 2G). 3.3. Local renal inflammation is reduced in SPAK KO-IgAN mice SPAK is a kinase that can be regulated by the transcription factor NF-κB [17] and is a member of the MAPKs superfamily that is extensively distributed across the body and promotes secretion of inflammatory-related proteins [19,31]. We therefore examined nuclear levels of p-SPAK, p-NF-κB, and p-p38 MAPK in the kidney, as these proteins contribute to several inflammatory and immune responses [7,17,32]. As shown in Fig. 3, staining of kidney sections with antibodies against these proteins showed that nuclear localization of p-SPAK (Fig. 3A–D and M), p-NF-κB p65 (Fig. 3E–H and N), and p-p38 MAPK (Fig. 3A–L and O) was significantly reduced in the glomerular (mainly localized in the mesangial area) and periglomerular region in SPAK KO-IgAN mice compared to WTIgAN mice in which the increased level of SPAK activation was proportionally correlated with the total number of mononuclear leukocytes that infiltrated in the periglomerular region (r ¼0.729,

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

217

Fig. 1. SPAK KO-IgAN mice are protected from proteinuria, renal function impairment, and renal lesions at day 36. (A–C) Albuminuria (A), serum BUN levels (B), and serum Cr levels (C) in WT and SPAK KO mice injected with IgA IC (IgAN) or saline. (D–G) Representative H&E staining of renal sections from the indicated groups of mice. Original magnification 400  (scale bar¼ 50 mm). The arrows indicate periglomerular mononuclear leukocyte infiltration. (H–J) Scoring of glomerular proliferation (H), glomerular sclerosis (I), and periglomerular mononuclear leukocyte infiltration (J). The data are expressed as the mean 7SEM for 12 mice per group. # not detectable. ns, no significant difference. *po 0.05, **po 0.01, and ***p o 0.005.

p o0.005). Because infiltration of mononuclear leukocytes into the kidney contributes to the pathogenesis of IgAN [7,10], we used immunohistochemical staining to evaluate the distribution of F4/ 80 þ macrophages (Fig. 4A–D and M), CD3 þ pan T cells (Fig. 4E–H and N) and CD4 þ helper T cells (Fig. 4I–L and O). Again, the SPAK KO-IgAN mice showed significantly reduced infiltration of macrophages and T cells in the kidney compared to WT-IgAN mice, which showed wide-spread renal infiltration of CD4 þ helper T cells in both the glomerular and tubulointerstitial compartments. 3.4. SPAK KO mice show reduced production of proinflammatory cytokines 3.4.1. In vivo Serum levels of IFN-γ, TNF-α, IL-6, MCP-1, and IL-1β in the mice were measured by ELISA and the results showed that, compared to WT-IgAN mice, SPAK KO-IgAN mice had dramatically reduced serum levels of IFN-γ (Fig. 5A), TNF-α (Fig. 5B), IL-6 (Fig. 5C), MCP-1 (Fig. 5D), and IL-1β (Fig. 5E). 3.4.2. Ex vivo We also examined TNF-α, IL-6, MCP-1, and IL-1β levels in the culture supernatant of peritoneal macrophages isolated from

untreated SPAK KO or WT mice after incubation of the cells for 24 h with preformed IgA IC. As shown in Fig. 6, after incubation with IgA IC, peritoneal macrophages from WT mice showed significantly increased secretion of TNF-α (Fig. 6A), IL-6 (Fig. 6B), MCP-1 (Fig. 6C), and IL-1β (Fig. 6D), but this effect was inhibited in peritoneal macrophages from SPAK KO mice. Furthermore, when primary cultures of MCs from the two mice strains were treated for 24 h with preformed IgA IC, levels of IL-6 (Fig. 7A) and MCP-1 (Fig. 7B) in the culture supernatants were significantly lower in the SPAK KO cultures. Collectively, these data suggest that SPAK plays an underlying pathogenic role in the IgA IC-mediated renal inflammatory response by stimulating the secretion of proinflammatory cytokines by primary MCs and macrophages. 3.5. Reduced activation of NF-κB/MAPKs signaling in renal intrinsic cells from SPAK KO mice MC proliferation is a key pathological features of IgAN [6,7,33] and tubulointerstital lesion/damage is closely contributed to impaired renal function [34–37]. A number of cytokines released by NF-κB and MAPKs are involved in regulating inflammatory and immune responses, including those in IgAN [7,38,39]. In addition, an NF-κB binding site on the SPAK promoter has been shown to

218

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

Fig. 2. Inhibition of splenic T cell activation/proliferation in SPAK KO-IgAN mice at day 36. (A) T cell proliferation measured as [3H]-thymidine incorporation. (B–D) Percentage of activated CD3 þ T cells (B), activated CD4 þ T cells (C), or activated CD8 þ T cells (D) in the total cell population in the spleen; activated cells are CD69 þ . (E–G) Number of CD4 þ T cells expressing IL-2 (E), IL-4 (F), or IFN-γ (G). The data are expressed as the mean 7SEM for 12 mice per group. ns, no significant difference. *p o0.05, **p o0.01, and ***p o 0.005.

