of October 16, 2018.This information is current as

of Lupus Nephritis in NZB/W MiceCCR1 Inhibition Ameliorates the Progression

Michel Peuchmaur, Dominique Berrebi and Karl BalabanianTharinger, Katia Mayol, Thierry Walzer, Pius Loetscher, Alexandre Bignon, Françoise Gaudin, Patrice Hémon, Hugo

http://www.jimmunol.org/content/192/3/886doi: 10.4049/jimmunol.1300123December 2013;

2014; 192:886-896; Prepublished online 23J Immunol 

MaterialSupplementary

3.DC1http://www.jimmunol.org/content/suppl/2013/12/24/jimmunol.130012

Referenceshttp://www.jimmunol.org/content/192/3/886.full#ref-list-1

, 17 of which you can access for free at: cites 47 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2014 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

CCR1 Inhibition Ameliorates the Progression of LupusNephritis in NZB/W Mice

Alexandre Bignon,*,† Francoise Gaudin,*,† Patrice Hemon,*,† Hugo Tharinger,‡

Katia Mayol,x Thierry Walzer,x Pius Loetscher,{ Michel Peuchmaur,‖,#

Dominique Berrebi,*,†,‖,# and Karl Balabanian*,†

Systemic lupus erythematosus is a chronic inflammatory autoimmune disease, the development of which is characterized by a pro-

gressive loss of renal function. Such dysfunction is associated with leukocyte infiltration in the glomerular and tubulointerstitial

compartments in both human and experimental lupus nephritis. In this study, we investigated the role of the Ccr1 chemokine re-

ceptor in this infiltration process during the progression of nephritis in the lupus-prone New Zealand Black/New Zealand White

(NZB/W) mouse model. We found that peripheral T cells, mononuclear phagocytes, and neutrophils, but not B cells, from nephritic

NZB/Wmice were more responsive to Ccr1 ligands than the leukocytes from younger prenephritic NZB/Wmice. Short-term treat-

ment of nephritic NZB/W mice with the orally available Ccr1 antagonist BL5923 decreased renal infiltration by T cells and macro-

phages. Longer Ccr1 blockade decreased kidney accumulation of effector/memory CD4+ T cells, Ly6C+ monocytes, and both M1

and M2 macrophages; reduced tubulointerstitial and glomerular injuries; delayed fatal proteinuria; and prolonged animal life-

span. In contrast, renal humoral immunity was unaffected in BL5923-treated mice, which reflected the unchanged numbers of

infiltrated B cells in the kidneys. Altogether, these findings define a pivotal role for Ccr1 in the recruitment of T and mononuclear

phagocyte cells to inflamed kidneys of NZB/W mice, which in turn contribute to the progression of renal injury. The Journal of

Immunology, 2014, 192: 886–896.

Systemic lupus erythematosus (SLE) is a chronic, inflamma-tory autoimmune disease characterized by the productionof autoantibodies against nuclear Ags; immune complex

deposition in multiple organs, including the kidney; complement

activation; leukocyte infiltration; and tissue damage (1–3). Leu-kocyte infiltration features in both human and experimental lupusnephritis and is strikingly associated with a progressive loss ofrenal function (4, 5). This infiltration includes T cells, B cells,neutrophils, and mononuclear phagocytes, and involves the glo-merular as well as the tubulointerstitial compartments of the kidney(6, 7). Infiltrating leukocytes constitute a source of inflammatoryand profibrotic mediators (e.g., cytokines, chemokines, extracel-lular matrix proteins), which contribute to the proliferation anddifferentiation of resident kidney cells and to matrix depositionleading to renal injury (7–10). Thus, preventing or reducing leu-kocyte recruitment into inflamed kidneys should inhibit the pro-gression of lupus nephritis.Chemokines and their seven-transmembrane G-protein–coupled

receptors are critical mediators in the inflammatory process andthus represent attractive therapeutic targets (11, 12). Many studiesrevealed that locally produced inflammatory chemokines (e.g.,Ccl2 to Ccl5, Cxcl10 to Cxcl13) govern the complex multistepprocess leading to renal leukocyte infiltration and injury in dif-ferent mouse models of nephropathy (13–19). In New ZealandBlack/New Zealand White (NZB/W) mice, which spontaneouslydevelop a severe autoimmune disease and more closely mimic thehuman SLE condition compared with other rodent models (20),we previously reported that splenic T, B, and myeloid cells fromnephritic mice migrated into noninflamed syngeneic kidneys (19).This process was enhanced if the recipient kidneys were chroni-cally inflamed, indicating that circulating leukocytes and thekidney both contribute to the inflammation in lupus nephritisthrough a self-maintenance mechanism. Elevated renal expressionof two Ccr1 chemokine receptor ligands: Ccl3 and Ccl5, in as-sociation with mononuclear phagocytes and T cell infiltration havebeen reported in NZB/W mice, as well as in other models of SLEand in human lupus nephritis (16, 19, 21). Thus, Ccr1 may me-diate the coordinated recruitment of inflammatory leukocytes to

*Universite Paris-Sud, Laboratoire “Cytokines, Chimiokines et Immunopathologie,”Unite Mixte de Recherche S996, 92140 Clamart, France; †INSERM, Laboratoired’Excellence en Recherche sur le Medicament et l’Innovation Therapeutique, 92140Clamart, France; ‡Institut Paris-Sud d’Innovation Therapeutique/Institut Federatif deRecherche 141, 92290 Chatenay-Malabry, France; xCentre International de Rechercheen Infectiologie, INSERM U1111, Ecole Normale Superieure, Universite Lyon 1,Centre National de la Recherche Scientifique, Unite Mixte de Recherche 5308,69365 Lyon Cedex 07, France; {Novartis Institutes for BioMedical Research, NovartisPharma AG, CH-4056 Basel, Switzerland; ‖Service d’Anatomie et de Cytologie Path-ologique, Unite Propre de Recherche de l’Enseignement Superieur Associe 1320,Hopital Robert Debre, 75019 Paris, France; and #Universite Denis Diderot, UniversiteParis 7, 75013 Paris France

Received for publication January 14, 2013. Accepted for publication November 19,2013.

This work was supported by the Assistance Publique-Hopitaux de Paris (Grant 07018)and the Agence Nationale de la Recherche (Grant 2010 JCJC 1104 01). A.B., F.G.,and K.B. are members of the Laboratory of Excellence in Research on Medica-tion and Innovative Therapeutics, supported by a grant from the Agence Nationalede la Recherche (ANR-10-LABX-33) under the program “Investissements d’Avenir”ANR-11-IDEX-0003-01. A.B. was the recipient of a fellowship from the FrenchMinistry and from the Fondation pour la Recherche Medicale (FDT20130928127).

A.B. and K.B. conceived and designed the experiments. A.B., F.G., P.H., H.T., M.P,D.B., and K.B. performed the experiments. A.B., M.P., D.B., and K.B. analyzed thedata. K.M., T.W., and P.L. contributed to reagents/materials/analysis tools. A.B., D.B.,and K.B. wrote the paper. K.B. provided funding. A.B., F.G., P.H., H.T., K.M., T.W.,P.L., M.P., D.B., and K.B. read and approved the final manuscript.

Address correspondence and reprint requests to Dr. Karl Balabanian. INSERM, UniteMixte de Recherche S996, Universite Paris-Sud, Laboratoire d’Excellence enRecherche sur le Medicament et l’Innovation Therapeutique, 32 Rue des Carnets,92140 Clamart, France. E-mail address: karl.balabanian@u-psud.fr

The online version of this article contains supplemental material.

Abbreviations used in this article: CMTMR, 5-(6)-(((4-chloromethyl) benzoyl) amino)tetramethylrhodamine; IHC, immunohistochemistry; NZB/W, New Zealand Black/NewZealand White; SLE, systemic lupus erythematosus; Treg, regulatory T cell.

Copyright� 2014 by TheAmericanAssociation of Immunologists, Inc. 0022-1767/14/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1300123

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

sites of renal injury, resulting in tubular and/or glomerular lesionsand tissue fibrosis. Indeed, accumulation of CCR1-positive mac-rophages and lymphocytes is found mainly in interstitial lesions ofrenal biopsies from SLE patients (22), and Ccr1 on myeloid cellsand some T cell subsets is thought to guide them to inflamed targetorgans such as the kidney in various rodent models of nephropathy(13, 14, 16–18, 23–28).In contrast to the prominent role of Ccr1 in renal leukocyte

infiltration in the lupus-like glomerulonephritis MRL-Fas(lpr)model (16, 17), nothing is known about the specific role of Ccr1in the development and progression of nephritis in NZB/W mice.Our aims were to 1) compare the expression of Ccr1 and itsligands in the kidney and splenic leukocytes of prenephritic andnephritic NZB/W mice; 2) investigate the functional relevance ofCcr1 in the renal recruitment of leukocytes in nephritic mice; and 3)assess the contribution of Ccr1 and its ligands to kidney damageusing the orally available Ccr1 antagonist BL5923.

