Susceptibility to T Cell-Mediated Injury in Immune Complex Disease ...

of February 11, 2018.This information is current as

Role of NF-AT PathwayActivation of Renin-Angiotensin System: TheImmune Complex Disease Is Linked to Local Susceptibility to T Cell-Mediated Injury in

Okumura, Yasuhiko Tomino, Chisei Ra and Jesús EgidoGuillermo Sanjuán, Marta Ruiz-Ortega, Takeshi Sugaya, KoOscar López-Franco, Purificación Hernández-Vargas, Yusuke Suzuki, Carmen Gómez-Guerrero, Isao Shirato,

http://www.jimmunol.org/content/169/8/4136doi: 10.4049/jimmunol.169.8.4136

2002; 169:4136-4146; ;J Immunol 

        average*  

4 weeks from acceptance to publicationSpeedy Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

?The JIWhy

Referenceshttp://www.jimmunol.org/content/169/8/4136.full#ref-list-1

, 27 of which you can access for free at: cites 69 articlesThis article

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 All rights reserved.Copyright © 2002 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 February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 11, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Susceptibility to T Cell-Mediated Injury in Immune ComplexDisease Is Linked to Local Activation of Renin-AngiotensinSystem: The Role of NF-AT Pathway1

Yusuke Suzuki,*‡ Carmen Gomez-Guerrero,* Isao Shirato,‡ Oscar Lopez-Franco,*Purificacion Hernandez-Vargas,* Guillermo Sanjuan,* Marta Ruiz-Ortega,* Takeshi Sugaya,†

Ko Okumura, §¶ Yasuhiko Tomino,‡ Chisei Ra,§� and Jesus Egido2*

FcR provides a critical link between ligands and effector cells in immune complex diseases. Emerging evidence reveals thatangiotensin (Ang)II exerts a wide variety of cellular effects and contributes to the pathogenesis of inflammatory diseases. Inanti-glomerular basement membrane Ab-induced glomerulonephritis (GN), we have previously noted that FcR-deficient mice(��/�) surviving from lethal initial damage still developed mesangial proliferative GN, which was drastically prevented by anAngII type 1 receptor (AT1) blocker. We further examined the mechanisms by which renin-Ang system (RAS) participates in thisimmune disease. Using bone marrow chimeras between��/� and AT1�/� mice, we found that glomerular injury in ��/� mice wasassociated with CD4� T cell infiltration depending on renal AT1-stimulation. Based on findings in cutaneous delayed-type hy-persensitivity, we showed that AngII-activated renal resident cells are responsible for the recruitment of effector T cells. We nextexamined the chemotactic activity of AngII-stimulated mesangial cells, as potential mechanisms coupling RAS and cellular im-munity. Chemotactic activity for T cells and Th1-associated chemokine (IFN-�-inducible protein-10 and macrophage-inflamma-tory protein 1�) expression was markedly reduced in mesangial cells from AT1�/� mice. Moreover, this activity was mainlythrough calcineurin-dependent NF-AT. Although IFN-�-inducible protein-10 was NF-�B-dependent, macrophage-inflammatoryprotein 1� was dominantly regulated by NF-AT. Furthermore, AT1-dependent NF-AT activation was observed in injured glo-meruli by Southwestern histochemistry. In conclusion, our data indicate that local RAS activation, partly via the local NF-ATpathway, enhances the susceptibility to T cell-mediated injury in anti-glomerular basement membrane Ab-induced GN. This novelmechanism affords a rationale for the use of drugs interfering with RAS in immune renal diseases.The Journal of Immunology,2002, 169: 4136–4146.