Fig. 3. Inhibition of renal NF-κB/MAPKs phosphorylation in SPAK KO-IgAN mice at day 36. (A–L) Immunohistochemical staining for p-SPAK (A–D), p-NF-κB p65 (E–H), or p-p38 MAPK (I–L); original magnification 400x. Arrowheads indicated positively-stained cells in glomerulus. Arrows indicated positively-stained cells in periglomerulus. (M–O) Scoring of cells stained for p-SPAK (M), p-NF-κB p65 (N), or p-p38 MAPK (O) in the glomerulus (cells/glomerular cross-section) and periglomerular region (cells/field). The anti-p-SPAK antibody can also react with p-OSR1, as indicated in the Material and Methods. The data are expressed as the mean 7 SEM for 12 mice per group. # not detectable. ns, no significant difference. *p o 0.05 and ***p o 0.005.

stimulate expression of a SPAK isoform [17] and SPAK has been shown to specifically activate the p38 MAPK pathway [40,41]. We therefore tested the effect of preformed IgA IC on NF-κB/MAPKs signaling in primary MCs isolated from untreated SPAK KO and WT mice by incubating the cells for 0, 30, or 60 min with the complexes. As shown in Fig. 7C and D, a significant increase in p-SPAK levels was seen over the course of the study in WT primary MCs, but not in SPAK KO primary MCs. In addition, WT primary MCs showed a significant increase in levels of phosphorylated IKKα/β,

IκB-α, and NF-κB p65, but this effect was greatly reduced or completely abolished in SPAK KO primary MCs. Similarly, the increase in p-JNK levels seen at 30 and 60 min in the WT cells was absent in the SPAK KO cells, and the increase in p-ERK1/2 and p-p38 MAPK levels seen in WT cells at 30 and 60 min was unaffected at 30 min, but abolished at 60 min in SPAK KO cells. Furthermore, when primary MCs were incubated for 24 h with IgA IC to evaluate the effect of SPAK on MC proliferation, evaluated using the MTT assay, the cells from SPAK KO mice showed significantly

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

219

Fig. 4. Inhibition of renal infiltration by inflammatory cells in SPAK KO-IgAN mice at day 36. (A–L) Immunohistochemistry staining for F4/80 þ macrophages (A–D), CD3 þ pan T cells (E–H), and CD4 þ helper T cells (I–L). Original magnification 400  . Arrowheads indicated positively-stained cells in glomerulus. Arrows indicated positively-stained cells in periglomerulus. (M–O) Scoring of stained cells in the glomerulus (cells/glomerular cross-section) and periglomerulus (cells/field). The data are expressed as the mean 7 SEM for 12 mice per group. # not detectable. ns, no significant difference. *p o 0.05 and ***p o 0.005.

Fig. 5. Reduced serum levels of proinflammatory cytokines in SPAK KO-IgAN mice at day 36. Serum levels of IFN-γ (A) TNF-α (B), IL-6 (C), MCP-1 (D), and IL-1β (E). The data are expressed as the mean 7 SEM for 12 mice per group. # not detectable. *p o0.05, **p o0.01, and ***p o 0.005.

220

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

Fig. 6. Reduced proinflammatory cytokine secretion by IgA IC-primed peritoneal macrophages from untreated WT or SPAK KO mice. Peritoneal macrophages were incubated for 24 h with or without IgA IC, then the culture supernatant was assayed for TNF-α (A), IL-6 (B), MCP-1 (C), or IL-1β (D). The data are expressed as the mean 7 SD for three separate experiments, each performed in triplicate. # not detectable. ns, no significant difference. **po 0.01 and ***p o 0.005.

lower proliferation than those from WT mice (Fig. 7E). Collectively, our data support the idea that SPAK activates NF-κB/MAPKs signal transduction in two major types of renal intrinsic cells in the IgA IC-mediated inflammatory response. Next, we examined whether incubation with preformed IgA IC for 0, 10, 30, or 60 min could trigger activation of the NF-κB/MAPKs signaling pathway in a MC line and a renal TEC line. After stimulation of the MC line with IgA IC, p-SPAK levels were significantly increased at 10 min and this increase persisted up to at least 60 min (Fig. 8A and B), while, in the TEC line, a significant increase was seen at 30 min and again persisted until at least 60 min (Fig. 8A and C). Phosphorylation of IKKα/β, IκB-α, and NF-κB p65 was significantly increased in both cell lines at 30 min and 60 min. In addition, p-ERK1/2 and p-JNK levels were significantly increased at 60 min in MCs and at 30 min and 60 min in renal TECs, while p-p38 MAPK levels in MCs were increased at 30 min and 60 min and in TECs at 60 min. Finally, to investigate the potential role of oxidative stress on the phosphorylation levels of SPAK in IgA IC treated cells, we tested the effect of antioxidant NAC on IgA IC-mediated SPAK phosphorylation. We found that IgA IC induced ROS production in MCs, and this effect was reduced by NAC (Fig. 8D). In addition, IgA IC induced SPAK phosphorylation, and this effect was also reduced by NAC (Fig. 8E). These results indicate that ROS is one of the upstream signals of SPAK phosphorylation in IgA treated mesangial cells.