Materials and MethodsMice and evaluation of proteinuria

NZB and NZW were purchased from Harlan Laboratories (Venray, TheNetherlands). (NZB 3 NZW) F1 (NZB/W) mice were bred from NZBfemales and NZW males in our conventional animal facility. Only NZB/Wfemale mice were used in this work. Urine samples were tested for pro-teinuria using Multistix 10 SG (Bayer Diagnostics, Puteaux, France) ona 0–4+ scale, corresponding to the following approximate protein con-centrations: 0, negative or trace; 1+, 30 mg/dl; 2+, 100 mg/dl; 3+, 300 mg/dl;and 4+, 2000 mg/dl. Mice were considered to have severe nephritis iftwo consecutive urine samples scored 3+. The 25- to 30-wk-old NZB/Wmice with severe proteinuria and renal inflammatory infiltrates were con-sidered nephritic, whereas 10- to 17-wk-old mice without proteinuria andkidney inflammation were considered young and prenephritic (5, 19).Ccr1- and Ccl3-deficient mice were kindly provided by Drs. C. Comba-diere and F. Sennlaub, respectively (29, 30). The study was conducted incompliance with the European Union guide for the care and use of labo-ratory animals and has been reviewed and approved by an appropriateinstitutional review committee (C2EA-26, Animal Care and Use Com-mittee, Villejuif, France).

Long-term BL5923 treatment of mice

BL5923 (Novartis, Basel, Switzerland) is a small-molecule antagonist witha high selectivity for CCR1 and an excellent pharmacokinetic profile (18,31). It has equal potency against human, mouse, and rat CCR1 (IC50 = 20,22 and 28 nM, respectively) in the binding assay (31). In addition, BL5923inhibited CCL3-induced calcium fluxes (IC50 = 16 nM) and chemotaxis(IC50 = 3 nM). In our hands, BL5923 has no effect on in vitro chemotaxisof spleen leukocytes from NZB/W mice in response to CCL2, CXCL-10,-12, and -13 (data not shown). The application of 60 mg/kg BL5923 byoral gavage was shown to result in efficacy in mouse and rat models (18,31). We confirmed this in NZB/W mice through a pilot assay of in vivohoming of labeled leukocytes by administering scale doses of BL5923ranging from 10 to 100 mg/kg. The maximal inhibition of leukocyte re-cruitment in the kidneys was obtained from the 60 mg/kg dose (data notshown). For curative treatment, groups of 10 NZB/W mice with severeproteinuria (3+) were randomized and treated blindly with either BL5923(60 mg/kg) or vehicle (hydroxyethylcellulose 0.5%) by oral gavage threetimes a week for 6 wk. Survival and proteinuria were evaluated during thetime course of the treatment. A shorter treatment was used to assess theeffects of BL5923 on renal damage. Groups of eight NZB/W mice withsevere proteinuria were treated by oral gavage twice per day (6 h betweendoses) with either BL5923 (60 mg/kg) or vehicle for 10 d. At the end of thetreatment, mice were sacrificed and analyzed for serum concentrations ofIgs and cytokines by ELISA, renal leukocyte infiltration by flow cytometry,and renal pathological features by immunohistochemistry (IHC).

Leukocyte preparation

To prepare cell suspensions enriched in renal leukocytes, one kidney fromeach mouse was digested for 1 h at 37˚C in humidified air with 5% CO2,using 0.1 mg/ml Liberase Blendzyme 3 (Roche Diagnostics, Mannheim,Germany), 0.4 mg/ml Collagenase D (Roche Diagnostics), and 400 U/mlDNase I (Sigma-Aldrich, St. Louis, MO) in DMEM supplemented with 1%BSA and 2.4 mM CaCl2. After standardized mechanical dissociation with

a gentleMACS Dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany),cells were filtered through a 70-mm-pore nylon mesh (BD Biosciences,San Jose, CA), and erythrocytes were lysed with an ammonium-chloride-potassium buffer. Cells were harvested from mouse spleens as described(32). Sorting of splenic CD3+, CD19+, F4-80+, and Gr-1+ cell subsets wascarried out on a FACSAria (BD Biosciences). Sorting purity was . 98%for each preparation. Leukocyte-enriched preparations were washed inPBS supplemented with 2% FCS and used for quantitative PCR, immu-nophenotyping, or functional assays.

Flow cytometric analyses and functional studies

Kidney and splenic leukocytes were preincubated with 5 mg/ml rat anti-mouse CD16/CD32 (Fc Blocking, eBioscience, San Diego, CA), washed inPBS, and incubated with the following Abs: Alexa Fluor 700 (AF-700)–conjugated rat anti-mouse CD45 (clone 30-F11) or rat anti-mouse Ly6G(clone 1A8), allophycocyanin-conjugated rat anti-mouse CD11c (cloneHL3), allophycocyanin-Cy7–conjugated rat anti-mouse Ly6C (clone AL-21), PE-Cy7-conjugated rat anti-mouse Gr-1 (clone RB6-8C5), PerCpCy5.5-conjugated rat anti-mouse CD19 (clone 1D3) or rat anti-mouseCD11b (clone M1/70), biotin-conjugated rat anti-mouse CD3 (clone145-2C11) or rat anti-mouse CD19 (clone 1D3) or rat anti-mouse NK1.1(clone PK136) followed by streptavidin-BDHorizon V500 (all reagentspurchased from BD) and AF-700–conjugated rat anti-mouse CD62L(clone MEL-14), allophycocyanin-conjugated rat anti-mouse Foxp3 (cloneFJK-16s), eFluor 450–conjugated rat anti-mouse CD4 (clone RM4-5),eFluor 780–conjugated rat anti-mouse CD44 (clone IM7), FITC-conjugatedhamster anti-mouse CD3 (clone 145-2C11) or rat anti-mouse F4-80 (cloneBM8), PE-Cy5–conjugated rat anti-mouse CD8 (clone 53-67), PE-Cy7–conjugated rat anti-mouse CMHII (clone M5/114.15.2), PerCp Cy5.5-conjugated rat anti-mouse CD25 (clone PC61.5) (all Abs purchased fromeBioscience) and allophycocyanin-conjugated rat anti-mouse F4-80 (cloneCI-A3-1; Serotec, Raleigh, NC). For Ccr1 detection, leukocytes were incu-bated with a PE-conjugated goat anti-mouse Ccr1 (C-20) or PE-conjugatednormal goat IgG (both from Santa Cruz Biotechnology, Santa Cruz, CA).Different flow cytometric gating strategies were used to define monocytes(CD32CD192NK1.12F4-80lowCD11b+Ly6G2), including inflammatory(SSClowLy6C+) and patrolling (SSClowLy6C2) monocytes, activated mac-rophages (CD32CD192NK1.12F4-80highCD11b+CD11c2) classified intoproinflammatory M1 macrophages (Ly6Ghigh) and alternatively activatedanti-inflammatory M2 macrophages (Ly6Glow), and CD4+ T cells, amongwhich were naive cells (CD3+CD4+CD442CD62L+), effector/memorycells (CD3+CD4+CD44+CD62L+/2), and regulatory T cell (Tregs) (CD3+

CD4+CD44+CD62L+/2CD25+Foxp3+). Corresponding leukocyte subsetsfrom the kidney or spleen of Ccr12/2 mice were used to assess nonspecificbackground staining. Analyses were carried out on a FACS Fortessa (BDBiosciences) using FlowJo software (TreeStar, Ashand, OR). At least30,000 events were collected for each analysis after gating on forward andside scatter and CD45 to select leukocytes. Chemotaxis of BL5923 (10mM)–loaded splenic leukocytes was performed using a Transwell (CorningCostar, Tewksbury, MA) assay upon induction by Ccl3 (R&D Systems,Minneapolis, MN) as previously described (19).

Cell labeling, syngeneic adoptive transfer and short-termtreatment

Splenic leukocytes from nephritic NZB/W donor mice were isolated usingLympholyte-M (Burlington, Ontario, Canada) density gradient centrifu-gation and then incubated for 30 min at 37˚C with 7.2 mM CellTrackerOrange 5-(6)-(((4-chloromethyl) benzoyl) amino) tetramethylrhodamine(CMTMR) (Molecular Probes, Grand Island, NY). CMTMR labeling ef-ficiency was . 99% in splenic mononuclear phagocytes (F4-80+), in-cluding macrophages (F4-80+CD11b+), B (CD19+), and T (CD3+) cells ofdonor mice. A 200-ml volume of 3 3 106 CMTMR-labeled cells con-taining ∼39% B cells, ∼28% T cells, and ∼33% myeloid cells, as deter-mined by flow cytometry, was injected i.v. into recipient nephritic NZB/Wmice. Recipient mice were pretreated twice per day (6 h between doses)with either BL5923 (60 mg/kg) or vehicle the day before and the day ofsyngeneic cell transfer. Kidneys were collected from recipient mice 18 hafter injection of the labeled cells, and renal lymphocytes were analyzedby flow cytometry to quantify kidney-homing of fluorescent leukocytes.

