Original Investigation

a 1-Acid Glycoprotein Attenuates Adriamycin-Induced Nephropathy via CD163 Expressing Induction

Rui Fujimura,1,2 Hiroshi Watanabe,1 Kento Nishida,1 Yukio Fujiwara,3 Tomoaki Koga,4 Jing Bi,1,2 Tadashi Imafuku,1,2 Kazuki Kobayashi,1 Hisakazu Komori,1 Masako Miyahisa,1 Hitoshi Maeda,1 Motoko Tanaka,5 Kazutaka Matsushita,5 Takashi Wada,6 Masafumi Fukagawa,7 and Toru Maruyama1

Abstract Background Recent clinical studies have shown that proteinuria is a critical factor in the progression of CKD and onset of cardiovascular disease. Inflammation and infiltration of into renal tissue are implicated as causes of proteinuria. a1-Acid glycoprotein (AGP), an acute-phase plasma , is leaked into the urine in patients with proteinuria. However, the relationship between urinary leakage of AGP, renal inflammation, and proteinuria remains unclear.

Methods Human AGP (hAGP) was exogenously administrated for 5 consecutive days to adriamycin-induced nephropathy model mice.

Results Adriamycin treatment increased urinary AGP, accompanied by decreased plasma AGP in mice. Exogenous hAGP administration to adriamycin-treated mice suppressed proteinuria, renal histologic injury, and inflammation. hAGP administration increased renal CD163 expression, a marker of anti-inflammatory macrophages. Similar changes were observed in PMA-differentiated THP-1 cells treated with hAGP. Even in the presence of LPS, hAGP treatment increased CD163/IL-10 expression in differentiated THP-1 cells.

Conclusions AGP alleviates proteinuria and renal injury in mice with proteinuric kidney disease via induction of CD163-expressing macrophages with anti-inflammatory function. The results demonstrate that endogenous AGP could work to protect against glomerular disease. Thus, AGP supplementation could be a possible new therapeutic intervention for patients with glomerular disease. KIDNEY360 1: 343–353, 2020. doi: https://doi.org/10.34067/KID.0000782019

Introduction persistent proteinuria and inflammation, for the pre- Approximately 13% of the worldwide population suf- vention of CKD progression and onset of CVD. fer from CKD, and the number of patients is increasing Recently, the involvement of macrophages in inflam- year by year (1). CKD not only leads to ESKD, but is mation in renal pathologies has been reported (8,9). also associated with a high risk of death from cardio- Macrophages have phenotypes that are classify them vascular disease (CVD) (2). In previous clinical trials, as M1 macrophages (classically activated macro- proteinuria was reported as a critical factor involved in phages) or M2 macrophages (alternatively activated the progression of CKD and onset of CVD (224). In macrophages). In general, M1 macrophages promote fact, nephrotic patients with massive proteinuria have inflammation and play a central role in host defense a 5.5-fold higher risk of CVD than matched controls (5). during infection. In contrast, M2 macrophages are in- It has been reported that inflammation and infiltration volved in anti-inflammation and tissue remodeling of immune cells into renal tissue are a cause of pro- (10). These phenotypes can be changed in response teinuria. In addition, the onset of proteinuria further to a disease state. M1 macrophages predominate in the induces inflammation in tubular cells (6,7). Therefore, early injury phase, whereas M2 macrophages predom- it is important to stop the vicious cycle induced by inate in the recovery and repair phase (11). Indeed,

1Department of Biopharmaceutics, Graduate School of Pharmaceutical Sciences, Kumamoto University, Kumamoto, Japan 2Program for Leading Graduate Schools “HIGO (Health life science: Interdisciplinary and Global Oriented) Program,” Kumamoto University, Kumamoto, Japan 3Department of Cell Pathology, Graduate School of Medical Sciences, Kumamoto University, Kumamoto, Japan 4Department of Medical Cell Biology, Institute of Molecular Embryology and Genetics, Kumamoto University, Kumamoto, Japan 5Department of Nephrology, Akebono Clinic, Kumamoto, Japan 6Department of Nephrology and Laboratory Medicine, Faculty of Medicine, Institute of Medical, Pharmaceutical and Health Sciences, Graduate School of Medical Sciences, Kanazawa University, Kanazawa, Japan 7Division of Nephrology, Endocrinology and Metabolism, Tokai University School of Medicine, Kanagawa, Japan

Correspondence: Hiroshi Watanabe or Toru Maruyama, Department of Biopharmaceutics, Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1, Oe-honmachi, Chuo-ku, Kumamoto 862-0973, Japan. E-mail: [email protected] or tomaru@ gpo.kumamoto-u.ac.jp www.kidney360.org Vol 1 May, 2020 Copyright © 2020 by the American Society of Nephrology 343 344 KIDNEY360