4. Discussion In the present study, we examined a possible pathophysiological role of SPAK in IgAN using a mouse model and cell models and showed that (1) SPAK deficiency inhibited development of IgAN and this was associated with inhibition of T cell activation and proliferation; (2) IgA IC-primed primary MCs isolated from untreated SPAK KO mice showed reduced production of proinflammatory cytokines and NF-κB/MAPKs activation; and (3) the

IgA IC-primed MC or TEC lines showed increased levels of p-SPAK, p-NF-κB p65, and p-p38 MAPK. We proposed that deposition of IgA IC in the glomerulus led to the activation of SPAK in MCs, macrophages and T cells of the glomerulus as well as renal TECs, and this effect at least partly involved the activation of NF-κB/ MAPKs signaling pathway, which in turn triggers MC proliferation and recruitment of mononuclear leukocytes into the kidney, and the activation of SPAK in renal TECs either directly by IgA IC or an indirect action of the activated MCs or infiltrating mononuclear leukocytes may further aggravate the disease process of IgAN. Mononuclear leukocyte infiltration in the glomeruli [7,10,42] and renal interstitial tissue [13,38] has been shown to play a key pathogenic role in IgAN, and high levels of systemic T cell activation are seen during progression of IgAN [7,10,43]. Furthermore, SPAK is known to be involved in the activation of CD4 þ T cells, [28,29] but it was unclear whether it had an impact on activation of macrophages. We therefore examined whether SPAK is involved in the activation and renal infiltration of mononuclear leukocytes in vivo and/or ex vivo, and found that SPAK deficiency resulted in: (1) reduced infiltration of macrophages and T cells into the periglomerular region (Fig. 4); (2) decreased production of proinflammatory cytokines by peritoneal macrophages (Fig. 6); and (3) inhibition of T cell proliferation, helper T cell activation, and Th1 polarization in the spleen (Fig. 2). Further studies are necessary to identify the renal intrinsic and immune cells types responsible for the effects of SPAK in the IgAN model, as well as the molecular mechanisms involved. In addition, NF-κB has been shown to be overexpressed in renal tubules and glomeruli in IgAN patients [38,44]. Consistent with this finding, our previous studies showed increased activation of NF-κB in a progressive IgAN mouse model [7,10]. In addition, IgA IC activate ERK [39,45] and p38 in human MCs, [45] and high levels of phosphorylated ERK and JNK have been observed in renal biopsies from IgAN patients [39,46]. Together, these findings suggest that the NF-κB/MAPKs signaling pathway plays a pathogenic role in IgAN. Importantly, SPAK has been shown to interact with

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

221

Fig. 7. IgA IC-primed primary MCs from untreated SPAK KO mice show reduced proinflammatory cytokine secretion, reduced phosphorylation of SPAK, NF-κB, and MAPKs, and reduced proliferation. (A and B) Primary MCs were incubated for 24 h with or without IgA IC, then the culture supernatant was assayed by ELISA for IL-6 (A) or MCP-1 (B). (C and D) Primary MCs were incubated for 0–60 min with or without IgA IC, then the cell lysate was assayed for phosphorylated SPAK, IKKα/β, IκB-α, NF-κB, ERK1/2, JNK, and p38 MAPK by Western blot analysis. (C) shows a typical result and (D) the semi-quantitative analysis of phosphorylated SPAK, IKKα/β, IκB-α, ERK1/2, JNK, and p38 MAPK normalized to β-actin levels, and phosphorylated NF-κB normalized to Lamin A levels. (E) Primary MCs were incubated for 24 h with or without IgA IC, then the cells were assayed for cell proliferation measured using the MTT assay. The data are expressed as the number of viable IgA IC-treated cells expressed as a percentage of that for the saline-treated cells and are the mean 7 SD for three separate experiments, each performed in triplicate. *p o 0.05,**p o0.01, and ***p o 0.005.

222

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

Fig. 8. Increased phosphorylation of SPAK, NF-κB, and MAPKs in IgA IC-primed mouse mesangial and renal tubuloepithelial cell lines. Cells were incubated for 0–60 min with or without IgA IC, then the cell lysates from the mesangial cell line (left panels) and renal tubuloepithelial cell line (right panels) were assayed for phosphorylated SPAK, IKKα/β, IκB-α, NF-κB, ERK1/2, JNK, and p38 MAPK by Western blot analysis. (A) shows a typical result and (B, C) the semi-quantitative analysis of (top to bottom) phosphorylated SPAK, IKKα/β, IκB-α, ERK1/2, JNK, and p38 MAPK normalized to β-actin levels, and phosphorylated NF-κB normalized to Lamin A levels. (D) ROS levels in MCs measured by 2’, 7’-dichlorofluorescein diacetate after incubated for 0–80 min with IgA IC. (E) Antioxidant NAC on IgA IC-mediated SPAK phosphorylation. MCs were incubated for 30 min with 10 mM of NAC and then 30 min with IgA IC, the cell lysate was assayed for phosphorylated SPAK by Western blot analysis. The data are expressed as the mean 7 SD for three separate experiments, each performed in triplicate. ns, no significant difference. *p o 0.05, **p o 0.01, and ***p o 0.005.