Quantitative RT-PCR analyses

Total cellular RNAwas isolated from renal or splenic cryosections or fromfreshly sorted splenic myeloid, T, and B cells using the RNeasy Plus MiniKit (QIAGEN, Courtaboeuf, France) and reverse transcribed with poly-d(T)-15 and Moloney murine leukemia virus reverse transcriptase (FisherBioblock, Illkirch, France). RNA was quantified using NanoDrop tech-nology (Thermo Scientific, Wilmington, DE). For each sample, RNA

The Journal of Immunology 887

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

quality was assessed from the 18S and 28S ribosomal RNAs, and RNAintegrity was assessed using a BioAnalyzer (Agilent Technologies, SantaClara, CA). Samples with RNA integrity numbers . 7 were processed forgene expression analyses. cDNA amplification (1 mg) was performed byquantitative real-time PCR on a LightCycler 480 instrument (RocheDiagnostics) with the LightCycler 480 SYBR Green detection kit (RocheDiagnostics) using forward (454–473) 59-GACCTTCAACACCCCAG-CCA-39 and reverse (689–705) 59-CGGACGATTTCCCTCTCAGC-39primers for b-actin (216 bp), forward (952–974) 59-TTAGCTTCCATG-CCTGCCTTATA-39 and reverse (1031–1051) 59 TCCACTGCTTCAGG-CTCTTGT-39 primers for Ccr1 (57 bp), forward (120–140) 59-GCCCT-TGCTGTTCTTCTCTGT-39 and reverse (357–377) 59-GGCATTCAGT-TCCAGGTCAGT-39 primers for Ccl3 (217 bp), and forward (119–138) 59-CACCTGCCTCACCATATGGC-39 and reverse (249–268) 59-GGCGGT-TCCTTCGAGTGACA-39 primers for Ccl5 (111 bp). Reactions were per-formed using the following amplification scheme: 95˚C for 10 min, and40 cycles of 95˚C for 20 s, 60˚C for 15 s, and 72˚C for 20 s. The disso-ciation curve method was used according to the manufacturer’s protocol (60–95˚C) to verify the presence of a single specific PCR product. Relativequantificationwas performedwith the standard curvemethod (33), and resultswere expressed as Ccr1/b-actin, Ccl3/b-actin, and Ccl5/b-actin ratios.

Histological analyses

Kidneys were fixed overnight in 4% paraformaldehyde and then embeddedin paraffin. Paraffin sections (5 mm) were stained with H&E, periodic acid–Schiff stain, and Masson’s trichrome. IHC was performed on paraffinsections using Abs against the pan–B cell marker B220 (RA3-6B2; BDBiosciences), the pan–T cell marker CD3 (3256-1; Epitomics, Burlingame,CA), the mononuclear phagocyte marker F4-80 (BM40088; Interchim, SanPedro, CA), the glomerular macrophage marker CD68 (MCA1957; AbDSerotec, Oxford, U.K.), the proliferating cell marker Ki-67 (ACK02, LeicaMicrosystems, Buffalo Grove, IL), or the chemokine Ccl3 (AP8527b;Abgent, San Diego, CA). The sections were washed and incubated with anappropriate biotinylated secondary Ab for 1 h at room temperature, andthen with streptavidin-HRP complex followed by 3,39-diaminobenzidine or3-amino-9-ethylcarbazole detection (LSAB kit, Dako France, Trappes,France). Sections were then counterstained with hematoxylin. Negativecontrols were carried out by applying the same procedure with omission ofthe primary Ab. Images were obtained on a Leica DMLB microscope (LeicaMicrosystems) equipped with standard optic objectives, at the indicatedmagnification, and were digitized directly with a Sony 3CCD color videocamera (DXC-990P, Surrey, U.K.).

Immunochemical staining was interpreted and scored simultaneouslyby three independent investigators (F.G., M.P., and D.B.), who were blindedto the protocol. Pathological changes in the kidney were assessed byevaluating glomerular activity (i.e., glomerular proliferation, karyorrhexis/fibrinoid necrosis, cellular crescents, hyaline deposits, and inflammatorycells), tubulointerstitial activity (i.e., interstitial inflammation, tubular cellnecrosis, and flattening and tubular distension), and the chronicity of thelesions (i.e., fibrous crescents, tubular atrophy, and interstitial fibrosis), aspreviously reported (34). Sections were scored using a 0–3 scale for glo-merular activity, as follows: 0 = no lesions, 1 = lesions in , 25% ofglomeruli, 2 = lesions in 25250% of glomeruli, and 3 = lesions in . 50%of glomeruli. Tubulointerstitial activity and lesion chronicity indices werescored using a 0–4 scale, as follows: 0 = no lesions, 1 = lesions in 1210%,2 = 11225%, 3 = .25250% and 4 = .502100%. The mean scores forindividual pathological features were summed to obtain the three mainscores: the glomerular activity score, the tubulointerstitial activity score,and the chronic lesion score.

For immunofluorescence studies, paraffin kidney sections were blockedin a solution of 5% BSA in PBS for 30 min and stained with an anti-IgG Ablabeled with Alexa 488 (1:200; Invitrogen, Grand Island, NY) for 1 h.Mounting was done using Vectashield medium (Vector Laboratories,Burlingame, CA) that included 49,6-diamidino-2-phenylindole fluoro-chrome for nuclear staining. Images were acquired on a Zeiss microscope(Axio Imager 2) using AxioVision software (Carl Zeiss, Oberkochem,Germany).

ELISA

Sera were harvested from blood obtained by retro-orbital puncture, with theanimals under anesthesia. Titers of dsDNA Abs were determined by ELISAas previously described, with some modifications (35). Briefly, sera werediluted 1:100 and incubated overnight at 4˚C in wells coated with 250 mg/ml dsDNA (Phage l DNA; Roche). The plates were then incubated withperoxidase-conjugated goat anti-mouse IgG Fc-specific Abs (Sigma-Aldrich). After incubation with the 2,2-azino-bis(3-ethylbenzothiazoline-

6-sulfonic acid) substrate (Sigma-Aldrich), the reaction was stopped byadding 0.05 M citric acid and 0.01% sodium azide, and the plates wereread at OD405 nm. Total levels of IgG, IgM, and IgA were determined asdescribed previously (32). Serum concentrations of TNF-a, IL-1b, IL-6,and IL-10 were assessed using a standardized ELISA (Quantikine R&DSystems, Rochester, MN). Sera from prenephritic and nephritic NZB/Wmice were used as negative and positive controls, respectively.

Statistical analysis

Unless specified, data are expressed as means 6 SD. The statistical sig-nificance between groups was evaluated using the unpaired two-tailedStudent t test and the log-rank test (Prism software; GraphPad, La Jolla,CA).

ResultsCcl3/Ccl5–Ccr1 axis expression increases in the kidneys ofnephritic mice

NZB/W mice with severe proteinuria had significantly higherabsolute numbers of renal leukocytes than did young, prenephriticNZB/W mice (Fig. 1A). Consistent with previous reports (19, 36),these cells included CD3+ T cells, myeloid cells including F4-80+

mononuclear phagocytes and Gr-1+ neutrophils, and, to a lesserextent, CD19+ B cells. Because the Ccl3/Ccl5–Ccr1 axis has beenimplicated in renal leukocyte recruitment in rodent models ofnephritis (14, 18, 19, 23, 37), we used real-time PCR to quantifyand compare the levels of Ccr1, Ccl3, and Ccl5 mRNAs in ne-phritic kidneys and nondiseased kidneys from young mice. Ccr1,Ccl3, and Ccl5 mRNA levels were significantly higher in nephriticNZB/W mouse kidneys than in those of prenephritic mice (Fig.1B). Previous studies have shown that kidney cells do not expressCcr1 and that renal Ccr1 synthesis mainly originates from infil-trated leukocytes (17, 18). Therefore, infiltrated leukocytes alonecould account for the higher levels of these mRNAs in the ne-phritic kidneys of NZB/W mice. Consistent with this assertion,surface Ccr1 receptor levels, that is, frequencies of Ccr1-positivecells and geometric mean fluorescence intensity values, were higherin the CD4+, CD8+, and Treg T cell subsets, proinflammatoryLy6C+ monocytes and M1 macrophages, and anti-inflammatoryM2 macrophages of nephritic mice than in the leukocytes of prene-phritic mice (Fig. 1C, Supplemental Fig. 1A). In contrast, patrol-ling Ly6C2 monocytes from nephritic and prenephritic mice hada similar pattern of membrane Ccr1, whereas renal B cells ex-hibited very low to undetectable levels of receptors. Next, weimmunostained kidney sections from prenephritic and nephriticNZB/W mice to identify the source of the Ccr1 ligand Ccl3.Total Ccl3 staining was significantly greater in nephritic thanin prenephritic mouse kidneys (Fig. 1D). Ccl3-positive cellswere mainly infiltrated leukocytes within glomerular and tubu-lointerstitial compartments, although some tubular epithelialand glomerular cells stained weakly for Ccl3. Unfortunately, wewere unable to study local expression of Ccr1 in the kidneyresulting from the lack of a suitable Ccr1 Ab for IHC analyses.Collectively, these findings reveal that leukocyte Ccr1 and Ccl3production is abnormally high in the kidneys of nephritic NZB/Wmice.