T he inflammatory cascade mediated by the Ab/immunecomplex (IC)3 has been redefined by intensive studies us-ing targeted gene disruption of IgR (FcR) (1, 2). These

studies have clarified that FcR is a critical molecule to initiatecellular response in type II and III inflammation. Studies with

mouse strains lacking functional FcRs (��/�) also demonstratedtheir crucial roles in the acute phase of anti-glomerular basementmembrane (GBM) Ab-induced glomerulonephritis (GN) (3, 4),which is one of the most important models available for assessinginflammatory mediators in both humoral and cellular immunity(5). Although ��/� mice with anti-GBM disease were protectedfrom acute polymorphonuclear cell (PMN) influx and subsequentlethal endothelial damage observed in their wild-type (WT) litter-mates (3), they still developed glomerular injury characterized bymesangial proliferation with monocyte/macrophage accumulationin the later phase of this disease (3).

In contrast, Holdsworth and Tipping and coworker (6) have con-vincingly demonstrated that glomerular-accumulated T cells are re-sponsible for the major component of glomerular injury, mainly glo-merular crescents, independent of the presence of autologous Ab inthis disease (7). Those T cells have a Th1 phenotype (6) and requireMHC class II expression by renal resident cells (RRC) for their suf-ficient effector response (8). Moreover, the crescentic formation isassociated with mesangial proliferation and macrophage infiltration(6), indicating that glomerular delayed-type hypersensitivity (DTH)injury closely resembles the lesions observed in the ��/� mice.

Angiotensin (Ang)II is a growth factor that regulates cell pro-liferation and extracellular matrix synthesis beyond its hemody-namic effect (9–11). Some studies have also revealed that AngIIparticipates in cellular recruitment and adhesion (11, 12). Indeed,studies with Ang-converting enzyme (ACE) inhibitors or Ang type1 receptor (AT1) blocker suggest that the renin-Ang system (RAS)contributes to the pathogenesis of inflammatory diseases, such as

*Renal and Vascular Laboratory, Fundacion Jimenez Dıaz, Autonoma University,Madrid, Spain; †Discovery Research Laboratory, Tanabe Seiyaku, Osaka, Japan; ‡Di-vision of Nephrology, Department of Internal Medicine, §Atopy (Allergy) ResearchCenter, and ¶Department of Immunology, Juntendo University School of Medicine,and �Department of Molecular Cell Immunology and Allergology, Advanced MedicalResearch Center, Nihon University School of Medicine, Tokyo, Japan

Received for publication November 1, 2001. Accepted for publication August5, 2002.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by research grants from Comunidad Autonoma de Madrid(08.4/0019/2000, 08.4/0017, and 2000, 08.9/0002/2000), Fondo de Investigacion Sani-taria (99/0425, 00/0111), European Union Concerted Action (BMH4-CT98-3631, DG12-SSMI), and Spanish Society of Nephrology. Y.S. is supported by funds from Japan HealthScience Foundation and Alumni Association of Juntendo University. O.L.-F. is a fellowfrom Fondo de Investigacion Saniteria.2 Address correspondence and reprint requests to Dr. Jesus Egido, Renal and VascularLaboratory, Fundacion Jimenez Dıaz, Autonoma University, Avenida de los ReyesCatolicos, 2, 28040-Madrid, Spain. E-mail address: [email protected] Abbreviations used in this paper: IC, immune complex; GBM, glomerular basementmembrane; GN, glomerulonephritis; anti-GBM GN, glomerular basement membrane Ab-induced GN; PMN, polymorphonuclear cell; WT, wild type; RRC, renal resident cell;DTH, delayed-type hypersensitivity; Ang, angiotensin; ACE, Ang-converting enzyme;AT1, Ang type 1 receptor; RAS, renin-Ang system; NTS, nephrotoxic serum; MC, mes-angial cell; CaN, calcineurin; CsA, cyclosporin A; IP, IFN-�-inducible protein; MIP,macrophage-inflammatory protein; BMC, bone marrow-derived cell.

The Journal of Immunology

Copyright © 2002 by The American Association of Immunologists, Inc. 0022-1767/02/$02.00

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

immune-mediated GN and allograft rejection (13, 14). However,the mechanisms of the beneficial effects of RAS blockade in thosediseases are still unclear.