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

(1) NF-κB [17,47], as NF-κB has a binding site that is essential for TNF-α induced SPAK promoter activity [17]; and (2) p38 MAPK [19], as SPAK can activate the p38 MAPK pathway to promote cell differentiation [18]. Furthermore, SPAK is involved in the production of proinflammatory cytokines [4,5,18]. In the present study, although secretion of IL-6 and MCP-1 was increased after incubation of primary MCs from WT mice with IgA IC, this effect was not seen with primary MCs from SPAK KO mice (Fig. 7), suggesting that SPAK activation is involved in the activation/proliferation of MCs and their release of proinflammatory cytokines. Consistent with this observation, we demonstrated activation of SPAK, NF-κB, and p38 MAPK in (1) glomeruli in predominantly mesangial areas of the kidney in WT-IgAN mice (Fig. 3); (2) IgA IC-primed primary MCs isolated from WT mice (Fig. 7); and (3) IgA IC-primed MC and TEC lines (Fig. 8). These findings support the involvement of NF-κB and p38 MAPK in the nephritogenic effect of SPAK in the development of IgAN. In addition, SPAK has been reported to play a downstream substrate of WNK kinases [1–3], which is a phosphorylation target of PI3-kinase-dependent AKT kinase [48–50]. AKT kinase also contributes to the proliferation of human MCs [51– 53]. Thus, in the present study, although NAC almost completely abolished ROS, the phosphorylation of SPAK was incompletely restored in MCs (Fig. 8D), and whether there are some other pathways, such as those of WNK and AKT kinases, involved in the phosphorylated SPAK in the renal intrinsic cells needs further investigation. In this study, we used a model induced passively by injection of IgA and PnC, but this model may not necessarily be related to the mechanism of injury in IgAN patients, which is now considered more likely to be mediated by abnormalities in IgA1 glycosylation, leading to IC formation [54–56]. However, the mouse model still may represent IgAN in humans to a certain degree, as it was the first experimental model to reproduce the characteristic granular immunofluorescence pattern of IgA and C3 mesangial deposits, with renal inflammation and fibrosis [6,7,10,20,21] and has been instrumental in: (1) elucidating the role of different IgA subtypes in IgA IC glomerular deposition; (2) clarifying the role of complement activation and IC deposition; (3) establishing the clearance kinetics of circulating IgA IC and the role of hepatic Kupffer cells in their elimination; (4) showing that the antigen in complexes with IgA is the primary determinant in renal injury; and (5) demonstrating synergy between extra-renal cytokines and IgA mesangial deposits in the development of renal injury and dysfunction and in the evolution of renal histopathologic changes initiated by IgA IC mesangial deposition, induction of mesangial proliferation leading to glomerulosclerosis, and interstitial inflammation and fibrosis. In addition, SPAK KO mice have been reported to be a model of hypotension in Gitelman syndrome, as they spontaneously show a low blood pressure [2], and whether this contributes to the protective effect against IgAN seen in SPAK KO mice deserves further investigation. The IgA IC used throughout the present study can directly cause inflammation in the renal TECs, and PnC (the antigen that was contained in the IgA IC) could play a major function in this kind of stimulation. It is suggested that in this IgA IC, the nature of antigen may play a key role in resultant injury of the renal TECs, [6,57–60] although the possibility of the plausible IgA receptor in the epithelial cells may also play a role in activation the renal TECs. In addition, the effect of blocking SPAK on the inflammation of the cells deserves further investigation. In summary, we report that p-SPAK levels are significantly increased in IgAN and that knockout of SPAK decreased the severity of the renal histopathology in the IgAN mouse model, suggesting that SPAK mediated the development of IgA nephropathy through the activation of NF-κB/MAPKs signaling pathway. SPAK plays a

223

pathogenic role in IgAN and that targeting SPAK-associated signaling pathways may be a strategy for the treatment of IgAN.

5. Conclusions We proposed that activation of SPAK and the NF-κB/MAPKs signaling pathway in MCs, macrophages and T cells of the glomerulus may be a mechanism underlying the pathogenesis of IgAN. The activation of SPAK in renal tubuloepithelial cells either directly by IgA IC or an indirect action of the activated MCs or infiltrating mononuclear leukocytes seen in the kidney may further aggravate the disease process of IgAN. Our results suggest that SPAK is a potential therapeutic target for the glomerular disorder.

Acknowledgments This study was supported by grants from the Ministry of Science and Technology (MOST 103-2321-B-016-002; 103-2314-B016-005-MY3) and the Medical Affairs Bureau (104-M012, 105006), Taipei, Taiwan, ROC.