Ccr1 is expressed by splenic T and myeloid cells from NZB/Wmice

As splenomegaly is a feature of nephritic NZB/W mice and thespleen constitutes a source of effector leukocytes (19), we nextassessed the levels of Ccr1, Ccl3, and Ccl5 mRNAs in spleencryosections from prenephritic and nephritic NZB/W mice.Transcripts encoding all three proteins were readily detectable inthe spleens of prenephritic mice (Fig. 2A). However, in accor-

888 ROLE OF Ccr1 IN LUPUS NEPHRITIS

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

dance with a previous study (15), splenic Ccr1, Ccl3, and Ccl5mRNA levels were significantly (30240%) higher in nephriticmice than in prenephritic mice. To identify which leukocytesubsets expressed Ccr1, T and B cells, mononuclear phagocytes,and neutrophils were sorted from the spleens of prenephritic andnephritic mice and analyzed for steady-state levels of Ccr1mRNA. Real-time PCR analysis revealed that Ccr1 mRNA levelswere highest in T and myeloid cells and that they were signifi-cantly higher in nephritic mice than in prenephritic mice (Fig. 2B).In contrast, Ccr1 transcript levels were very low in splenic B cellsfrom both prenephritic and nephritic mice. Flow cytometricanalysis of Ccr1 on the surface of these cells also corroborated theRT-PCR results. The Ccr1 geometric mean fluorescence intensityvalues and the fractions of Ccr1-positive cells in the splenicT cells, mononuclear phagocytes, and neutrophils of nephriticNZB/W mice were 2- to 3-fold higher than in those obtained fromprenephritic mice (Fig. 2C, Supplemental Fig. 1B). This findingencompassed CD4+ and CD8+ T cells; Tregs; Ly6C+, but notLy6C2, monocytes; and M1/M2 macrophages. As expected, Ccr1was undetectable on the surface of B cells from either prenephriticor nephritic mice. Taken together, these data indicate that Ccr1gene expression is upregulated in the splenic T and myeloid cellsof nephritic mice, which led us to investigate the functionalconsequences of this anomaly.

Ccr1 is functional on splenic T and mononuclear phagocytecells from NZB/W mice

We compared the in vitro Ccl3-induced migration of splenicleukocytes obtained from prenephritic and nephritic NZB/W mice,

using a Transwell-based chemotaxis assay. Whereas random mi-

gration for cells from both types of mice was roughly comparable

(Fig. 3A), Ccl3 (102100 nM) stimulated T cell and mononuclear

phagocyte migration across the Transwell membrane, resulting in

bell-shaped dose–response patterns for cells from both types of

mice. However, the migratory responses of T cells and mononu-

clear phagocytes from nephritic mice were greater than those of

prenephritic mice at all Ccl3 concentrations tested, suggesting that

the leukocytes of nephritic mice were more sensitive to Ccl3 than

those of prenephritic mice. Migratory responsiveness to Ccl3 was

abolished by the specific Ccr1 antagonist BL5923, indicating that

this Ccl3-stimulated migration was Ccr1 dependent. These results

are consistent with the higher levels of Ccr1 receptors found on

the surfaces of T cells and mononuclear phagocytes from nephritic

mice. In both prenephritic and nephritic NZB/W, B cells were

refractory to Ccl3-promoted chemotaxis (data not shown). To

exclude potential interference with Ccl3-enhanced leukocyte

chemotaxis by the splenic microenvironment (e.g., production of

soluble proinflammatory mediators), T cells and mononuclear

FIGURE 1. Kidney expression of Ccr1 and its ligands in nephritic NZB/W mice. (A) Distribution of T cells (CD3), B cells (CD19), mononuclear

phagocytes (F4-80), and neutrophils (Gr-1) in kidneys of prenephritic and nephritic NZB/W mice was determined by flow cytometry. (B) Levels of Ccr1,

Ccl3, and Ccl5 transcripts in renal cryosections from prenephritic and nephritic NZB/W mice were assessed by real-time PCR. Each individual sample was

run in triplicate. Results are expressed as Ccr1/b-actin, Ccl3/b-actin, and Ccl5/b-actin ratios. (C) Membrane expression of Ccr1 on T cells, including CD4+,

CD8+ and Tregs, B cells, Ly6C+ and Ly6C2 monocytes, and M1/M2 macrophages from kidneys of prenephritic and nephritic NZB/W mice was determined

by flow cytometry using a PE-conjugated goat anti-mouse Ccr1 Ab. Corresponding leukocyte subsets from the kidneys of Ccr12/2 mice were used to assess

nonspecific background staining (shaded area). (D) Kidney sections from prenephritic and nephritic NZB/W mice were immunostained for Ccl3 (red,

shown by white arrows) and those of Ccl32/2 mice were used as a negative control. Tissues were counterstained with hematoxylin. Original magnifications

were 3400 (i, iii, v) and 3630 (ii, iv, and vi). Scale bars denote 50 and 25 mm, respectively, for 3400 and 3630 magnifications. Results represent the

mean6 SD of four independent experiments (A and B) or are representative of three to four independent determinations (C and D). *p, 0.05, **p, 0.005,

***p , 0.0005 compared with prenephritic NZB/W mice.

The Journal of Immunology 889

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

phagocytes were sorted from the spleens of prenephritic and ne-phritic mice and analyzed for their ability to migrate towarda maximally effective concentration of Ccl3 (50 nM) (Fig. 3B).Ccl3-stimulated chemotaxis was enhanced in cells from nephriticmice compared with the cells of prenephritic mice, and this effectwas inhibited by BL5923. Although Ccl3 also stimulated themigration of the corresponding leukocyte subsets from Ccr12/2

mice, this likely occurred through Ccl3 receptors other than Ccr1,such as Ccr5. Indeed, cells from Ccr12/2 mice remained com-pletely insensitive to BL5923 treatment, emphasizing the speci-ficity of this antagonist for Ccr1. Overall, these findings indicatethat the increased expression of Ccr1 on NZB/W leukocytes isassociated with enhanced migration to Ccl3 gradients.

Ccr1 blockade reduces the recruitment of T and macrophagesto the inflamed kidney

We then investigated whether fluorescently labeled splenic leu-kocytes isolated from donor nephritic NZB/W mice would berecruited into the kidneys of nephritic syngeneic mice pretreatedwith BL5923. Short-term BL5923 treatment reduced by ∼70% thenumbers of donor T cells and mononuclear phagocytes, includingmacrophages, but not of B cells, that infiltrated the kidneys ofrecipient mice (Fig. 3C). These findings, combined with those

displayed in Figs. 1 and 2, support the idea that abnormal renalhoming of T and macrophages in nephritic mice depends on thepresence of functional Ccr1 on these leukocytes.

Long-term BL5923-based treatment prolongs the lifespan ofNZB/W mice

To explore whether the heightened activity of the Ccl3/Ccl5–Ccr1axis in NZB/W mice could constitute a valid therapeutic target inlupus, we initiated long-term BL5923 treatment in NZB/W micewith already advanced nephritis. Mice with grade 3+ proteinuriawere randomized to receive BL5923 or vehicle for 6 wk. Invehicle-treated mice, proteinuria continued to increase rapidlyduring the next 4 wk of follow-up (Fig. 4A). In contrast, pro-teinuria progressed more slowly to the fatal stage in BL5923-treated mice. The difference between the two groups becamesignificant as early as 1 wk after treatment initiation and remainedso until 5 wk of follow-up. In addition, BL5923 treatment mark-edly prolonged the lifespan of nephritic mice (Fig. 4B). Thus,taken together with its ability to inhibit renal leukocyte homing,the beneficial effects of BL5923 treatment suggest that therapeuticCcr1 blockade slows disease progression by inhibiting the infil-tration of T and myeloid cells into the inflamed kidneys of ne-phritic mice.

FIGURE 2. Spleen expression of Ccr1 and its ligands in nephritic NZB/W mice. (A and B) Levels of Ccr1, Ccl3, and Ccl5 transcripts in spleen cry-

osections (A) and of Ccr1 mRNAs in sorted T and B cells, mononuclear phagocytes, and neutrophils from the spleen (B) of prenephritic and nephritic

NZB/W mice were assessed by real-time PCR. Each individual sample was run in triplicate. Results are expressed as Ccr1/b-actin, Ccl3/b-actin, and

Ccl5/b-actin ratios. (C) Membrane expression of Ccr1 on splenic T cells, including CD4+, CD8+, and Tregs, B cells, Ly6C+ and Ly6C2 monocytes, and

M1/M2 macrophages from prenephritic and nephritic NZB/W mice. Background fluorescence (shaded area) was assessed in Ccr12/2 splenic leukocytes.

Results represent the mean6 SD of four independent experiments (A and B) or are representative of four independent determinations (C). *p, 0.05, **p,0.005 compared with prenephritic NZB/W mice.