previous reports have demonstrated that administration of Animal Studies M2 macrophages induced by IL-4 or IL-4/IL-13 is effective All animal experiments were performed according to the against renal injury resulting from ischemia reperfusion or guidelines, principles, and procedures for the care and use adriamycin (ADR)-induced nephropathy in mice (11,12). of laboratory animals of Kumamoto University. All animals These studies suggest that regulation of macrophage phe- were housed in an environment with controlled tempera- notypes in the kidney could be important in reducing renal ture, a 12-hour /dark cycle, and free access to food inflammation and proteinuria. and water. a1-Acid glycoprotein (AGP), also known as orosomucoid, Six-week-old male BALB/c mice were obtained from is a heavily glycosylated serum protein with five branched Japan SLC, Inc. (Shizuoka, Japan). After randomizing mice N-glycans containing negatively charged sialic acids (13,14). by body weight, mice were slowly injected intravenously Therefore, AGP has the lowest pI value of all plasma pro- with saline or ADR (15 mg/kg body weight). The first teins (pI52.7–3.2). In addition, AGP is classified as one of administration of saline or AGP (5 mg/mouse) was intra- the major acute-phase in mammals and is known peritoneally administered between 30 minutes and 1 hour that its serum concentration increases two- to five-fold after ADR injection, allowing for time for the ADR to during inflammation (15). In patients with proteinuria, it disappear from the blood. Subsequently, saline or hAGP has been observed that AGP leaks into urine (16,17). On the was administered at the same time for 4 consecutive days. other hand, AGP has been reported to possesses unique No mice died under these experimental conditions. On day biologic activities. For example, in vitro experiments have 7 and day 21 after ADR administration, urine was collected demonstrated that AGP modulates the activation of neu- by metabolic cage for a total of 16 hours from 4:00 PM to 8:00 trophils, lymphocytes, and (214,18–21). Daemen AM, and the urine volume measured. Mice were sacrificed et al. (22) reported that exogenously administered AGP at day 7 or day 21 after ADR administration. Kidneys and attenuates renal ischemia reperfusion injury in vivo, al- livers were harvested after perfusion with 0.9% NaCl. A part though its renoprotective mechanisms remain unclear. Im- of the left kidney was used for quantitative RT-PCR and portantly, we have previously found that AGP weakly histologic analysis. activates the Toll-like receptor 4 (TLR4)/CD14 pathway and increases the expression of CD163, a cysteine-rich scav- Renal Function enger receptor in macrophages, particularly of the M2 Urinary albumin was measured by ELISA (Shibayagi, phenotype (23,24). Interestingly, CD163-overexpressing Japan). BUN, creatinine, and total cholesterol were mea- fl macrophages inhibit LPS-induced in ammation in vitro sured using FUJI DRI-CHEM (Fujifilm, Japan). (25). Furuhashi et al. (26) also found that CD163-expressing macrophages are associated with IL-10 production as an anti- inflammatory cytokine in anti-glomerular basement mem- Histologic Analysis brane GN rats. These findings have led to the idea that Sections of renal tissue were stained with periodic acid- AGP may potentiate the suppression of renal inflammation Schiff. Quantitative evaluation of glomerulosclerosis and and proteinuria by changing the macrophage phenotype to damaged tubules was performed according to the report one with enhanced CD163 expression. of Cao et al. (28). Imaging and analyses were performed The purpose of this study is to clarify whether exoge- using a Keyence BZ-X710 microscope (Keyence, Osaka, – nously administered human AGP (hAGP) attenuates renal Japan). The images were randomly acquired, with 12 fi damage in ADR-induced nephropathy mice. In addition, 17 high-power elds collected for each mouse and then possible mechanisms of this renoprotective effect are exam- 21 glomeruli randomly traced within them. The percentage ined. Here, we provide evidence that administration of of glomerulosclerosis was computed from the area that was hAGP induces CD163-expressing macrophages in the kid- periodic acid-Schiff-positive in the total area of the same ney, which leads to reduced proteinuria and renal inflam- glomerulus. For analysis of damaged tubules, 10 high- fi mation in ADR-induced nephropathy mice. power eld images were randomly acquired of the renal cortex of each mouse. Three people counted damaged tubules, which revealed intraluminal casts, vacuolization, and droplets, and the percentage of damaged tubules was Materials and Methods calculated using the number of damaged tubules divided by the total number of tubules. This analysis was performed in Purification of AGP hAGP was purified from supernatant of human plasma a blind manner. fraction V donated by KM Biologics (Kumamoto, Japan). Supernatant was diluted with acetate buffer (10 mM) and Immunostaining Assay injected onto an ion exchange column (Hitrap CM,Q col- Infiltration of macrophages was confirmed by immuno- umn; GE Healthcare, Little Chalfont, UK) at a flow rate of staining used anti-EMR1 antibody (F4/80; Dako, Santa 3 ml/min on an AKTAprime Plus System (GE Healthcare). Clara, CA). For each mouse, 10–12 high-power fields were The bound materials were eluted with acetate buffer (10 randomly imaged and the number of F4/801 cells was mM) containing NaCl (0.5 mol/L). hAGP was eluted in the counted. The expression of nephrin and the renal localiza- first peak. The eluted sample was dialyzed against distilled tion of AGP were confirmed by immunofluorescence using water at 4°C, freeze-dried, and stored at 220°C (27). The nephrin and ORM1 antibodies. Deparaffinized sections purity of hAGP was confirmed by SDS-PAGE. The sec- were inactivated with Histo VT One (Nacalai Tesque, Kyoto, ondary structure of hAGP was confirmed by circular Japan) and incubated with nephrin (1:100; R&D Systems, dichroism. Inc., Minneapolis, MN) or ORM1 antibody (1:50; Proteintech, KIDNEY360 1: 343–353, May, 2020 AGP prevents proteinuria via CD163+ macrophages, Fujimura et al. 345

A BC ADR Sacrifice 4000 ** * 400 ** *

3000 300 ** ** 2000 200 01234 7 21 (Day) (mg / dL) (mg / mmol) 1000 100 hAGP Total plasma cholesterol

Urinary albumin/ creatinine 0 0

ADR ADR ADR Control Control Control

ADR+hAGP ADR+hAGP ADR+hAGP Day7 Day21

D Control ADR ADR+hAGP E Control ADR ADR + hAGP

40 ** 60 150 P=0.0507 ** ** ** ** * 30 40 100 20 20 50

10 (% of Control) Damaged Tubles (%) Fluorescence Intensity Glomerulosclerosis (%) 0 0 0

ADR ADR ADR Control Control Control

ADR+ hAGP ADR+ hAGP ADR+ hAGP F G 25 ** ** 250 * P=0.0503 Plasma 20 200 urine 15 150 Day7 Day21 Day7 Day21 10 100 Control ADR 5 50 )

200 3 15 0 0 ORM2 mRNA (Fold change) ORM1 mRNA (Fold change) 150 ADR ADR 10 Control Control 100 ADR+ hAGP ADR+ hAGP

Urine AGP 5 150 50 P=0.0504 nd nd (Integrated Intencity (×10

Plasma AGP (% of Control) 0 100 0

ADR ADR ADR ADR Control Control Control Control 50 Day7 Day21 Day7 Day21

0 ORM3 mRNA (Fold change)

ADR Control

ADR+ hAGP

Figure 1. | Exogenous hAGP administration suppresses proteinuria and renal injury in ADR mice. (A) Schedule of human a1-acid glycoprotein (hAGP) treatment. BALB/c mice were administered with hAGP (5 mg/mouse) for 5 consecutive days after adriamycin (ADR) injection. Mice were sacrificed at day 21. (B) Urinary albumin to creatinine ratio at day 7 and day 21 after ADR injection. (C) Total plasma cholesterol at day 21 after ADR injection. (D) Representative periodic acid-Schiff-stained sections at day 21 after ADR injection, and quantitation of evaluation of glomerulosclerosis and tubule damage. Scale bars, 100 mm (control n55; ADR and ADR1hAGP, n56). (E) Representative 346 KIDNEY360

Rosemont, IL) overnight at 4°C, followed by incubation with separated by 10% SDS-PAGE. Proteins were transferred Alexa 488 donkey anti-goat IgG(H1L) (1:500; Abcam plc, onto polyvinylidene difluoride membranes (Immobilon-P; Cambridge, UK) or Alexa 546 goat anti-rabbit IgG(H1L) (1: Millipore, Billerica, MA) by wet electroblotting. The mem- 200; Thermo Fisher Scientific, Waltham, MA) for 90 minutes branes were incubated in ORM1 antibody (1:5000; Protein- at room temperature. Images were obtained on a Keyence tech). Antibody signals were detected with LAS-4000 BZ-X710 microscope (Keyence). minutes (GE Healthcare).