References [1] S.S. Yang, T. Morimoto, T. Rai, M. Chiga, E. Sohara, M. Ohno, K. Uchida, S.H. Lin, T. Moriguchi, H. Shibuya, Y. Kondo, S. Sasaki, S. Uchida, Molecular pathogenesis of pseudohypoaldosteronism type II: generation and analysis of a Wnk4 (D561A/ þ) knockin mouse model, Cell Metab. 5 (2007) 331–344. [2] S.S. Yang, Y.F. Lo, C.C. Wu, S.W. Lin, C.J. Yeh, P. Chu, H.K. Sytwu, S. Uchida, S. Sasaki, S.H. Lin, SPAK-knockout mice manifest Gitelman syndrome and impaired vasoconstriction, J. Am. Soc. Nephrol. 21 (2010) 1868–1877. [3] S.H. Lin, I.S. Yu, S.T. Jiang, S.W. Lin, P. Chu, A. Chen, H.K. Sytwu, E. Sohara, S. Uchida, S. Sasaki, S.S. Yang, Impaired phosphorylation of Na(þ )-K(þ )  2Cl (  ) cotransporter by oxidative stress-responsive kinase-1 deficiency manifests hypotension and Bartter-like syndrome, Proc. Natl. Acad. Sci. USA 108 (2011) 17538–17543. [4] Y. Yan, H. Laroui, S.A. Ingersoll, S. Ayyadurai, M. Charania, S. Yang, G. Dalmasso, T.S. Obertone, H. Nguyen, S.V. Sitaraman, D. Merlin, Overexpression of Ste20related proline/alanine-rich kinase exacerbates experimental colitis in mice, J. Immunol. 187 (2011) 1496–1505. [5] Y. Zhang, E. Viennois, B. Xiao, M.T. Baker, S. Yang, I. Okoro, Y. Yan, Knockout of Ste20-like proline/alanine-rich kinase (SPAK) attenuates intestinal inflammation in mice, Am. J. Pathol. 182 (2013) 1617–1628. [6] T.K. Chao, A. Rifai, S.M. Ka, S.M. Yang, H.A. Shui, Y.F. Lin, H.K. Sytwu, W.H. Lee, J. T. Kung, A. Chen, The endogenous immune response modulates the course of IgA-immune complex mediated nephropathy, Kidney Int. 70 (2006) 283–297. [7] S.M. Yang, S.M. Ka, K.F. Hua, T.H. Wu, Y.P. Chuang, Y.W. Lin, F.L. Yang, S.H. Wu, S. S. Yang, S.H. Lin, J.M. Chang, A. Chen, Antroquinonol mitigates an accelerated and progressive IgA nephropathy model in mice by activating the Nrf2 pathway and inhibiting T cells and NLRP3 inflammasome, Free Radic. Biol. Med. 61 (2013) 285–297. [8] J.M. Ambruzs, P.D. Walker, C.P. Larsen, The histopathologic spectrum of kidney biopsies in patients with inflammatory bowel disease, Clin. J. Am. Soc. Nephrol. 9 (2014) 265–270. [9] L.A. van Es, E. de Heer, L.J. Vleming, A. van der Wal, M. Mallat, I. Bajema, J. A. Bruijn, J.W. de Fijter, GMP-17-positive T-lymphocytes in renal tubules predict progression in early stages of IgA nephropathy, Kidney Int. 73 (2008) 1426–1433. [10] S.M. Ka, T.T. Hsieh, S.H. Lin, S.S. Yang, C.C. Wu, H.K. Sytwu, A. Chen, Decoy receptor 3 inhibits renal mononuclear leukocyte infiltration and apoptosis and prevents progression of IgA nephropathy in mice, Am. J. Physiol. Ren. Physiol. 301 (2011) F1218–F1230. [11] A. Rifai, L.D. Dworkin, IgA nephropathy: markers of progression and clues to pathogenesis, Kidney Int. 73 (2008) 1338–1340. [12] M. Walsh, A. Sar, D. Lee, S. Yilmaz, H. Benediktsson, B. Manns, B. Hemmelgarn, Histopathologic features aid in predicting risk for progression of IgA nephropathy, Clin. J. Am. Soc. Nephrol. 5 (2010) 425–430. [13] M.C. Falk, G. Ng, G.Y. Zhang, G.C. Fanning, L.P. Roy, K.M. Bannister, A.C. Thomas, A.R. Clarkson, A.J. Woodroffe, J.F. Knight, Infiltration of the kidney by alpha beta and gamma delta T cells: effect on progression in IgA nephropathy, Kidney Int. 47 (1995) 177–185. [14] Y. Kanamaru, M. Arcos-Fajardo, I.C. Moura, T. Tsuge, H. Cohen, M. Essig, F. Vrtovsnik, C. Loirat, M. Peuchmaur, L. Beaudoin, P. Launay, A. Lehuen, U. Blank, R.C. Monteiro, Fc alpha receptor I activation induces leukocyte recruitment and promotes aggravation of glomerulonephritis through the FcR