890 ROLE OF Ccr1 IN LUPUS NEPHRITIS

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Therapeutic Ccr1 inhibition reduces renal injuries in nephriticNZB/W mice

To investigate the ability of BL5923 to hamper the progression ofrenal damage, we treated NZB/W mice with established severe 3+proteinuria with either BL5923 or vehicle for 10 d and then an-alyzed the kidneys of the treated animals. At this endpoint, thedifference in proteinuria severity between both groups was sig-nificant, but most mice from both groups were still alive (Fig. 4).Like nephritic mice (Fig. 1A), vehicle-treated mice exhibited renalaccumulation of T cells, in particular the CD4+ subset, myeloidcells, and, to a lesser extent, B cells (Fig. 5, Supplemental Fig.2A). In BL5923-treated mice, the absolute numbers of CD4+

T cells, monocytes, macrophages, and granulocytes were signifi-cantly decreased in the kidneys. Analyses of T cell and monocyte/macrophage subsets indicated that effector/memory CD4+ T cells,Ly6C+ monocytes, and both M1 and M2 macrophages were tar-geted by BL5923 treatment. This result was associated with lowerrenal levels of Ccr1, Ccl3, and Ccl5 mRNAs (Supplemental Fig.2B) and decreased levels of serum inflammatory cytokines such asTNF-a and IL-1b (Supplemental Fig. 3). In contrast, B cells,Tregs, and Ly6C2 monocytes were not affected by Ccr1 antago-nism (Fig. 5, Supplemental Fig. 2). Histological analyses con-firmed these results and revealed beneficial effects of Ccr1blockade on tubulointerstitial and glomerular injuries (Fig. 6,Table I). Indeed, kidney tissue sections from vehicle-treated miceshowed marked glomerulonephritis; tubular dilatation; tubularprotein cast deposition; Ki-67–positive proliferation of tubularepithelial, interstitial, and glomerular cells; and infiltration ofB cells, T cells, and mononuclear phagocytes in the interstitial andglomerular areas. In contrast, sections from BL5923-treated micedisplayed significantly less glomerular and tubulointerstitialdamage; less karyorrhexis/fibrinoid necrosis; absence of glomer-ular crescents, proliferating leukocytes and tubular epithelial cells;

and fewer glomerular macrophages as well as interstitial T cellsand mononuclear phagocytes. Note that the mean scores for the

FIGURE 3. Functional expression of Ccr1 on

splenic leukocytes from nephritic NZB/W mice.

(A and B) Total spleen cells (A) or sorted splenic

T cells and mononuclear phagocytes (B) of NZB/W

(prenephritic versus nephritic) and Ccr12/2 mice

were tested for their ability to migrate in response

to 102100 nM (A) or 50 nM (B) Ccl3. Inhibition

of Ccr1-mediated chemotaxis by BL5923 (10 mM)

added to both chambers is shown. Transmigrated

cells recovered in the lower chamber were stained

with mAbs specific for CD3 and F4-80 Ags and

counted by flow cytometry. Results (mean6 SD) are

from three independent experiments and expressed

as the percentage of input total cells that migrated to

the lower chamber. (C) NZB/W mice with advanced

nephritis were injected i.v. with CMTMR-labeled

leukocytes isolated from spleens of donor NZB/W

mice. Recipient mice were pretreated twice per day

with either 60 mg/kg BL5923 or vehicle (6 h be-

tween doses) on the day before and the day of

cell transfer. Kidneys were collected 18 h after in-

jection of cells, and the infiltrated leukocytes were

analyzed by flow cytometry. Results (mean 6 SD)

are expressed as the number of fluorescent T cells, B

cells, or mononuclear phagocytes including macro-

phages (F4-80+CD11b+) recruited to the kidneys of

recipient NZB/W mice (treated or not with 60 mg/kg

BL5923; n = 11 per group) per 105 transferred leu-

kocytes and are from eight independent experiments.

*p , 0.05, **p , 0.005, ***p , 0.0005 com-

pared with leukocytes from prenephritic NZB/W or

Ccr12/2 mice or with vehicle-treated NZB/W mice.

FIGURE 4. BL5923 administration delays death in nephritic NZB/W

mice. Groups of 10 NZB/W mice with severe proteinuria (3+) were treated

by oral gavage three times per week for 42 d with either 60 mg/kg BL5923

or vehicle. (A) Proteinuria (mean grade6 SD) was assessed once per week

during the time course of the therapeutic protocol. *p , 0.05 compared

with vehicle-treated NZB/W mice. (B) Survival rate. ↑, End of treatment.

The p value is that obtained in the log-rank test.

The Journal of Immunology 891

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

glomerular activity, the tubulointerstitial activity, and the chro-nicity of the lesions were roughly comparable between BL5923-treated and untreated mice, that is, before the onset of treatment(Table I, Fig. 6B). Vehicle- and BL5923-treated mice exhibitedno substantial differences in tubular or glomerular infiltration ofB cells, and BL5923 treatment did not affect systemic humoralimmune responses or intraglomerular IgG deposition in nephriticNZB/W mice (Fig. 6, Supplemental Fig. 2C, 2D). Overall, thesefindings indicate that long-term BL5923 treatment delays tubu-lointerstitial and glomerular injuries in NZB/W mice, most likelyby attenuating Ccr1-mediated interstitial recruitment of T andmononuclear phagocyte cells.

DiscussionBy combining pharmacological and functional approaches, weunveiled a pathogenic role for Ccr1 in lupus nephritis in NZB/Wmice. Our findings revealed elevated levels of Ccr1 receptors onperipheral mononuclear phagocytes, neutrophils, and T cells innephritic NZB/W mice, which were associated with greater leu-kocyte chemotaxis to Ccl3. Syngeneic adoptive transfer experi-ments using ex vivo–labeled splenic leukocytes combined withshort-term Ccr1 blockade by the orally available BL5923 antag-onist identified a role for Ccr1 in the abnormal homing of mac-

rophages and T cells into the kidney. Longer BL5923 treatmentin mice with established nephritis attenuated Ccr1-driven kid-ney accumulation of effector/memory CD4+ T cells, Ly6C+

monocytes, and both M1 and M2 macrophages, and reducedtubulointerstitial and glomerular injuries. These processes wereassociated with renal disease improvement, which resulted indelayed onset of fatal proteinuria and a prolonged lifespan. Toour knowledge, this is the first evidence in NZB/W mice that Ccr1blockade can slow renal disease progression by selective attenu-ation of interstitial and glomerular recruitment of T and mono-nuclear phagocyte cells.Our results extend previous studies indicating that functional

Ccr1 receptors are present on mononuclear phagocytes, neu-trophils, and T cells and that Ccr1 inactivation prevents the in-terstitial infiltration of Ccr1-expressing leukocytes in variousnonlupus rodent models (13, 14). In this study, we unraveled thatCcr1 receptor biosynthesis is upregulated in splenic T and myeloidcells from nephritic NZB/W mice, which likely arises as a con-sequence of the disease process. Transcriptional expression ofCcr1 has been reported in splenic macrophages and T cells fromMRL-Fas(lpr) mice (17). In line with this, our real-time PCRanalyses revealed increased levels of Ccr1 mRNAs in T and my-eloid cells sorted from the spleens of nephritic NZB/W mice

FIGURE 5. Ccr1 blockade attenuates renal accumulation of effector/memory T cells and macrophages in nephritic NZB/W mice. Groups of eight

NZB/W mice with severe proteinuria were treated by oral gavage twice per day with either 60 mg/kg BL5923 or vehicle (6 h between doses) for 10 d. At the

end of the treatment, kidneys were analyzed by flow cytometry for the frequencies (A) and absolute numbers (B) of total monocytes (Mono) expressing the

Ly6C marker or not; macrophages (Mac), including M1 and M2 subsets; and CD4+ T cells, including the naive, effector/memory (Eff/Mem) and Treg

subsets. Results represent the mean6 SD (B) or are representative (A) of eight mice in each group. *p, 0.05, **p, 0.005, ***p, 0.0005 compared with

vehicle-treated NZB/W mice.