Kidney Cell Suspension Quantitative RT-PCR Kidney cell suspension for isolating renal macrophages Total RNA from kidney and liver isolated by the RNAiso was prepared according to the report of Cao et al. (29). plus (Takara Bio) was reverse-transcribed with PrimeScript Kidneys were harvested after perfusion with HBSS (without RT Master Mix (Takara Bio). Quantitative RT-PCR analy- calcium) and DMEM containing 1 mg/ml collagenase IV sis was performed in an iCycler thermal cycler (Bio-Rad and 100 mg/ml DNase I (Worthington Biochemical Corp, Laboratories, Hercules, CA) using SYBR Premix Ex TaqII Lakewood, NJ). Kidney tissues were cut into a paste and (Takara Bio). The quantitative RT-PCR method has been shaken for 40 minutes at 37°C in DMEM containing colla- previously described (30) and the primers used are listed in genase IV and DNase I. Cell suspensions were passed the Supplemental Table. GAPDH was used as an internal 3 through a 75-mm stainless mesh and centrifuged at 180 control. g for 3 minutes. The pellets were suspended in red blood cell lysis buffer and centrifuged at 300 3 g for 5 minutes. After washing with PBS(2), cell suspensions were incu- Statistical Analyses bated for 15 minutes at 4°C in blocking buffer (2% FBS and The means for two group data were compared by the m unpaired t test. The means for more than two groups were 2 mM EDTA) containing 10 g/ml fragment crystallizable ’ blocker. Cell suspensions were stained with anti-CD45.2, compared by one-way ANOVA followed by Tukey s mul- tiple comparison. Probability values of P,0.05 or P,0.01 F4/80, and antibodies to T cell, B cell, and natural killer cell fi lineages (TCRb, TCRgd, CD3«, CD19, and CD49b). were considered to be signi cant.

Isolation of Renal Macrophages Results Kidney cell suspensions were gated on forward scatter Exogenous hAGP Administration Suppresses Proteinuria and and SSC to remove doublets, and exclude dead cells using Renal Injury in ADR Mice propidium iodide. Then, cells were gated on leukocytes After adriamycin injection (ADR mice), hAGP (5 mg/ using an anti-CD45.2 antibody and T cell, B cell, and natural mouse) was intraperitoneally administrated for 5 consecu- killer cell lineages excluded. Macrophages were isolated tive days (Figure 1A, treatment schedule). The results show using anti-F4/80 antibody. After sorting, RNA isolation that the urinary albumin to creatinine ratio in ADR mice was was performed using an NucleoSpin RNA XS column significantly reduced from 7 to 21 days in the hAGP-treated (Takara Bio, Otsu, Japan). group (Figure 1B). Furthermore, the increased plasma cho- lesterol level observed in the ADR group was significantly Cell Culture suppressed by hAGP administration (Figure 1C). The renal Human THP-1 cells were grown in RPMI 1640 (Gibco, function of each mouse was measured (BUN and serum Grand Island, NY) containing 10% FBS, and 1% penicillin creatinine), but no significant differences were observed fi and streptomycin (Thermo Fisher Scienti c). THP-1 cells between the control, ADR, and ADR-treated with hAGP were maintained at 37°C in a 5% CO2 atmosphere. THP-1 groups (Supplemental Figure 1B). Renal histologic analy- 3 5 cells were seeded at 8 10 cells/well in six-well plates, and ses at day 21 showed that hAGP significantly suppressed differentiated by addition of 50 nM PMA for 48 hours. Cells glomerulosclerosis and tubular atrophy in ADR mice were then washed in PBS and incubated for 24 hours. hAGP (Figure 1D). AGP distribution in the kidney was confirmed – fi (0.5 2.5 mg/ml), which was quanti ed by bicinchoninic by immunofluorescence staining using anti-AGP antibody. acid protein assay, was added and cells incubated for a fur- Fluorescence was largely detected in the glomerulus in ther 48 hours. In the experiments with stimulation by LPS, control mice, but was markedly decreased in the ADR- hAGP (1.0 mg/ml), or IL-4 (10 ng/ml) and IL-13 (10 ng/ml), treated group. In contrast, similar levels of fluorescence in was added to the cells for 24 hours, and then cells stimulated glomeruli were observed in control mice and ADR mice with LPS (100 ng/ml) for 2 hours. administered with hAGP (Figure 1E). Because plasma AGP levels are upregulated during in- Western Blotting flammation, we examined the change in levels of endoge- AGP in plasma and urine was confirmed by western nous mouse AGP (mAGP) in plasma of ADR mice by blotting for each mouse. Both plasma and urine samples western blotting. Although ADR injection increased the (0.5 ml/lane) were boiled for 3 minutes at 100°C and expression of AGP mRNA (Orm1, Orm2, and Orm3) in liver

fluorescence photomicrographs showing AGP localization in kidney at day 21. The glomeruli are enclosed in a white dotted line. 5 Scale bars, 100 mm. Red: AGP (n 4 per group) (F) Expression of mouse a1-acid glycoprotein (mAGP) mRNA in liver was determined by quantitative RT-PCR (control n55; ADR and ADR1hAGP, n56). (G) Level of endogenous mAGP in plasma and urine determined by western blot analysis (control, n53; ADR, n54). nd, not detected. All data are mean 6 SE *P,0.05, **P,0.01; by one-way ANOVA and post hoc Tukey’s multiple comparison. KIDNEY360 1: 343–353, May, 2020 AGP prevents proteinuria via CD163+ macrophages, Fujimura et al. 347

A 20 * * Control ADR ADR+hAGP 15

10 cells per hpf) Macrophages

+ 2 5 (×10 F4/80

0

ADR Control

ADR + hAGP B C D * 5 ** ** 4 40 ** * * 4 3 30 3 2 20 2

1 10 1 IL-1 β mRNA (Fold change) TGF- β mRNA (Fold change) 0 0 CD163 mRNA (Fold change) 0

ADR ADR ADR Control Control Control

ADR + hAGP ADR + hAGP ADR + hAGP E F G 15 200 15 * ** ** 150 10 10

100

5 5 50 IL-10 mRNA (Fold change) iNOS mRNA (Fold change) CD206 mRNA (Fold change) 0 0 0

ADR ADR ADR Control Control Control

ADR + hAGP ADR + hAGP ADR + hAGP

Figure 2. | Exogenous hAGP administration exerts an anti-inflammatory effect in kidneys of ADR mice. (A) Representative micrographs of 1 1 kidneys at day 21 after ADR injection, stained with antibody against F4/80 macrophages, and quantitative evaluation of F4/80 cells. Scale bars, 100 mm(n54 per group) (B–G) Expression levels of TGF-b, IL-1b, CD163, CD206, inducible nitric oxide synthase (iNOS), and IL-10 mRNAs in kidney at day 21 after ADR injection were determined by quantitative RT-PCR (control, n55; ADR and ADR1hAGP, n56). All data are mean 6 SE; *P,0.05; by one-way ANOVA and post hoc Tukey’s multiple comparison.