224

T.-J. Lin et al. / Free Radical Biology and Medicine 99 (2016) 214–224

gamma adaptor, Eur. J. Immunol. 7 (2007) 1116–1128. [15] Y. Ikezumi, T. Suzuki, T. Karasawa, H. Hasegawa, T. Yamada, N. Imai, I. Narita, H. Kawachi, K.R. Polkinghorne, D.J. Nikolic-Paterson, M. Uchiyama, Identification of alternatively activated macrophages in new-onset paediatric and adult immunoglobulin A nephropathy: potential role in mesangial matrix expansion, Histopathology 58 (2011) 198–210. [16] H.J. Lin, C.P. Wu, C.K. Peng, S.H. Lin, S. Uchida, S.S. Yang, K.L. Huang, ith-NoLysine Kinase 4 mediates alveolar fluid regulation in hyperoxia-induced lung injury, Crit. Care Med. 43 (2015) 412–419. [17] Y. Yan, G. Dalmasso, H.T. Nguyen, T.S. Obertone, L. Charrier-Hisamuddin, S. V. Sitaraman, D. Merlin, Nuclear factor-kappaB is a critical mediator of Ste20like proline-/alanine-rich kinase regulation in intestinal inflammation, Am. J. Pathol. 173 (2008) 1013–1028. [18] Y. Yan, D. Merlin, Ste20-related proline/alanine-rich kinase: a novel regulator of intestinal inflammation, World J. Gastroenterol. 14 (2008) 6115–6121. [19] M.H. Chiu, H.S. Liu, Y.H. Wu, M.R. Shen, C.Y. Chou, PAK mediates KCC3-enhanced cervical cancer tumorigenesis, FEBS J. 281 (2014) 2353–2365. [20] V. Montinaro, L. Aventaggiato, T. Cavallo, A. Rifai, Evolution of renal injury in a chronic model of IgA immune-complex-associated nephropathy, Nephrol. Dial. Transpl. 10 (1995) 2035–2042. [21] A. Rifai, P.A. Small Jr, P.O. Teague, E.M. Ayoub, Experimental IgA nephropathy, J. Exp. Med. 150 (1979) 1161–1173. [22] S.M. Ka, X.R. Huang, H.Y. Lan, P.Y. Tsai, S.M. Yang, H.A. Shui, A. Chen, Smad7 gene therapy ameliorates an autoimmune crescentic glomerulonephritis in mice, J. AM Soc. Nephrol. 18 (2007) 1777–1788. [23] T. Moriguchi, S. Urushiyama, N. Hisamoto, S. Iemura, S. Uchida, T. Natsume, K. Matsumoto, H. Shibuya, WNK1 regulates phosphorylation of cation-chloride-coupled cotransporters via the STE20-related kinases, SPAK and OSR1, J. Biol. Chem. 280 (2005) 42685–42693. [24] S.M. Yang, S.M. Ka, H.L. Wu, Y.C. Yeh, C.H. Kuo, K.F. Hua, G.Y. Shi, Y.J. Hung, F. C. Hsiao, S.S. Yang, Y.S. Shieh, S.H. Lin, C.W. Wei, J.S. Lee, C.Y. Yang, A. Chen, Thrombomodulin domain 1 ameliorates diabetic nephropathy in mice via anti-NF-κB/NLRP3 inflammasome-mediated inflammation, enhancement of NRF2 antioxidant activity and inhibition of apoptosis, Diabetologia 57 (2014) 424–434. [25] S.M. Ka, H.K. Sytwu, D.M. Chang, S.L. Hsieh, P.Y. Tsai, A. Chen, Decoy receptor 3 ameliorates an autoimmune crescentic glomerulonephritis model in mice, J. Am. Soc. Nephrol. 