Although the precise mechanisms remain undefined, previousstudies have demonstrated that Ab deposition onto GBM stronglyactivates the intrarenal (15) and systemic RAS (16, 17), inducinghemodynamic changes in a dose-dependent manner. Surprisingly,the AT1 blocker drastically attenuated glomerular injury and ac-cumulation of macrophages in ��/� GN (3). Furthermore, the ac-tivated RAS may participate early in the pathogenesis of this dis-ease. Those findings highlighted a certain role of RAS in immunerenal injury.

We hypothesized that RAS activation plays an essential role inthe susceptibility of local cellular immune response. In the kidneyand lymphocytes, AngII exerts its biological effects mainly viaAT1 (18). In rodents, AT1 exists in two isoforms, AT1A andAT1B, regulated by two different genes. The murine AT1A is theisoform predominantly expressed in most tissues (19). AngII viaAT1A triggers the proliferation of splenic lymphocytes followingsystemic cellular immune responses in mice (20). A recent studyhas shown that renal AT1 may play a role in the progression of thisimmune disease (21). However, the implication of AngII on T cellrecruitment in the kidney as well as the molecular mechanismsinvolved is not yet defined. To circumvent the potential contribu-tion of FcR in this process, some particular experiments were de-signed. For analyzing relevant cell types (resident or inflammatorycells) and the RAS system (local or systemic), we generated bonemarrow chimeras between ��/� and AT1A-deficient (AT1�/�)mice and induced anti-GBM GN. Based on the in vivo findings, wealso examined the potential mechanisms coupling RAS activationand the cellular immune response, such as chemokine expressionand T cell recruitment. In addition, we further studied intracellularevents involved in cell signaling with special attention to transcrip-tional factors as mediators of the AngII-induced inflammatory pro-cess (11, 12), including NF-�B (22, 23) and NF-AT (24, 25). Ourpresent findings show a novel mechanism in the pathogenesis of ICdisease and propose the potential therapeutical interest of RASblockade in immune renal diseases.

Materials and MethodsMice

FcR �-chain-deficient (��/�) and AT1AR-deficient (AT1�/�) mice weregenerated by a homologous recombination method. Construction of target-ing vectors and generation of these knockout mice have been previouslydescribed in detail (3, 4, 16). In all in vivo experiments, we used femaleanimals weighing 18–23 g. Both ��/� and AT1�/� mice have a C57BL/6background. Although WT littermates of each strain (��/�, AT1�/�) andC57BL/6 mice were analyzed in all experiments, no significant differenceswere found in the kinetics of proteinuria and glomerular/interstitium dam-age, as noted in our previous studies (3, 16). Thus, results of C57BL/6 micematched for age were shown as representative controls of WT.

Generation of the bone marrow chimeras

Bone marrow transplantation (BMT) was performed to 6- to 8-wk-old fe-male ��/�, AT1�/�, and WT mice. Bone marrows were collected fromeach mouse strain and treated with Gay’s solution to exclude RBC con-tamination for protection against vascular (thrombotic) injury, and thenwere transplanted i.v. (3–5 � 107 bone marrow cells) to mice which hadbeen irradiated at 800 rad x-ray with renal protection by lead plates. Be-cause all mice had a C57BL/6 background, there were no symptoms ofgraft-vs-host-disease in any of them. For an additional 4 wk, transplantedanimals were kept in air-conditioned clean cages. We generated three dif-ferent bone marrow chimeras as follows: �A (bone marrow from ��/� toirradiated AT1�/�), �W (��/� to WT), and WA (WT to AT1�/�). In apilot study, WW (WT to WT), �� (��/� to ��/�), and AA (AT1�/� toAT1�/�) mice were also generated as controls.