(Figure 1F), plasma mAGP tended to decrease between 7 Exogenous hAGP Administration Exerts Anti-inflammatory and 21 days. In contrast to plasma levels, urinary mAGP Effects in Kidneys of ADR Mice was increased in ADR mice (Figure 1G). These data suggest To evaluate the inflammatory condition, kidney sections that although mAGP production is increased in renal in- were stained with anti-F4/80 antibody, a representative flammation induced by ADR injection, plasma mAGP levels macrophage marker. The results show that hAGP admin- do not increase due to its higher levels of urinary excretion. istration significantly decreases macrophage infiltration 348 KIDNEY360

A B Control ADR ADR+hAGP Control ADR ADR+hAGP AGP Nephrin

15 5 Nephrin AGP 4 150 200 * 10 3 * * * 150 100 2 5 100 1 50

Damaged Tubles (%) 50 Glomerulosclerosis (%) (% of Control) 0 0 (% of Control) Fluorescence Intensity Fluorescence Intensity 0 0 ADR ADR Control Control ADR ADR Control Control ADR + hAGP ADR + hAGP ADR + hAGP ADR + hAGP

C D E Control ADR ADR+hAGP 4 2.0 * 3 1.5

2 1.0

15 1 0.5

0 IL-1 β mRNA (Fold change) 0 10 TGF- β mRNA (Fold change)

ADR ADR Control Control cells per hpf) Macrophages

+ 2 5 ADR + hAGP ADR + hAGP (×10 F4/80 0

ADR Control

ADR + hAGP

Figure 3. | Exogenous hAGP administration induces an M2-like macrophage phenotype at day 7 in ADR mice. (A) Representative periodic acid-Schiff-stained sections at day 7 after ADR injection, and quantitative of evaluation of glomerulosclerosis and tubule damage (control, n55; ADR and ADR1hAGP, n56). Scale bars, 100 mm. (B) Representative fluorescence photomicrographs showing nephrin and AGP localization in glomeruli at day 7. The glomeruli are enclosed in a white dotted line. Green: nephrin; red: AGP (Control, n53; ADR and ADR1hAGP, n55). (C) 1 Representative micrographs of kidney at day 7 after ADR injection, stained with antibody against F4/80 macrophages and quantitative 1 evaluation of F4/80 cells (n54 per group). Scale bars, 100 mm. (D and E) Levels of expression of TGF-b and IL-1b mRNAs in kidney at day 7 after ADR injection were determined by quantitative RT-PCR (control, n55; ADR and ADR1hAGP, n56). All data are mean 6 SE; *P, 0.05; by one-way ANOVA and post hoc Tukey’s multiple comparison.

(Figure 2A). It has been reported that the expression of TGF-b ADR mice treated with hAGP. Interestingly, hAGP admin- increases in damaged glomerular cells including podo- istration significantly increased CD163 mRNA levels but cytes, whereas IL-1b is produced by macrophages infiltrat- did not increased CD206 mRNA levels in kidney. Both of ing into the glomerulus (31,32). Therefore, levels of mRNA these molecules are representative M2 macrophage markers for TGF-b and IL-1b were measured. As shown in Figure 2, (Figure 2, D and E), whereas the expression of inducible B and C, TGF-b and IL-1b were significantly decreased in nitric oxide synthase (iNOS), which is used as an M1 KIDNEY360 1: 343–353, May, 2020 AGP prevents proteinuria via CD163+ macrophages, Fujimura et al. 349

A 2.0 1.5 *

1.5 5.09 Lin 1.0

FSC-A Macs CD45.2 1.0 47.9 0.5 0.5

FSC-A FSC-A F4/80 0 0 CD206 mRNA (Fold change) CD163 mRNA (Fold change)

Control ADR ADR B Control Control Control + hAGP ADR + hAGP ADR + hAGP AGP

4 2.5 150 * 2.0 3 100 1.5 2 1.0 50 1 (% of Control) 0.5 Fluorescence Intensity CD206 mRNA (Fold change) 0 CD163 mRNA (Fold change) 0 0 Control Control Control Control Control Control +hAGP +hAGP +hAGP

Figure 4. | Exogenous administration of hAGP induces CD163-expressing renal macrophages. (A) The mRNA expression of CD163 and CD206 in renal macrophages at day 7 after ADR injection (control, n54; ADR and ADR1hAGP, n56). (B) Representative fluorescence photomicrographs showing AGP localization in control and control1hAGP mice. The expression of CD163 and CD206 mRNAs in renal macrophages in control and control1hAGP mice (control, n55; control1hAGP, n55). All data are mean 6 SE; *P, 0.05; by one-way ANOVA and post hoc Tukey’s multiple comparison or unpaired t test. macrophage marker, significantly decrease in the hAGP- evaluate the macrophage phenotype. As shown in administered group (Figure 2F). Moreover, in the hAGP- Figure 4A, the mRNA expression of CD163 was decreased administered group, there was significantly increased in macrophages from ADR mice, whereas hAGP adminis- expression of IL-10 compared with control mice (Figure 2G). tration tended to increase the mRNA expression of CD163 (Figure 4A). Next, we administered hAGP to control mice and measured the mRNA expression of CD163 and CD206 Exogenous hAGP Administration Induces in the isolated macrophages. Although the glomerular fluo- a CD163-Expressing Macrophage Phenotype at Day 7 in rescence intensities of AGP were not significantly changed ADR Mice between control and hAGP-treated control mice, hAGP Although hAGP administration significantly decreased administration significantly increased CD163-expressing re- the urinary albumin to creatinine ratio at day 7 (Figure 1B), nal macrophages, even in the control mice (Figure 4B). neither renal histologic damage nor infiltration of macro- phages were found in any of the three groups (control, ADR, and ADR 1 hAGP) (Figure 3, A and C). Compared with hAGP Treatment Switches PMA-Differentiated THP-1 Cells control mice, nephrin expression as an indicator of podocyte into CD163-Expressing Macrophages injury was decreased in ADR mice treated with/or without To further study the effect of hAGP on the plasticity of hAGP at day 7 (Figure 3B). On the other hand, the locali- macrophage phenotypes, PMA-differentiated THP-1 cells, zation of AGP to the glomerulus was increased by hAGP a macrophage-like cell type, were treated with hAGP. Be- administration in ADR mice at day 7 (Figure 3B). Under the cause hAGP circulates at 0.5 mg/ml in the plasma of healthy same experimental conditions, hAGP administration sup- humans and levels increase to 1.0–2.5 mg/ml during in- pressed adriamycin-induced TGF-b mRNA expression flammation, in vitro experiments were performed using (Figure 3E), whereas no significant change in IL-1b mRNA these plasma concentrations of hAGP. Expression of expression was found in any of the three groups (Figure 3E). CD163 mRNA was increased by hAGP in a dose- Because the infiltration of macrophages was not changed dependent manner, consistent with our previous results between the ADR and hAGP-treated ADR groups (23). In addition, we found that hAGP dose-dependently (Figure 3C), we isolated macrophages from kidney to decreased the expression of CD206 and iNOS mRNA 350 KIDNEY360