18 (2007) 2473–2485. [26] X. Zhou, Z. Yin, X. Guo, D.P. Hajjar, J. Han, Inhibition of ERK1/2 and activation of liver X receptor synergistically induce macrophage ABCA1 expression and cholesterol efflux, J. Biol. Chem. 285 (2010) 6316–6326. [27] S.M. Ka, C.W. Cheng, H.A. Shui, W.M. Wu, D.M. Chang, Y.C. Lin, A. Chen, Mesangial cells of lupus-prone mice are sensitive to chemokine production, Arthritis Res. Ther. 9 (2007) R67. [28] J. Cardone, G. Le Friec, P. Vantourout, A. Roberts, A. Fuchs, I. Jackson, T. Suddason, G. Lord, J.P. Atkinson, A. Cope, A. Hayday, C. Kemper, Complement regulator CD46 temporally regulates cytokine production by conventional and unconventional T cells, Nat. Immunol. 11 (2010) 862–871. [29] H. Yamamoto, A.F. Fara, P. Dasgupta, C. Kemper, CD46: the ‘multitasker’ of complement proteins, Int J. Biochem. Cell Biol. 45 (2013) 2808–2820. [30] Y. Li, J. Hu, R. Vita, B. Sun, H. Tabata, A. Altman, SPAK kinase is a substrate and target of PKC-θ in T-cell receptor-induced AP-1 activation pathway, EMBO J. 23 (2004) 1112–1122. [31] S.S. Yang, C.L. Huang, H.E. Chen, C.S. Tung, H.P. Shih, Y.P. Liu, Effects of SPAK knockout on sensorimotor gating, novelty exploration, and brain area-dependent expressions of NKCC1 and KCC2 in a mouse model of schizophrenia, Prog. Neuropsychopharmacol. Biol. Psychiatry 61 (2015) 30–36. [32] B. Zhou, D. Wang, X. Feng, Y. Zhang, Y. Wang, J. Zhuang, X. Zhang, G. Chen, E. Delpire, D. Gu, H. Cai, WNK4 inhibits NCC protein expression through MAPK ERK1/2 signaling pathway, Am. J. Physiol. Ren. Physiol. 302 (2012) F533–F539. [33] H. Tamouza, F. Vende, M. Tiwari, M. Arcos-Fajardo, F. Vrtovsnik, M. Benhamou, R.C. Monteiro, I.C. Moura, Transferrin receptor engagement by polymeric IgA1 induces receptor expression and mesangial cell proliferation: role in IgA nephropathy, Contrib. Nephrol. 157 (2007) 144–147. [34] G. Bollée, D. Anglicheau, A. Loupy, J. Zuber, N. Patey, D. Mac Gregor, F. Martinez, M.F. Mamzer-Bruneel, R. Snanoudj, E. Thervet, C. Legendre, L. H. Noël, High-dosage intravenous immunoglobulin-associated macrovacuoles are associated with chronic tubulointerstitial lesion worsening in renal transplant recipients, Clin. J. Am. Soc. Nephrol. 3 (2008) 1461–1468. [35] K. Zuo, Y. Wu, S.J. Li, F. Xu, C.H. Zeng, Z.H. Liu, Long-term outcome and prognostic factors of idiopathic membranous nephropathy in the Chinese population, Clin. Nephrol. 79 (2013) 445–453. [36] Working Group of the International IgA Nephropathy Network and the Renal Pathology Society, R. Coppo, S. Troyanov, R. Camilla, R.J. Hogg, D.C. Cattran, H. T. Cook, J. Feehally, I.S. Roberts, A. Amore, C.E. Alpers, J. Barratt, F. Berthoux, S. Bonsib, J.A. Bruijn, V. D’Agati, G. D'Amico, S.N. Emancipator, F. Emma, F. Ferrario, F.C. Fervenza, S. Florquin, A.B. Fogo, C.C. Geddes, H.J. Groene, M. Haas, A.M. Herzenberg, P.A. Hill, S.I. Hsu, J.C. Jennette, K. Joh, B.A. Julian, T. Kawamura, F.M. Lai, L.S. Li, P.K. Li, Z.H. Liu, S. Mezzano, F.P. Schena, Y. Tomino, P.D. Walker, H. Wang, J.J. Weening, N. Yoshikawa, H. Zhang, The