Genotype exchange in peripheral blood of each bone marrow chimera(�A and �W) was determined by PCR with purified genomic DNA from

peripheral blood before and 5 wk after BMT (QIAamp Blood kit; Qiagen,Hilden, Germany). Primers used for murine FcR �-chain were as follows:specific primer for exon 3 (5�-GGAATTCGCTGCCTTTCGGACCTGGAT-3�) and exon 2 (5�-GGAATTCGATGCTGTCCTGTTTTTGTA-3�)and for the created neo �-chain (4) of which exon 2 was replaced (5�-GCCAACGCTATGTCCTGATAG-3�). PCR was simultaneously performedwith these primers under the following conditions: 94°C for 1 min, 57°C for1 min, and 72°C for 1.5 min with 33 cycles. After the confirmation of genotypeexchange, these chimeras were subjected to the experiments.

Experimental protocol for anti-GBM GN

The method for preparation of nephrotoxic serum (NTS) was previouslydescribed (3). In the present study, we used batches of NTS different fromthose of our previous study (3) to avoid the possibility that RAS-relatedinjury in ��/� mice was dependent on the batch. Anti-GBM GN was in-duced by i.v. injection of NTS through the tail vein in mice which werepreimmunized with rabbit IgG and IFA (Difco, Detroit, MI) 4 days beforethe administration of NTS, and followed until day 150. Because a prelim-inary study showed that NTS at a dose of 20 �l/20 g body weight wassufficient to cause proteinuria and severe renal damage in WT mice, weused this dose of NTS (1� NTS) for general experiments and a 3-foldhigher dose (60 �l/20 g) for the excess NTS model (3� NTS model) in��/� mice and their chimeras. No mice developed anaphylactic symptomsafter the injection of NTS.

Urinary protein was determined at days 1, 3, 5, 7, and 10 and once aweek after day 14 until day 50, and every 10 days after day 50 by Knight’smethod, as previously described (3). To be sure of the disease kinetics inacute phase (before day 50), we also checked their spontaneous urine pro-duction when we moved them for cleaning cages and therefore confirmedthat urinary protein, depicted in figures, well-represented the outcome ofthe disease. Kidneys were perfused with cold saline and removed undergeneral anesthesia. For the evaluation of the effect of RAS blockade, ��/�

mice were treated with the AT1 blocker valsartan (Novartis Pharmaceuti-cals, Tokyo, Japan; 10 mg/kg/day orally) 24 h before the injection of NTS.

Renal histopathological studies

Kidney sections, fixed in 10% formaldehyde, were stained with periodicacid Schiff’s reagent in 4-�m-thick sections to assess histological changesby light microscopy. Frozen renal sections were used for immunofluores-cence for rabbit and murine IgG, C3, CD4� T cells, and then stained withFITC-labeled Abs (ICN Pharmaceuticals, Frankfurt, Germany; DAKO,Barcelona, Spain; and BD PharMingen, San Diego, CA). Mesangial pro-liferation was evaluated by the numbers of mesangial cells (MC) in oneglomerular tuft (score 0, 0–2 cells; 1, 3–4; 2, 5–6; 3, 7–8; 4, �8). Glo-merular endothelial damage was scored by the percentage of fibrin depo-sition occupancy in one glomerulus (score 0, 0%; 1, 0–25; 2, 25–50; 3,50–75; 4, �75). At least 25 glomeruli of one animal and five animals ofeach group were examined. The mean scores of each group were expressedin Table I as follows: 0–1, (�); 1–2, (�); 2–3, (��); 3–4, (���). Forthe evaluation of CD4� T cells, at least 30 glomeruli per section wereexamined using a blinded protocol as previously described (26). The resultswere expressed as cells per glomerular section.

Preparation of murine MC and AngII stimulation

Murine MC (WT and AT1�/�) were cultured from isolated glomeruli byseveral sieving techniques and different centrifugation as previously de-scribed (23), and maintained in RPMI 1640 medium (Life Technologies,Grand Island, NY) containing 10% FCS, 1 mM L-glutamine, and 100�g/ml penicillin/streptomycin. MC were characterized by phase contrastmicroscopy and immunohistochemistry (positive staining for desmin andvimentin, and negative staining for keratin and factor VIII Ag) (23, 27).Confluent cells between the first and third passages were used for assays.