892 ROLE OF Ccr1 IN LUPUS NEPHRITIS

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

compared with those of younger prenephritic mice. Importantly,this was corroborated at the protein level by the increased levelsof Ccr1 receptors on splenic T cells, including CD4+, CD8+, andTreg subsets, proinflammatory Ly6C+ monocytes and macrophages(M1 and M2), and neutrophils from nephritic mice, and by theirenhanced Ccl3-stimulated chemotaxis in vitro. Note that Ccl3promoted typical bell-shaped dose–response chemotactic patternsof leukocytes from NZB/W mice. This finding suggests that en-hanced migration of leukocytes from nephritic mice results froman increased expression of Ccr1 rather than from an impairedreceptor inactivation. One can speculate that such increased ex-pression and signaling through Ccr1 accounts for renal T andmyeloid cell accumulation in nephritic NZB/W mice.One important question that arises from these findings concerns

the molecular mechanism responsible for Ccr1 upregulation inmononuclear phagocytes, neutrophils, and T cells from nephriticmice. The different cytokine and sex hormone milieu in NZB/Wmice may affect the expression and activity of Ccr1. High se-rum concentrations of TNF-a, TGF-b, and IFN-g have been re-ported in lupus nephritis, and these cytokines are known to induceCcr1 gene expression, notably in myeloid and T cells (38, 39).Estrogens that regulate inflammation may amplify this phenom-enon (40, 41). However, epigenetic (e.g., promoter hypomethyl-ation) or posttranscriptional (e.g., microRNA) regulatory mechanismscould account for the increased steady-state levels of Ccr1mRNAs

in nephritic NZB/W mice. Further investigation is needed to de-termine the specific relevance of each of these factors in Ccr1biology.A second question that arises from our findings pertains to the

cellular mechanism of Ccr1-mediated accumulation of myeloid andT cells in the nephritic kidneys of NZB/Wmice. This accumulationmay result from leukocyte influx, retention, and/or proliferation.Short-term treatment of nephritic NZB/W mice with BL5923significantly reduced the number of fluorescently labeled T cellsand macrophages found in the kidneys of recipient mice. Theseresults are consistent with a Ccr1-dependent homing of leukocytesfrom the blood to the kidney. In line with this, in vitro and in vivostudies have reported a role for Ccr1 in T cell and mononuclearphagocyte adhesion and transendothelial migration (14, 17, 18, 27,42, 43). Thus, BL5923-induced inhibition of renal T cell andmacrophage infiltration could be related to reduced Ccr1-mediatedadhesion of leukocytes to activated peritubular capillaries. How-ever, Ki-67 staining of renal tissue sections revealed fewer pro-liferating leukocytes in long-term BL5923-treated mice. Consis-tent with these results, Ninichuk and coworkers (18) showed thatBL5923 treatment in vitro reduced proliferation of the murinemacrophage J774 cell line in the presence of serum, whereasBL5923 treatment in vivo reduced the number of interstitial pro-liferating cells (which likely included some leukocytes) in ne-phropathy of type 2 diabetic db/db mice. This could constitute

FIGURE 6. Therapeutic Ccr1 inhibition reduces re-

nal damage in nephritic NZB/W mice. Groups of eight

NZB/W mice with severe proteinuria were treated as

described in Fig. 5. (A) Kidney sections from vehicle-

or BL5923-treated mice were stained for the indicated

markers. Original magnifications 3400 (i, iii, v, vii, ix,

i9, iii9, v9, vii9, and ix9) and 31000 (ii, iv, vi, viii, x, ii9,

iv9, vi9, viii9, and x9). Scale bars denote 50 mm and 12

mm for 3400 and 31000 magnifications, respectively.

Arrows indicate infiltrating leukocytes and proliferating

cells. (B) Intermediate magnification images of repre-

sentative kidney sections stained with PAS solution

(i–iii). Images show inflammatory renal damage with

interstitial infiltrates (arrows), glomerulonephritis, and

intratubular protein cast deposition (*) in a vehicle-

treated mouse (ii), and renal histology with signifi-

cantly fewer glomerular and tubulointerstitial injuries

in untreated or BL5923-treated mice (i and iii). Original

magnification 3250. High-magnification image of

representative glomeruli from a vehicle-treated mouse

(v) showing glomerular crescents (*), severe endoca-

pillary proliferation involving all the glomerular tuft

(arrowhead), hyaline deposits (arrows), and karyor-

rhexis. Images of representative glomeruli from un-

treated (iv) or BL5923-treated (vi) mice, showing only

mild endocapillary proliferation (arrowhead), either

focal (iv) or global (vi). Original magnification 31250.

Scale bars denote 100 mm and 8 mm for 3250 and

31250 magnifications, respectively. Results are repre-

sentative of eight mice in each group.

The Journal of Immunology 893

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

another mechanism by which BL5923 reduces the number oftubulointerstitial and glomerular T and myeloid cells in nephriticNZB/W mice. These observations support the assertion that Ccr1promotes leukocyte extravasation into the renal interstitial com-partment and cell proliferation after kidney infiltration.In B cells, however, Ccr1 and its ligands were clearly dispensable

for the homing, retention, and likely function of these cells in thenephritic kidneys of NZB/Wmice. Several convergent observationssupport this idea. First, very low levels of Ccr1 mRNA and proteinwere detected in splenic and renal B cells from both prenephriticand nephritic NZB/W mice. As a consequence, these cells wereinsensitive to Ccl3-promoted chemotaxis. Second, short-termBL5923treatment of nephritic NZB/Wmice did not prevent the recruitmentof B cells to the inflamed kidneys. Third, longer Ccr1 blockadealtered neither the absolute number of B cells nor their interstitialand glomerular partitioning in the kidneys of nephritic NZB/Wmice. Finally, BL5923-mediated blockade of Ccr1 did not affectsystemic or renal humoral autoimmunity in nephritic NZB/Wmice.These findings suggest differential chemokine receptor require-ments for B cells versus T and myeloid cells in their renal homingand compartmentalization in nephritic NZB/W mice (19, 35, 44).For instance, the ligands of CXCR4 and CXCR5, which are up-regulated in the nephritic kidneys of NZB/W mice, are likely to bemajor contributors to this pathological process.The third important question concerns the nature of the T or

myeloid cell populations that drive kidney pathogenesis in a Ccr1-dependent manner in this model of lupus nephritis. The use ofCcr1-deficient mice or small-molecule Ccr1 antagonists efficientlyreduced interstitial infiltration of T cells, mononuclear phagocytes,and/or neutrophils in different mouse models of nephropathy (13,14, 17, 18, 23, 27, 28, 43). In particular, s.c. injections of the Ccr1antagonist BX471 in MRL-Fas(lpr) mice led to reduced numbersof macrophages, T cells, and Ki-67 positive proliferating cells andreduced fibrosis in the interstitium. However, treatment did notmodulate proteinuria or markers of glomerular injury (17). In thisarticle, we showed that a 10-d interventional treatment using theorally available Ccr1 antagonist BL5923 resulted in decreasednumbers of effector/memory CD4+ T cells, Ly6C+ monocytes,both M1 and M2 macrophages, and neutrophils in the kidneys ofNZB/W mice. Concomitantly, Ccr1 blockade reduced the in situnumbers of interstitial T cells and mononuclear phagocytes, aswell as glomerular macrophages, and improved tubulointerstitial,

and, surprisingly, glomerular injuries. We speculate that Ccr1 in-hibition reduces the renal recruitment of Ly6C+ monocytes andtheir differentiation toward M1-polarized macrophages, whichdepends on the local inflammatory environment encompassingcytokines (e.g., INF-g, TNF-a), chemokines (e.g., Ccl3, Ccl5),and renal cell necrosis (10). Consequently, BL5923 treatmentwould decrease proinflammatory M1 and profibrotic M2 polari-zation of recruited and resident renal macrophages, which is as-sociated with a reduction in renal cell damage, as well as a delayin the vicious circle of renal inflammation and parenchymal loss,thus resulting in prolonged survival of NZB/W mice (10, 45, 46).In contrast, Tregs and Ly6C2 monocytes were not affected byBL5923 treatment, suggesting that Ccr1 is not a dominant majorreceptor for their kidney infiltration and accumulation. The ben-eficial effect of BL5923 treatment also suggests that Tregs andLy6C2 monocytes are not major players in kidney pathogenesis inthis model of lupus nephritis. Therefore, selectively modulatingthe number of interstitial and glomerular mononuclear phag-ocytes, neutrophils, or T cells using clodronate-loaded liposomesor Ab-depleting strategies, in combination with noninvasive oraladministration of BL5923, could help to determine the functionalcontribution of each leukocyte subpopulation to the progression ofautoimmune nephropathy in NZB/W mice.Another related finding in our work is that therapeutic Ccr1

blockade led to a reduced content of Ccr1 and its cognate ligandsCcl3 and Ccl5 in the kidneys of NZB/Wmice. These findings are inline with previous studies (17, 18), which indicate that nonimmuneresident kidney cells do not express Ccr1 and that Ccr1 expressionoriginates from infiltrated T and myeloid cells. Hence, BL5923-mediated reduction of Ccr1 expression in the kidney was likelydue to the attenuation of Ccr1-promoted T and myeloid cell in-filtration. This process could also account for the decreased rep-resentation of Ccr1 ligands in nephritic kidneys. Indeed, wedetected more Ccl3 and Ccl5 protein and mRNA in injured kid-neys from nephritic mice than in control kidneys from prenephriticmice. Infiltrated leukocytes constituted the major source of renalCcl3, whereas tubular and glomerular cells exhibited only weakCcl3 staining. In agreement, previous studies have reported that Ccl3and Ccl5 can be produced by infiltrating leukocytes, includingmononuclear phagocytes, neutrophils, and T cells (14, 17, 47). Ofinterest, we observed a significant decrease in serum inflamma-tory cytokines TNF-a and IL-1b upon BL5923 exposure. In