[37] [38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49] [50]

[51]

[52]

[53]

[54]

[55] [56]

[57]

[58]

[59]

[60]

Oxford IgA nephropathy clinicopathological classification is valid for children as well as adults, Kidney Int. 77 (2010) 921–927. K.N. Lai, L.Y. Chan, J.C. Leung, Mechanisms of tubulointerstitial injury in IgA nephropathy, Kidney Int. Suppl. 94 (2005) S110–S115. G.E. Silva, R.S. Costa, R.C. Ravinal, L.Z. Ramalho, M.A. Dos Reis, T.M. Coimbra, M. Dantas, NF-kB expression in IgA nephropathy outcome, Dis. Markers 31 (2011) 9–15. H. Tamouza, J.M. Chemouny, L. Raskova Kafkova, L. Berthelot, M. Flamant, M. Demion, L. Mesnard, E. Paubelle, F. Walker, B.A. Julian, E. Tissandié, M. K. Tiwari, N.O. Camara, F. Vrtovsnik, M. Benhamou, J. Novak, R.C. Monteiro, I. C. Moura, The IgA1 immune complex-mediated activation of the MAPK/ERK kinase pathway in mesangial cells is associated with glomerular damage in IgA nephropathy, Kidney Int. 82 (2012) 1284–1296. A.M. Johnston, G. Naselli, L.J. Gonez, R.M. Martin, L.C. Harrison, H. J. DeAizpurua, SPAK, a STE20/SPS1-related kinase that activates the p38 pathway, Oncogene 19 (2000) 4290–4297. T.C. Polek, M. Talpaz, T. Spivak-Kroizman, The TNF receptor, RELT, binds SPAK and uses it to mediate p38 and JNK activation, Biochem. Biophys. Res. Commun. 343 (2006) 125–134. E. Sabadini, A. Castiglione, G. Colasanti, F. Ferrario, R. Civardi, G. Fellin, G. D’Amico, Characterization of interstitial infiltrating cells in Berger’s disease, Am. J. Kidney Dis. 12 (1988) 307–315. L. Zhang, Y. Wang, X. Shi, H. Zou, Y. Jiang, A higher frequency of CD4⁺CXCR5⁺ T follicular helper cells in patients with newly diagnosed IgA nephropathy, Immunol. Lett. 158 (2014) 101–108. J.C. Leung, S.C. Tang, L.Y. Chan, W.L. Chan, K.N. Lai, Synthesis of TNF-alpha by mesangial cells cultured with polymeric anionic IgA – role of MAPK and NFkappaB, Nephrol. Dial. Transpl. 23 (2008) 72–78. C. Gómez-Guerrero, O. López-Franco, Y. Suzuki, G. Sanjuán, P. HernándezVargas, J. Blanco, J. Egido, Nitric oxide production in renal cells by immune complexes: role of kinases and nuclear factor-kappaB, Kidney Int. 62 (2002) 2022–2034. M.H. De Borst, J. Prakash, W.B. Melenhorst, M.C. van den Heuvel, R.J. Kok, G. Navis, H. van Goor, Glomerular and tubular induction of the transcription factor c-Jun in human renal disease, J. Pathol. 213 (2007) 219–228. Y. Yan, G. Dalmasso, H.T. Nguyen, T.S. Obertone, S.V. Sitaraman, D. Merlin, Ste20-related proline/alanine-rich kinase (SPAK) regulated transcriptionally by hyperosmolarity is involved in intestinal barrier function, PLoS One 4 (2009) e5049. C.J. Cheng, C.L. Huang, Activation of PI3-kinase stimulates endocytosis of ROMK via Akt1/SGK1-dependent phosphorylation of WNK1, J. Am. Soc. Nephrol. 22 (2011) 460–471. J.A. McCormick, D.H. Ellison, The WNKs: atypical protein kinases with pleiotropic actions, Physiol. Rev. 91 (2011) 177–471. H. Nishida, E. Sohara, N. Nomura, M. Chiga, D.R. Alessi, T. Rai, S. Sasaki, S. Uchida, Phosphatidylinositol 3-kinase/Akt signaling pathway activates the WNK-OSR1/SPAK-NCC phosphorylation cascade in hyperinsulinemic db/db mice, Hypertension 60 (2012) 981–990. F. Das, N. Ghosh-Choudhury, N. Dey, A. Bera, M.M. Mariappan, B.S. Kasinath, G. Ghosh Choudhury, High glucose forces a positive feedback loop connecting Akt kinase and FoxO1 transcription factor to activate mTORC1 kinase for mesangial cell hypertrophy and matrix protein expression, J. Biol. Chem. 289 (2014) 32703–32716. N. Zheng, D. Wang, H. Ming, H. Zhang, X. Yu, BAFF promotes proliferation of human mesangial cells through interaction with BAFF-R, BMC Nephrol. 15 (2015) 72. F. Das, N. Ghosh-Choudhury, M.M. Mariappan, B.S. Kasinath, G.G. Choudhury, Hydrophobic motif site-phosphorylated protein kinase CβII between mTORC2 and Akt regulates high glucose-induced mesangial cell hypertrophy, Am. J. Physiol. Cell Physiol. 310 (2016) C583–C596. Y. Suzuki, H. Suzuki, J. Yasutake, Y. Tomino, Paradigm shift in activity assessment of IgA nephropathy – optimizing the next generation of diagnostic and therapeutic maneuvers via glycan targeting, Expert Opin. Biol. Ther. 15 (2015) 583–593. K. Kiryluk, J. Novak, The genetics and immunobiology of IgA nephropathy, J. Clin. Invest 124 (2014) 2325–2332. J.K. Boyd, C.K. Cheung, K. Molyneux, J. Feehally, J. Barratt, An update on the pathogenesis and treatment of IgA nephropathy, Kidney Int. 81 (2012) 833–843. A. Chen, L.F. Sheu, M.F. Shaio, T.H. Huang, S.L. Ding, W.H. Lee, Anti-idiotype antibody directly interferes with glomerular IgA immune complex deposition, Nephron 78 (1998) 464–473. A. Chen, S.L. Ding, L.F. Sheu, Y.B. Song, S.D. Shieh, M.F. Shaio, W.Y. Chou, Y.S. Ho, Experimental IgA nephropathy. Enhanced deposition of glomerular IgA immune complex in proteinuric states, Lab. Investig. 70 (1994) 639–647. A. Chen, C.H. Wei, W.H. Lee, C.Y. Lin, Experimental IgA nephropathy: factors influencing IgA-immune complex deposition in the glomerulus, Springer Semin. Immunopathol. 16 (1994) 97–103. H. Imai, A. Chen, R.J. Wyatt, A.: Rifai, Lack of complement activation by human IgA immune complexes, Clin. Exp. Immunol. 73 (1988) 479–483.