After a 48-h starvation, both WT and AT1�/� MC were stimulated withAngII 10�6 M in serum-free medium for 3, 6, 12, and 24 h. Supernatantsfrom stimulated MC were collected for chemotaxis assays. For inhibitionassays, MC were preincubated with a NF-�B inhibitor, 10 �M parthenolide(Sigma-Aldrich, Madrid, Spain) for 1.5 h (28), or with calcineurin (CaN)/NF-AT inhibitors, 1 �M cyclosporin A (CsA; Sigma-Aldrich), or 10 �MCaN autoinhibitory peptide 457–482 (29) (Calbiochem, Darmstadt, Ger-many) for 2 h. After exposure to these inhibitors, the culture medium wasremoved and cells were washed with serum-free medium and subjected tothe experiments.

4137The Journal of Immunology

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Chemotaxis assays

The chemotactic activity of MC supernatants was evaluated in 24-wellTranswell chemotaxis chambers (Costar, High Wycombe, U.K.) as previ-ously described (30). The lower wells were loaded in triplicate with 600 �lof the supernatants and covered with a 5-�m pore-size polycarbonatemembrane. Upper compartments were loaded with 100 �l of the cell sus-pension containing 5 � 105 T cells (Jurkat cell; ATCC TIB-152; AmericanType Culture Collection, Manassas, VA). The chambers were incubated at37°C for 4 h to assess chemotaxis of T cells. Migrating cells in the lowercompartment were counted by flow cytometry. Specific chemotaxis datarepresent the fold-increase of the average number of migrated cells witheach MC supernatant vs the stimulation medium alone (serum-free mediumwith AngII 10�6 M).

RNA extraction and mRNA expression analyses

Total mesangial RNA was obtained by the TRIzol method (Life Technol-ogies). One microgram RNA from stimulated MC was reverse-transcribedand then amplified with a commercial kit (Promega, Buckinghamshire,U.K.), with the use of 0.5 �Ci [�-32P]dCTP (3000 Ci/mmol, Amersham,Arlington Heights, IL) and 20 pmol specific primers for mouse IFN-�-inducible protein (IP)-10 (sense, 5�-CAACCCAAGTGCTGCC-3�; anti-sense, 5�-GGGAATTCACCATGCTTGACCA-3�; fragment, 475 bp, ref.AF 227743) (31), mouse macrophage-inflammatory protein (MIP) 1�(sense, 5�-GCTGTCCTCC-TCTGCACCAT-3�; antisense, 5�-CTGCCGGCTTCGCTTGGTTA-3�; fragment, 189 bp, ref. NM 002983) (32), andmouse GAPDH (sense, 5�-CCGGTGCTGAGTATGTAGTG-3�; antisense,5�-CAGTCTTCTGAGTGGCAGTG-3�; fragment, 289 bp, ref. AK013857). The amplifications were conducted with annealing temperaturesof 61°C (IP-10), 62°C (MIP1�), or 59°C (GAPDH). The optimum numberof amplification cycles used for semiquantitative RT-PCR (30, 32, and 25,respectively) was chosen on the basis of pilot experiments (data notshown). In some cases, PCR products of IP-10 and MIP1� were purifiedfrom low-melting temperature agarose gel, radiolabeled with RandomPrimed DNA Labeling kit (Roche, Indianapolis, IN), and used as cDNAprobes for hybridization in Northern blot analysis. The expression ofGAPDH was used as internal control. Aliquots of each reaction were runon 4% acrylamide-bisacrylamide gels. The gels were dried and exposed toX-OMAT AS films (Eastman Kodak, Madrid, Spain). Autoradiogramswere quantified by the Image Quant scanning densitometry (MolecularDynamics, Sunnyvale, CA). The density of each gene was compared afterthe individual correction by density of GAPDH.