Table I. Kidney biopsy index of nephritic NZB/W mice

Treatment Untreated Vehicle BL5923

Glomerular activity indexGlomerular proliferation 2.5 (2–3) 2.75 (2–3) 2.25 (1–3)Karyorrhexis/fibrinoid necrosis 0 1.38 (1–3) 0.50 (0–2)Cellular crescents 0 1.50 (1–2) 0***Hyaline deposits 0 1.75 (1–3) 0.38 (0–1)**Polymorphonuclear leukocytes 0.75 (0–1) 1.0 (0–2) 0.38 (0–1)*Score 3.25 (2–4) 8.38 (6–11) 3.5 (1–7)**

Tubulointerstitial activity indexTubular cell lesions 0 1.25 (1–2) 0***Tubular distension 0.5 (0–1) 1.88 (1–3) 0.25 (0–1)**Interstitial inflammation 0.25 (0–1) 1.75 (1–2) 1.0 (0–2)*Score 0.75 (0–2) 4.88 (3–7) 1.25 (0–2)***

Chronicity lesions indexFibrous crescents 0 0.25 (0–1) 0.13 (0–1)Tubular atrophy 0 0 0Interstitial fibrosis 0 0.38 (0–1) 0.13 (0–1)Score 0 0.63 (0–2) 0.25 (0–2)

Total score 3.75 (3–4) 13.88 (10–18) 5.0 (2–11)**

Data are means, with ranges in parentheses.*p , 0.05, **p , 0.005, ***p , 0.0005 compared with vehicle treatment.

894 ROLE OF Ccr1 IN LUPUS NEPHRITIS

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

combination with the observation that BL5923 reduced the num-ber of T and mononuclear phagocyte cells, which likely constitutea source of Ccr1 ligands and other inflammatory and profibroticmediators, in damaged kidneys, these data suggest that Ccr1contributes to leukocyte recruitment to the kidney and amplifica-tion of the inflammatory response, which involves renal cells andleukocytes themselves and participates in nephritis progression.Finally, our findings suggest that blocking the Ccr1-mediated

influx of T and myeloid cells is sufficient to improve lupus ne-phritis in NZB/W mice. Long-term BL5923 treatment reducedtubulointerstitial and glomerular injuries, delayed proteinuria, andprolonged the lifespan of nephritic NZB/W mice. Thus, by pre-venting renal infiltration of T cells, mononuclear phagocytes, andneutrophils, BL5923 might reduce detrimental inflammation inboth tubulointerstitial and glomerular compartments, and result inprolonged survival. In contrast to the effects of BX471 adminis-tration in the MRL-Fas(lpr) model (17), BL5923 treatment did notimprove renal fibrosis in the 10-d window of treatment in NZB/Wmice. Additional work is required to decipher the functional roleand renal distribution of proinflammatory (M1), as well as thebalance of anti-inflammatory (M2c) and profibrotic (M2a) mac-rophages and their contribution to renal fibrosis (10), together withCcr1 as a master regulator of kidney infiltration during the pro-gression of lupus nephritis. Immunotherapies aimed at specificcell types are of interest in SLE because they could be used toimprove current treatments based only on nonspecific immuno-suppressive drugs. Our data provide a rationale for potentiallyadding BL5923 to the treatment arsenal of patients with SLE. Bylimiting Ccr1 activation, oral treatment with BL5923 would at-tenuate the aberrant interstitial and glomerular accumulation ofleukocytes in the kidneys and thus reduce the severity of lupus-related nephropathy.

AcknowledgmentsWe thank H. Gary and L. Bouchet-Delbos (Institut Paris-Sud d’Innovation

Therapeutique/Institut Federatif de Recherche 141, Chatenay-Malabry,

France) for excellent technical assistance; Drs. C. Combadiere (INSERM,

UMR_S945, UPMC, Universite Paris 6, Paris, France) and F. Sennlaub

(INSERM, UMR_S968, UPMC, Universite Paris 6, Institut de la Vision,

Paris, France) for providing Ccr12/2 and Ccl32/2 mice, respectively; Prof.

A. Dalloul, Dr. V. Machelon, and Dr. F. Bachelerie (Universite Paris-Sud,

Laboratoire “Cytokines, Chimiokines et Immunopathologie,” INSERM

UMR_S996, Clamart, France) for critical reading of the manuscript; and

Drs. J. Harriague and K. Liittschwager (4Clinics, Paris, France) for editing

the manuscript.

DisclosuresP.L. is an employee of Novartis Pharma AG, which holds patents for

BL5923. The funders had no role in study design, data collection and anal-

ysis, decision to publish, or preparation of the manuscript.

References1. Rahman, A., and D. A. Isenberg. 2008. Systemic lupus erythematosus. N. Engl.

J. Med. 358: 929–939.2. Tsokos, G. C. 2011. Systemic lupus erythematosus. N. Engl. J. Med. 365: 2110–

2121.3. Lech, M., and H. J. Anders. 2013. The pathogenesis of lupus nephritis. J. Am.

Soc. Nephrol. 24: 1357–1366.4. Saxena, R., T. Mahajan, and C. Mohan. 2011. Lupus nephritis: current update.

Arthritis Res. Ther. 13: 240.5. Davidson, A., and C. Aranow. 2010. Lupus nephritis: lessons from murine

models. Nat. Rev. Rheumatol. 6: 13–20.6. Holdsworth, S. R., and P. G. Tipping. 2007. Leukocytes in glomerular injury.

Semin. Immunopathol. 29: 355–374.7. Borchers, A. T., N. Leibushor, S. M. Naguwa, G. S. Cheema, Y. Shoenfeld,

and M. E. Gershwin. 2012. Lupus nephritis: a critical review. Autoimmun. Rev.

12: 174–194.

8. Teramoto, K., N. Negoro, K. Kitamoto, T. Iwai, H. Iwao, M. Okamura, andK. Miura. 2008. Microarray analysis of glomerular gene expression in murinelupus nephritis. J. Pharmacol. Sci. 106: 56–67.

9. Anders, H. J., V. Vielhauer, M. Frink, Y. Linde, C. D. Cohen, S. M. Blattner,M. Kretzler, F. Strutz, M. Mack, H. J. Grone, et al. 2002. A chemokine receptorCCR-1 antagonist reduces renal fibrosis after unilateral ureter ligation. J. Clin.Invest. 109: 251–259.

10. Anders, H. J., and M. Ryu. 2011. Renal microenvironments and macrophagephenotypes determine progression or resolution of renal inflammation and fi-brosis. Kidney Int. 80: 915–925.

11. Thelen, M., and J. V. Stein. 2008. How chemokines invite leukocytes to dance.Nat. Immunol. 9: 953–959.

12. Ohl, K., and K. Tenbrock. 2011. Inflammatory cytokines in systemic lupuserythematosus. J. Biomed. Biotechnol. 2011: 432595.

13. Lionakis, M. S., B. G. Fischer, J. K. Lim, M. Swamydas, W. Wan, C. C. RichardLee, J. I. Cohen, P. Scheinberg, J. L. Gao, and P. M. Murphy. 2012. Chemokinereceptor Ccr1 drives neutrophil-mediated kidney immunopathology and mor-tality in invasive candidiasis. PLoS Pathog. 8: e1002865.

14. Furuichi, K., J. L. Gao, R. Horuk, T. Wada, S. Kaneko, and P. M. Murphy. 2008.Chemokine receptor CCR1 regulates inflammatory cell infiltration after renalischemia-reperfusion injury. J. Immunol. 181: 8670–8676.

15. Turner, J. E., H. J. Paust, S. B. Bennstein, P. Bramke, C. Krebs, O. M. Steinmetz,J. Velden, F. Haag, R. A. Stahl, and U. Panzer. 2012. Protective role for CCR5 inmurine lupus nephritis. Am. J. Physiol. Renal Physiol. 302: F1503–F1515.

16. Perez de Lema, G., H. Maier, E. Nieto, V. Vielhauer, B. Luckow, F. Mampaso,and D. Schlondorff. 2001. Chemokine expression precedes inflammatory cellinfiltration and chemokine receptor and cytokine expression during the initiationof murine lupus nephritis. J. Am. Soc. Nephrol. 12: 1369–1382.

17. Anders, H. J., E. Belemezova, V. Eis, S. Segerer, V. Vielhauer, G. Perez deLema, M. Kretzler, C. D. Cohen, M. Frink, R. Horuk, et al. 2004. Late onsetof treatment with a chemokine receptor CCR1 antagonist prevents progres-sion of lupus nephritis in MRL-Fas(lpr) mice. J. Am. Soc. Nephrol. 15: 1504–1513.

18. Ninichuk, V., A. G. Khandoga, S. Segerer, P. Loetscher, A. Schlapbach,L. Revesz, R. Feifel, A. Khandoga, F. Krombach, P. J. Nelson, et al. 2007. Therole of interstitial macrophages in nephropathy of type 2 diabetic db/db mice.Am. J. Pathol. 170: 1267–1276.