Plasma AGP concentration (Human) Normal Inflammation

0.5 1.0 2.5 (mg/mL)

ACDB ** 20 ** * 1.5 ** ** 1.5 ** 30 ** ** * ** ** ** ** 15 * 1.0 1.0 20 10 0.5 0.5 10 5

0 0 0 0 IL-10 mRNA (Fold change) (Fold IL-10 mRNA iNOS mRNA (Fold change) (Fold iNOS mRNA CD206 mRNA (Fold change) (Fold CD206 mRNA CD163 mRNA CD163 mRNA (Fold change) 0hAGP 0.5 1.0 2.5 0hAGP 0.5 1.0 2.5 0hAGP 0.5 1.0 2.5 0hAGP 0.5 1.0 2.5

E F ** 2.5 ** 20 ** * ** ** 2.0 ** 15 1.5 10 1.0

5 0.5

0 0 CD206 mRNA (Fold change) (Fold CD206 mRNA CD163 mRNA CD163 mRNA (Fold change) IL-4 IL-4 hAGP IL-4 IL-4 hAGP IL-13 IL-13

Figure 5. | AGP treatment switches PMA-differentiated THP-1 cells into M2-like macrophages. (A–D) The expression of CD163, CD206, iNOS, and IL-10 mRNAs in PMA-differentiated THP-1 cells treated with AGP (0–2.5 mg/ml) for 48 hours (n56 per group). (E and F) Expression of CD163 and CD206 mRNAs in PMA-differentiated THP-1 cells treated with AGP (1.0 mg/ml) or IL-4 (20 ng/ml), and IL-13 (20 ng/ml) or IL-4 (30 ng/ml), for 24 hours (n53 per group). All data are mean 6 SE *P, 0.05, **P,0.01; by one-way ANOVA and post hoc Tukey’s multiple comparison.

(Figure 5, B and C). These in vitro data are consistent with Discussion the changes in renal expression of CD163 and iNOS mRNA In this study, we found that AGP potentiates the change shown in Figures 2 and 4. Interestingly, hAGP also dose- of macrophage phenotype to CD163-expressing macro- dependently increased the expression of IL-10 mRNA phages with anti-inflammatory functions in vivo, thereby (Figure 5D). The changes of CD163 and CD206 expression reducing proteinuria, inflammation, and renal damage. in macrophages treated with hAGP were compared with Although AGP-induced CD163-expressing macrophages those in IL-4/IL-13- or IL-4-treated macrophages, a repre- have M2-like properties, they differ from IL-4/IL-13- sentative M2 macrophage phenotype. IL-4/IL-13 or IL-4 induced M2 macrophages (Figure 5). Both CD163 and treatment did not result in an increase in CD163 expression CD206 are normally used as markers of M2 macrophages in macrophages, whereas CD206 expression was signifi- (33,34), but the changes in expression of each marker are not cantly increased (Figure 5, E and F). always consistent during macrophage activation switching. For example, Porcheray et al. showed that IL-4 induced CD206 but not CD163 in human -derived macro- The AGP-Induced CD163-Expressing Macrophage phages. In contrast, IL-10 induced CD163 but not CD206 Phenotype has Anti-Inflammatory Properties under the same conditions (33). These results are similar to Next,wecomparedthephenotypicandanti- the characteristics of hAGP-treated macrophages, as shown inflammatory properties of macrophages treated with in Figure 5. Moreover, Furuhashi et al. (26) previously AGP or IL-4/IL-13 under inflammatory conditions. As reported that adipose-derived stromal cells induce both 1 1 shown in Figure 6, hAGP- or IL-4/IL-13-induced macro- CD163 macrophages and CD206 macrophages in cres- 1 phages maintained their phenotypes even after LPS chal- centic GN rats, and that CD163 macrophages had higher 1 lenge for 2 hours. The expression of IL-10 mRNA was levels of IL-10 expression compared with CD206 macro- measured after stimulation with LPS for 2 hours after treat- phages. This observation is in agreement with the our 1 ment with either AGP or IL-4/IL-13. Interestingly, AGP- results, which show that hAGP-induced CD163 macro- treated macrophages had a greater ability to induce IL-10 phages have increased IL-10 production ability compared 1 expression compared with IL-4/IL-13-treated macrophages with IL-4/IL-13-induced CD206 macrophages (Figure 6). (Figure 6D). Previously, we reported that hAGP increases CD163 and KIDNEY360 1: 343–353, May, 2020 AGP prevents proteinuria via CD163+ macrophages, Fujimura et al. 351

hAGP or IL-4 +IL-13 LPS

02426 (hr)

A B C D

** ** ** ** ** 40 ** ** 5 ** 2.0 ** ** 1.5 ** ** ** ** ** 4 1.5 30 1.0 3 1.0 20 2 0.5 10 1 0.5 0 0 0 0 IL-10 mRNA (Fold change) (Fold IL-10 mRNA IL-4 hAGP IL-4 hAGP change) (Fold iNOS mRNA IL-4 hAGP IL-4 hAGP CD163 mRNA (Fold change) (Fold CD163 mRNA IL-13 change) (Fold CD206 mRNA IL-13 IL-13 IL-13 LPS LPS LPS LPS

Figure 6. | The hAGP-induced macrophage phenotype demonstrates anti-inflammatory properties. (A–D) The expression of CD163, CD206, iNOS, and IL-10 mRNAs in PMA-differentiated THP-1 cells stimulated with LPS (100 ng/ml) for 2 hours after treatment with AGP (1.0 mg/ml), or IL-4 (10 ng/ml) and IL-13 (10 ng/ml), for 24 hours (n54 per group). All data are mean 6 SE *P, 0.05, **P,0.01; by one-way ANOVA and post hoc Tukey’s multiple comparison.