Twenty-five micrograms of denatured RNA were electrophoresed andtransferred to nylon membranes (Genescreen; New England Nuclear, Bos-ton, MA). The membranes were prehybridized for 6 h at 42°C in 50%formamide, 1% SDS, 5� SSC, 1� Denhardt’s solution, 0.1 mg/ml dena-tured salmon sperm DNA, and 50 mM PBS, pH 6.5. The hybridization wasperformed at 55°C for 20 h with 10% dextran sulfate and 1 � 106 cpm/mlof labeled denatured cDNA probe. Membranes were washed, autoradio-graphed, and films were scanned using the scanning densitometry (Molec-ular Dynamics). Relative amounts of mRNA were established in relation to28S rRNA.

Extraction of nuclear proteins and EMSA

Nuclear extracts were obtained as previously described (33) and the activ-ity of transcription factors was evaluated by EMSA. Briefly, frozen kidneypieces were pulverized in a metallic chamber and resuspended in a coldextraction buffer (20 mM HEPES-NaOH (pH 7.6), 20% (v/v) glycerol, 0.35

M NaCl, 5 mM MgCl2, 0.1 mM EDTA, 1 mM DTT, 0.5 mM PMSF, 1�g/ml pepstatin A). The homogenate was vigorously shaken, and the in-soluble materials were precipitated by centrifugation at 12,000 rpm for 30min at 4°C. Supernatants were dialyzed overnight against a binding buffercontaining 20 mM HEPES-NaOH (pH 7.6), 20% (v/v) glycerol, 0.1 mMNaCl, 5 mM MgCl2, 0.1 mM EDTA, 1 mM DTT, and 0.5 mM PMSF.These dialysates were cleared by centrifugation at 10,000 � g for 15 minat 4°C and stored in aliquots at �80°C until use. Protein concentration wasquantified by the bicinchoninic acid method (Pierce, Rockford, IL).

NF-AT consensus oligonucleotides (5�-CGCCCAAAGAGGAAAATTTGTTTCATA-3�) (Santa Cruz Biotechnology, Santa Cruz, CA) were[32P]-end-labeled by incubation for 10 min at 37°C with 10 U T4 polynu-cleotide kinase (Promega) in a reaction containing 10 �Ci [�-32P]ATP(3000 Ci/mmol; Amersham), 70 mM Tris-HCl, 10 mM MgCl2, and 5 mMDTT. The reaction was stopped by the addition of EDTA to a final con-centration of 0.05 M. Nuclear proteins (10 �g) were equilibrated for 10min in a binding buffer containing 4% glycerol, 1 mM MgCl2, 0.5 mMEDTA, 0.5 mM DTT, 50 mM NaCl, 10 mM Tris-HCl (pH 7.5), and 2 �gpoly(dI-dC) for a 20-�l final volume. When competition and supershiftassays were performed, the cold probe and Abs (anti-NF-ATc4; 0.5�g)(Santa Cruz Biotechnology) were added to this buffer 30 and 60 minbefore the addition of the labeled probe. Labeled probe (0.035 pmol) wasadded to the reaction and incubated for 30 min at room temperature. Thereaction was stopped by the addition of gel loading buffer (250 mM Tris-HCl, 0.2% bromophenol blue, 0.2% xylene cyanol, and 40% glycerol) andrun on a nondenaturing, 4% acrylamide gel at 100 V at room temperaturein 89 mM Tris-borate, 2 mM EDTA (pH 8.0; TBE) (22).

Southwestern histochemistry

This technique was developed to detect the in situ distribution and DNA-binding activity of transcriptional factors (34). NF-AT consensus oligonu-cleotide was digoxigenin-labeled with a 3�-terminal transferase (Boeh-ringer Mannheim, Mannheim, Germany). Paraffin-embedded tissuesections were fixed in 0.5% paraformaldehyde and incubated with 0.1mg/ml DNase I. The DNA binding reaction was performed by incubationwith 50 pmol of the labeled DNA probe in buffer containing 0.25% BSAand 1 �g/ml poly(dI-dC). The sections were then incubated with alkalinephosphatase-conjugated anti-digoxigenin Ab, and colorimetric detectionwas performed as described. Preparations without probe were used as neg-ative controls, and mutant-labeled probe and excess of unlabeled probewere used to test the specificity of the technique.