19. Adalid-Peralta, L., A. Mathian, T. Tran, L. Delbos, I. Durand-Gasselin,D. Berrebi, M. Peuchmaur, J. Couderc, D. Emilie, and S. Koutouzov. 2008.Leukocytes and the kidney contribute to interstitial inflammation in lupus ne-phritis. Kidney Int. 73: 172–180.

20. Perry, D., A. Sang, Y. Yin, Y. Y. Zheng, and L. Morel. 2011. Murine models ofsystemic lupus erythematosus. J. Biomed. Biotechnol. 2011: 271694.

21. Cockwell, P., A. J. Howie, D. Adu, and C. O. Savage. 1998. In situ analysis ofC-C chemokine mRNA in human glomerulonephritis. Kidney Int. 54: 827–836.

22. Furuichi, K., T. Wada, N. Sakai, Y. Iwata, K. Yoshimoto, M. Shimizu,K. Kobayashi, K. Takasawa, H. Kida, S. I. Takeda, et al. 2000. Distinct ex-pression of CCR1 and CCR5 in glomerular and interstitial lesions of humanglomerular diseases. Am. J. Nephrol. 20: 291–299.

23. Vielhauer, V., E. Berning, V. Eis, M. Kretzler, S. Segerer, F. Strutz, R. Horuk,H. J. Grone, D. Schlondorff, and H. J. Anders. 2004. CCR1 blockade reducesinterstitial inflammation and fibrosis in mice with glomerulosclerosis and ne-phrotic syndrome. Kidney Int. 66: 2264–2278.

24. Owen, J. L., M. F. Criscitiello, S. Libreros, R. Garcia-Areas, K. Guthrie,M. Torroella-Kouri, and V. Iragavarapu-Charyulu. 2011. Expression of the in-flammatory chemokines CCL2, CCL5 and CXCL2 and the receptors CCR1-3and CXCR2 in T lymphocytes from mammary tumor-bearing mice. Cell.Immunol. 270: 172–182.

25. Ajuebor, M. N., C. M. Hogaboam, T. Le, A. E. Proudfoot, and M. G. Swain.2004. CCL3/MIP-1alpha is pro-inflammatory in murine T cell-mediated hepa-titis by recruiting CCR1-expressing CD4(+) T cells to the liver. Eur. J. Immunol.34: 2907–2918.

26. Schaller, M. A., L. E. Kallal, and N. W. Lukacs. 2008. A key role for CCchemokine receptor 1 in T-cell-mediated respiratory inflammation. Am. J.Pathol. 172: 386–394.

27. Anders, H. J., V. Ninichuk, and D. Schlondorff. 2006. Progression of kidneydisease: blocking leukocyte recruitment with chemokine receptor CCR1 antag-onists. Kidney Int. 69: 29–32.

28. Ramos, M. V., C. Auvynet, L. Poupel, M. Rodero, M. P. Mejias, C. A. Panek,S. Vanzulli, C. Combadiere, and M. Palermo. 2012. Chemokine receptor CCR1disruption limits renal damage in a murine model of hemolytic uremic syn-drome. Am. J. Pathol. 180: 1040–1048.

29. Gao, J. L., and P. M. Murphy. 1995. Cloning and differential tissue-specificexpression of three mouse beta chemokine receptor-like genes, including thegene for a functional macrophage inflammatory protein-1 alpha receptor. J. Biol.Chem. 270: 17494–17501.

30. Cook, D. N., M. A. Beck, T. M. Coffman, S. L. Kirby, J. F. Sheridan,I. B. Pragnell, and O. Smithies. 1995. Requirement of MIP-1 alpha for an in-flammatory response to viral infection. Science 269: 1583–1585.

31. Revesz, L., B. Bollbuck, T. Buhl, J. Dawson, R. Feifel, R. Heng, P. Hiestand,H. Sparrer, A. Schlapbach, and R. Waelchli. 2006. Bridged piperazines andpiperidines as CCR1 antagonists with oral activity in models of arthritis andmultiple sclerosis. Lett. Drug Des. Discov. 3: 689–694.

32. Balabanian, K., E. Brotin, V. Biajoux, L. Bouchet-Delbos, E. Lainey,O. Fenneteau, D. Bonnet, L. Fiette, D. Emilie, and F. Bachelerie. 2012. Properdesensitization of CXCR4 is required for lymphocyte development and periph-eral compartmentalization in mice. Blood 119: 5722–5730.

The Journal of Immunology 895

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

33. Rutledge, R. G., and C. Cote. 2003. Mathematics of quantitative kinetic PCR andthe application of standard curves. Nucleic Acids Res. 31: e93.

34. Hill, G. S., M. Delahousse, D. Nochy, and J. Bariety. 2005. Class IV-S versusclass IV-G lupus nephritis: clinical and morphologic differences suggestingdifferent pathogenesis. Kidney Int. 68: 2288–2297.

35. Balabanian, K., J. Couderc, L. Bouchet-Delbos, A. Amara, D. Berrebi,A. Foussat, F. Baleux, A. Portier, I. Durand-Gasselin, R. L. Coffman, et al. 2003.Role of the chemokine stromal cell-derived factor 1 in autoantibody productionand nephritis in murine lupus. J. Immunol. 170: 3392–3400.

36. Theofilopoulos, A. N., and F. J. Dixon. 1985. Murine models of systemic lupuserythematosus. Adv. Immunol. 37: 269–390.

37. Ramos, C. D., C. Canetti, J. T. Souto, J. S. Silva, C. M. Hogaboam,S. H. Ferreira, and F. Q. Cunha. 2005. MIP-1alpha[CCL3] acting on the CCR1receptor mediates neutrophil migration in immune inflammation via sequentialrelease of TNF-alpha and LTB4. J. Leukoc. Biol. 78: 167–177.

38. Bonecchi, R., N. Polentarutti, W. Luini, A. Borsatti, S. Bernasconi, M. Locati,C. Power, A. Proudfoot, T. N. Wells, C. Mackay, et al. 1999. Up-regulation ofCCR1 and CCR3 and induction of chemotaxis to CC chemokines by IFN-gamma in human neutrophils. J. Immunol. 162: 474–479.

39. Banas, B., B. Luckow, M. Moller, C. Klier, P. J. Nelson, E. Schadde, M. Brigl,D. Halevy, H. Holthofer, B. Reinhart, and D. Schlondorff. 1999. Chemokine andchemokine receptor expression in a novel human mesangial cell line. J. Am. Soc.Nephrol. 10: 2314–2322.

40. Mo, R., J. Chen, A. Grolleau-Julius, H. S. Murphy, B. C. Richardson, andR. L. Yung. 2005. Estrogen regulates CCR gene expression and function inT lymphocytes. J. Immunol. 174: 6023–6029.

41. Lengi, A. J., R. A. Phillips, E. Karpuzoglu, and S. A. Ahmed. 2007. Estrogenselectively regulates chemokines in murine splenocytes. J. Leukoc. Biol. 81:1065–1074.

42. Eis, V., B. Luckow, V. Vielhauer, J. T. Siveke, Y. Linde, S. Segerer, G. Perez DeLema, C. D. Cohen, M. Kretzler, M. Mack, et al. 2004. Chemokine receptorCCR1 but not CCR5 mediates leukocyte recruitment and subsequent renal fi-brosis after unilateral ureteral obstruction. J. Am. Soc. Nephrol. 15: 337–347.

43. Ninichuk, V., O. Gross, C. Reichel, A. Khandoga, R. D. Pawar, R. Ciubar,S. Segerer, E. Belemezova, E. Radomska, B. Luckow, et al. 2005. Delayedchemokine receptor 1 blockade prolongs survival in collagen 4A3-deficient micewith Alport disease. J. Am. Soc. Nephrol. 16: 977–985.

44. Ishikawa, S., T. Sato, M. Abe, S. Nagai, N. Onai, H. Yoneyama, Y. Zhang,T. Suzuki, S. Hashimoto, T. Shirai, et al. 2001. Aberrant high expression ofB lymphocyte chemokine (BLC/CXCL13) by C11b+CD11c+ dendritic cells inmurine lupus and preferential chemotaxis of B1 cells towards BLC. J. Exp. Med.193: 1393–1402.

45. Nelson, P. J., A. J. Rees, M. D. Griffin, J. Hughes, C. Kurts, and J. Duffield. 2012.The renal mononuclear phagocytic system. J. Am. Soc. Nephrol. 23: 194–203.

46. Lin, S. L., A. P. Castano, B. T. Nowlin, M. L. Lupher, Jr., and J. S. Duffield.2009. Bone marrow Ly6Chigh monocytes are selectively recruited to injuredkidney and differentiate into functionally distinct populations. J. Immunol. 183:6733–6743.

47. Turner, J. E., H. J. Paust, O. M. Steinmetz, A. Peters, C. Meyer-Schwesinger,F. Heymann, U. Helmchen, S. Fehr, R. Horuk, U. Wenzel, et al. 2008. CCR5deficiency aggravates crescentic glomerulonephritis in mice. J. Immunol. 181:6546–6556.

896 ROLE OF Ccr1 IN LUPUS NEPHRITIS

by guest on October 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from