IL-10 via weak stimulation of the TLR4/CD14 pathway (23). because AGP is a highly negative charged protein resulting Subsequently, Nakamura et al. reported that ORM1, one of from the presence of sialylated N-glycans (pI of 2.8–3.8) (13). the AGP variants, induces M2b monocytes in vitro (35) and Hjalmarsson et al. (42) reported that the administration produces IL-10, which is known to be induced by TLR4 of hAGP increased negative charges at the ESL and im- signals. Therefore, the AGP-induced CD163-expressing proved podocyte morphology in rats with puromycin macrophages found in this study may correspond to M2b aminonucleoside-induced nephrosis. Our study found that macrophages, which are also known as regulatory macro- localization of AGP in the glomerulus was increased by phages. It is recognized that regulatory macrophages pro- hAGP treatment at day 7 (Figure 3B). Because it has been duce high levels of IL-10, which limits inflammation by reported that damaged glomerular cells such as podocytes suppressing the production and activation of various proin- produced TGF-b, the increased glomerular localization of flammatory cytokines. Although IL-4/IL-13-induced mac- AGP might contribute to the inhibition of TGF-b mRNA rophages are also effective against inflammatory disease, expression in ADR mice (Figure 3D). In addition, these data they have the potential to increase the risk of tissue fibrosis suggest that administered hAGP might be localized for at (36,37). On the other hand, regulatory macrophages have least 3 days in the glomerulus and hence might contribute to low fibrotic potential compared with IL-4/IL-13-induced the suppression of proteinuria at day 7. On the other hand, macrophages (38). Therefore, it is unlikely that hAGP- increased glomerular localization of hAGP was not found in induced CD163-expressing macrophages have the potential hAGP-treated control mice (Figure 4B). This result indicates to induce renal fibrosis. This is supported by our recent that the accumulation of AGP may require a decreasing study in which exogenously administered hAGP sup- negative charge in the glomerulus. Further investigation pressed unilateral ureter obstruction-induced renal fibrosis would be needed to elucidate the role of exogenously ad- in a mouse model (27). ministered hAGP on glomerular negative charge selectivity Administration of hAGP inhibited urinary albumin in in the future. In addition, the anti-AGP (ORM1) antibody ADR mice between days 7 and 21. However, renal histologic used in this study has cross reactivity with hAGP and damage and infiltration of macrophages in ADR mice were mAGP. Therefore, we could not clearly distinguish whether markedly weaker at day 7 than at day 21. Reflecting these the localization of AGP in the glomerulus came from en- data, the expression of IL-1b mRNA was not changed in dogenous or exogenous AGP under our experimental con- ADR mice because IL-1b is produced by infiltrating glo- ditions. Further study may be needed using hAGP- or merular macrophages. These results imply that other mech- mAGP-specific antibodies if such antibodies are commer- anisms may be involved in the decrease in urinary albumin cially available. in hAGP-treated mice at day 7. The glomerular barrier has In conclusion, this study indicates that AGP has a unique both size and negative charge selectivity. Negative charge mechanism in regulating inflammation and has the poten- selectivity is a result of the endothelial cell surface layer tial to suppress urinary protein levels by targeting the ESL. (ESL), which functions to suppress albumin leakage into The results showed that endogenous AGP could work to urine (39241). Interestingly, AGP is localized at the ESL protect against glomerular disease. Because proteinuria is and contributes to the negative charge barrier in glomeruli an independent risk factor for CKD progression and CVD 352 KIDNEY360

onset (2,3), it is speculated that alleviation of proteinuria and individuals from 7 cohort studies in Japan. Am J Epidemiol 178: 1–11, 2013 protection against renal tissue damage by AGP could be ~ a new therapeutic strategy for CKD such as nephrosis, and 5. Ordonez JD, Hiatt RA, Killebrew EJ, Fireman BH: The increased risk of coronary heart disease associated with nephrotic syn- diabetic nephropathy and CVD. Future investigation is drome. Kidney Int 44: 638–642, 1993 needed to evaluate if AGP is also protective if given in 6. Abbate M, Zoja C, Remuzzi G: How does proteinuria cause a later phase of the disease. progressive renal damage? J Am Soc Nephrol 17: 2974–2984, 2006 7. Silverstein DM: Inflammation in chronic kidney disease: Role in Author Contributions the progression of renal and cardiovascular disease. Pediatr Nephrol 24: 1445–1452, 2009 R. Fujimura, H. Watanabe, and T. Maruyama conceptualized the 8. Wang Y, Harris DC: Macrophages in renal disease. J Am Soc study; J. Bi, R. Fujimura, Y. Fujiwara, T. Imafuku, K. Kobayashi, Nephrol 22: 21–27, 2011 T. Koga, H. Komori, M. Miyahisa, K. Nishida, and H. Maeda were 9. Cao Q, Wang Y, Harris DC: Pathogenic and protective role of responsible for investigation; R. Fujimura was responsible for for- macrophages in kidney disease. Am J Physiol Renal Physiol 305: mal analysis, visualization, and the original draft of the manuscript; F3–F11, 2013 10. Liu YC, Zou XB, Chai YF, Yao YM: Macrophage polarization in H. Watanabe and T. Maruyama were responsible for funding inflammatory diseases. Int J Biol Sci 10: 520–529, 2014 acquisition, project administration, and supervision; R. Fujimura, 11. Lee S, Huen S, Nishio H, Nishio S, Lee HK, Choi BS, Ruhrberg C, Y. Fujiwara, T. Koga, H. Watanabe, and T. Maruyama were Cantley LG: Distinct macrophage phenotypes contribute to responsible for methodology; R. Fujimura, H. Watanabe, and kidney injury and repair. J Am Soc Nephrol 22: 317–326, 2011 T. Maruyama were responsible for data curation; R. Fujimura, 12. Wang Y, Wang YP, Zheng G, Lee VW, Ouyang L, Chang DH, Mahajan D, Coombs J, Wang YM, Alexander SI, Harris DC: Ex Y. Fujiwara, T. Koga, H. Watanabe, and T. Maruyama were re- vivo programmed macrophages ameliorate experimental chronic sponsible for resources; and R. Fujimura, H. Watanabe, Y. Fujiwara, inflammatory renal disease. Kidney Int 72: 290–299, 2007 T. Koga, H. Maeda, M. Tanaka, K. Matsushita, T. Wada, M. 13. Fournier T, Medjoubi-N N, Porquet D: Alpha-1-acid glycopro- Fukagawa, and T. Maruyama reviewed and edited the manuscript. tein. Biochim Biophys Acta 1482: 157–171, 2000 14. Hochepied T, Berger FG, Baumann H, Libert C: Alpha(1)-acid glycoprotein: An acute phase protein with inflammatory and Disclosures immunomodulating properties. Cytokine Growth Factor Rev 14: M. Fukagawa reports grants and personal fees from Kyoto Kirin 25–34, 2003 and Ono Pharmaceutical, and personal fees from Bayer Japan. 15. Edwards DJ, Lalka D, Cerra F, Slaughter RL: Alpha1-acid gly- M. Tanaka reports personal fees from Torii and Kyowa Kirin. J. Bi, coprotein concentration and protein binding in trauma. Clin R. Fujimura, Y. Fujiwara, T. Imafuku, K. Kobayashi, T. Koga, Pharmacol Ther 31: 62–67, 1982 H. Komori, H. Maeda, T. Maruyama, K. Matsushita, M. Miyahisa, 16. Arai H, Tomizawa S, Maruyama K, Seki Y, Kuroume T: Reversed- phase high-performance liquid chromatography for analysis of K. Nishida, T. Wada, and H. Watanabe have nothing to disclose. urinary proteins: Diagnostic significance of alpha 1-acid glyco- protein. Nephron 66: 278–284, 1994 Funding 17. Bennett MR, Pleasant L, Haffner C, Ma Q, Haffey WD, Ying J, This work was supported in part by Grant-in-Aid for Challenging Wagner M, Greis KD, Devarajan P: A novel biomarker panel to Exploratory Research grant 16K15320, Japan Society for the Pro- identify steroid resistance in childhood idiopathic nephrotic syndrome. Biomark Insights 12: 1177271917695832, 2017 motion of Science Fellows grant 18J23070, and the KUMAYAKU 18. Costello MJ, Gewurz H, Siegel JN: Inhibition of neutrophil ac- Alumni Research Foundation. tivation by alpha1-acid glycoprotein. Clin Exp Immunol 55: 465–472, 1984 19. Laine´ E, Couderc R, Roch-Arveiller M, Vasson MP, Giroud JP, References Raichvarg D: Modulation of human polymorphonuclear neu- 1. Hill NR, Fatoba ST, Oke JL, Hirst JA, O’Callaghan CA, Lasserson trophil functions by alpha 1-acid glycoprotein. Inflammation 14: DS, Hobbs FDR: Global prevalence of chronic kidney disease - A 1–9, 1990 systematic review and meta-analysis. PLoS One 11: e0158765, 20. Chiu KM, Mortensen RF, Osmand AP, Gewurz H: Interactions of 2016 alpha1-acid glycoprotein with the immune system. I. Purification 2. Matsushita K, van der Velde M, Astor BC, Woodward M, Levey and effects upon lymphocyte responsiveness. Immunology 32: AS, de Jong PE, Coresh J, Gansevoort RT; Chronic Kidney Disease 997–1005, 1977 Prognosis Consortium: Association of estimated glomerular fil- 21. Bories PN, Guenounou M, Fe´ger J, Kodari E, Agneray J, Durand G: tration rate and albuminuria with all-cause and cardiovascular Human alpha 1-acid glycoprotein-exposed macrophages release mortality in general population cohorts: A collaborative meta- interleukin 1 inhibitory activity. Biochem Biophys Res Commun analysis. Lancet 375: 2073–2081, 2010 147: 710–715, 1987 3. van der Velde M, Matsushita K, Coresh J, Astor BC, Woodward M, 22. Daemen MA, Heemskerk VH, van’t Veer C, Denecker G, Wolfs Levey A, de Jong P, Gansevoort RT, van der Velde M, Matsushita TG, Vandenabeele P, Buurman WA: Functional protection by K, Coresh J, Astor BC, Woodward M, Levey AS, de Jong PE, acute phase proteins alpha(1)-acid glycoprotein and alpha(1)- Gansevoort RT, Levey A, El-Nahas M, Eckardt KU, Kasiske BL, antitrypsin against ischemia/reperfusion injury by preventing Ninomiya T, Chalmers J, Macmahon S, Tonelli M, Hemmelgarn B, apoptosis and inflammation. Circulation 102: 1420–1426, 2000 Sacks F, Curhan G, Collins AJ, Li S, Chen SC, Hawaii Cohort KP, 23. Komori H, Watanabe H, Shuto T, Kodama A, Maeda H, Watanabe Lee BJ, Ishani A, Neaton J, Svendsen K, Mann JF, Yusuf S, Teo KK, K, Kai H, Otagiri M, Maruyama T: a(1)-Acid glycoprotein up- Gao P, Nelson RG, Knowler WC, Bilo HJ, Joosten H, Kleefstra N, regulates CD163 via TLR4/CD14 protein pathway: Possible Groenier KH, Auguste P, Veldhuis K, Wang Y, Camarata L, protection against hemolysis-induced oxidative stress. J Biol Thomas B, Manley T; Chronic Kidney Disease Prognosis Con- Chem 287: 30688–30700, 2012 sortium: Lower estimated glomerular filtration rate and higher 24. Mosser DM: The many faces of macrophage activation. J Leukoc albuminuria are associated with all-cause and cardiovascular Biol 73: 209–212, 2003 mortality. A collaborative meta-analysis of high-risk population 25. Alvarado-Vazquez PA, Bernal L, Paige CA, Grosick RL, Moracho cohorts. Kidney Int 79: 1341–1352, 2011 Vilrriales C, Ferreira DW, Ulecia-Moro´n C, Romero-Sandoval EA: 4. Nagata M, Ninomiya T, Kiyohara Y, Murakami Y, Irie F, Sairenchi Macrophage-specific nanotechnology-driven CD163 over- T, Miura K, Okamura T, Ueshima H; EPOCH-JAPAN Research expression in human macrophages results in an M2 phenotype Group: Prediction of cardiovascular disease mortality by pro- under inflammatory conditions. Immunobiology 222: 900–912, teinuria and reduced kidney function: Pooled analysis of 39,000 2017 KIDNEY360 1: 343–353, May, 2020 AGP prevents proteinuria via CD163+ macrophages, Fujimura et al. 353

26. Furuhashi K, Tsuboi N, Shimizu A, Katsuno T, Kim H, Saka Y, 34. Biswas SK, Mantovani A: Macrophage plasticity and interaction Ozaki T, Sado Y, Imai E, Matsuo S, Maruyama S: Serum-starved with lymphocyte subsets: Cancer as a paradigm. Nat Immunol adipose-derived stromal cells ameliorate crescentic GN by 11: 889–896, 2010 promoting immunoregulatory macrophages. J Am Soc Nephrol 35. Nakamura K, Ito I, Kobayashi M, Herndon DN, Suzuki F: Oro- 24: 587–603, 2013 somucoid 1 drives opportunistic infections through the polari- 27. Bi J, Watanabe H, Fujimura R, Nishida K, Nakamura R, Oshiro S, zation of monocytes to the M2b phenotype. Cytokine 73: 8–15, Imafuku T, Komori H, Miyahisa M, Tanaka M, Matsushita K, 2015 Maruyama T: A downstream molecule of 1,25-dihydroxyvitamin 36. Borthwick LA, Barron L, Hart KM, Vannella KM, Thompson RW, D3, alpha-1-acid glycoprotein, protects against mouse model of Oland S, Cheever A, Sciurba J, Ramalingam TR, Fisher AJ, Wynn renal fibrosis. Sci Rep 8: 17329, 2018 TA: Macrophages are critical to the maintenance of IL-13- 28. Cao Q, Wang C, Zheng D, Wang Y, Lee VW, Wang YM, Zheng G, dependent lung inflammation and fibrosis. Mucosal Immunol Tan TK, Yu D, Alexander SI, Harris DC, Wang Y: IL-25 9: 38–55, 2016 induces M2 macrophages and reduces renal injury in 37. Jakubzick C, Kunkel SL, Puri RK, Hogaboam CM: Therapeutic proteinuric kidney disease. J Am Soc Nephrol 22: 1229–1239, targeting of IL-4- and IL-13-responsive cells in pulmonary fibrosis. 2011 Immunol Res 30: 339–349, 2004 29. Cao Q, Wang Y, Wang XM, Lu J, Lee VW, Ye Q, Nguyen H, Zheng 38. Fleming BD, Mosser DM: Regulatory macrophages: Setting the G, Zhao Y, Alexander SI, Harris DC: Renal F4/801 threshold for therapy. Eur J Immunol 41: 2498–2502, 2011 CD11c1 mononuclear phagocytes display 39. Haraldsson B, Nystro¨m J, Deen WM: Properties of the glomerular phenotypic and functional characteristics of macrophages in barrier and mechanisms of proteinuria. Physiol Rev 88: 451–487, health and in adriamycin nephropathy. J Am Soc Nephrol 26: 2008 349–363, 2015 40. Hjalmarsson C, Johansson BR, Haraldsson B: Electron micro- 30. Enoki Y, Watanabe H, Arake R, Fujimura R, Ishiodori K, Imafuku scopic evaluation of the endothelial surface layer of glomerular T, Nishida K, Sugimoto R, Nagao S, Miyamura S, Ishima Y, capillaries. Microvasc Res 67: 9–17, 2004 Tanaka M, Matsushita K, Komaba H, Fukagawa M, Otagiri M, 41. Jeansson M, Bjo¨rck K, Tenstad O, Haraldsson B: Adriamycin alters Maruyama T: Potential therapeutic interventions for chronic glomerular endothelium to induce proteinuria. J Am Soc Nephrol kidney disease-associated sarcopenia via indoxyl sulfate- 20: 114–122, 2009 induced mitochondrial dysfunction. J Cachexia Sarcopenia 42. Hjalmarsson C, Lidell ME, Haraldsson B: Beneficial effects of Muscle 8: 735–747, 2017 orosomucoid on the glomerular barrier in puromycin 31. Kim JH, Kim BK, Moon KC, Hong HK, Lee HS: Activation of the aminonucleoside-induced nephrosis. Nephrol Dial Transplant TGF-beta/Smad signaling pathway in focal segmental glomer- 21: 1223–1230, 2006 ulosclerosis. Kidney Int 64: 1715–1721, 2003 Received: November 26, 2019 Accepted: March 12, 2020 32. Matsumoto K: Production of interleukin-1 by glomerular mac- rophages in nephrotoxic serum nephritis. Am J Nephrol 10: 502–506, 1990 R.F. and H.W. contributed equally to this work. 33. Porcheray F, Viaud S, Rimaniol AC, Le´one C, Samah B, Dereuddre-Bosquet N, Dormont D, Gras G: Macrophage K360: This article contains supplemental material online at http:// activation switching: An asset for the resolution of inflammation. kidney360.asnjournals.org/lookup/suppl/doi:10.34067/ Clin Exp Immunol 142: 481–489, 2005 KID.0000782019/-/DCSupplemental.

a1-acid glycoprotein attenuates adriamycin-induced nephropathy via CD163

expressing macrophages induction

Rui Fujimura, Hiroshi Watanabe, Kento Nishida, Yukio Fujiwara, Tomoaki Koga,

Jing Bi, Tadashi Imafuku, Kazuki Kobayashi, Hisakazu Komori,

Masako Miyahisa, Hitoshi Maeda, Motoko Tanaka, Kazutaka Matsushita,

Takashi Wada, Masafumi Fukagawa, and Toru Maruyama

Supplemental Information

Supplemental Figure 1

A

26 ConControl trol ADR 24 ADR ADR+AGPADR+hAGP 22 20 18

Body weight (g) 16 14 0 3 6 9 12 15 18 21 Days after injected ADR

B Serum Cr ** 250250 3030 0.3) 0.3 ** dL 200200 ) (mg/ 0.2 dL 2020 0.2 150150

100100 creatinine creatinine 0.1 BUN (mg/ 1010 0.1 Kidney weight (mg) 5050 Serum Serum

00 0.00 00

Ctrl ADR ADR Ctrl ADRADR ADR Control Control ADR Control ADR+hAGP Control ADR+hAGP ADR+hAGP ADR+hAGP ADR+hAGP ADR+hAGP

Supplemental Fig.1. Biochemical parameters in ADR mice at day 21.

(A) Body weight after administration of ADR. (B) Level of blood urea nitrogen (BUN), serum creatinine and kidney weight at day 21 after administration of ADR. All data are mean ± SE. **P<0.01; by one-way

ANOVA and post hoc Tukey’s multiple comparison. (Control n=5, ADR, ADR+hAGP n=6)

Supplemental Table

Primers used in qRT-PCR.

Target Forward (5’→3’) Reverse (5’→3’) Mouse GAPDH AACTTTGGCATTGTGGAAGG ACACATTGGGGGTAGGAACA Orm1 ACACAATAGAGCTTCGGGAGTC ATATCTGGCCTTTTGGCATAGA Orm2 GGCTGTCCTAAACCCTGATT GGTCCCATTCTCTCTCTGGA Orm3 TTCTCATTGCTGTGGCTGAC GGTCCCATTCTCTCTCTGGA TGF-b TGGAAGTGGATCCA GCAGGAGCGCACACG CGCGCCCAAGG ATCATGTTGGAC IL-1b TGAGCTGAAAGCTCTCCACC CTGATGTACCAGTTGGGGAA CD163 ATGGGCTAACTCCAGCGCCG GATCCATCTGAGCAGGTCACTCCA CD206 GCCAGAGACATAACAGCA CAGGTTTCCTTTCAGTCCT iNOS AGTCAACTGCAAGAGAACGGA GAAGAGAAACTTCCAGGGGCA Human GAPDH GGTGAAGGTCGGAGTCAACG ACCATGTAGTTGAGGTC AATGAAGG CD163 ACAGCGGCTTGCAGTTTCCT ATGGCCTCCTTTTCCATTCCAG CD206 CGAGGAAGAGGTTCGGTTCACC GCAATCCCGGTTCTCATGGC iNOS CGTGGAGACGGGAAAGAAGT CCTGGGTCCTCTGGTCAAAC IL-10 AGGGAAGAAATCGATGACAGC TCAAGGCGCATGTGAACTC