Assessment of cutaneous DTH

Mice were immunized i.p. with Ag (250 �g goat Ig) emulsified in CFA (8,35). After 7 days, immunized mice were challenged with the same Ag (250�g) in the hind footpad. For both induction of anti-GBM GN and DTH in��/� mice, those mice were immunized with rabbit IgG 3 days after DTHpreimmunization and were injected with 3-fold higher NTS at the sametime of the Ag challenge. DTH responsiveness was determined 24-h post-challenge by measuring the dorsal-ventral thickness difference of the Ag-injected left footpad and the saline-injected right footpad, as a control,using a micrometer (Mitutoyo, Kanagawa, Japan).

Table I. Receptor phenotypes and glomerular lesions in each mouse straina

Phenotype of FcR and AT1 Phenotype of Glomerular Lesion

FcR AT1 1� NTS 3� NTS

BMC RRC BMC RRC end.damage mes.prolif mes.prolif Crescents

WT � � � � ��� � nd nd��/� � � � � � � ��� ��AT1�/� � � � � ��� � nd nd�A � � � � � � � ��W � � � � � � ��� ��WA � � � � ��� � nd nd

a 3� NTS, 3-fold higher amounts of NTS; nd, not done, because most of them died; end.damage, glomerular endothelial damage; mes.prolif, mesangial proliferation;crescents, glomerular crescents.

4138 ROLE OF LOCAL RAS IN T CELL-MEDIATED IC DISEASE

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Statistical analysis

Results are expressed as mean � SD and were analyzed by ANOVA (seeFig. 1) and Mann-Whitney test (see Figs. 3–5) for comparison of quanti-tative variables. Statistical significance was established as p � 0.05 (two-tailed curve).

ResultsFcR on bone marrow-derived cells (BMCs) plays a crucial rolein acute glomerular injury in anti-GBM disease

We have previously demonstrated that FcR and AT1 are criticalmolecules in the induction of this disease (3). However, the rele-vant cell types expressing these receptors remain unknown. There-fore, to clarify this feature, we generated bone marrow chimerasbetween mice strains lacking each receptor. Their receptor pheno-types (FcR and AT1) of BMCs and RRCs are summarized in TableI. We also generated control mice which were transplanted withbone marrow from the same mice strain (WW, ��, and AA) undersame irradiation conditions, and induced this disease. We couldnot find any significant difference in their disease phenotypes (uri-nary protein and renal pathology) from mice without transplanta-tion (WT, ��/�, and AT1�/�, respectively; data not shown), in-dicating that bone marrows in recipient animals were functionallyreconstituted by transplantation and the irradiation condition maynot elicit significant alteration of this disease.

Next, we analyzed the evolution of anti-GBM GN in these an-imals. During the acute phase of the disease, WT, AT1�/�, andWA mice showed severe proteinuria peaking at day 7 (Fig. 1A)with glomerular endothelial damage associated with fibrin deposits(Fig. 2A, a–c) (Table I). However, proteinuria peak in AT1�/� andWA mice was significantly less than in WT ( p � 0.05) mice. AllWT and most AT1�/� (62%) and WA (67%) mice died with mas-sive ascites before day 35, while ��/�, �W, and �A mice werecompletely protected from proteinuria (Fig. 1A) and endothelialdamage (Fig. 2A, d–f). These data further confirm a critical im-plication of FcR on BMCs in the acute glomerular damage of thisdisease.

AT1-stimulation induces glomerular injury associated withCD4� T cell infiltration in ��/� mice

In contrast to that shown above, ��/� and �W mice injected withhigher amounts of NTS (3� NTS) developed moderate proteinuria(Fig. 1B) and glomerular injury characterized by mesangial pro-liferation, cellular infiltration, and glomerular enlargement withoccasional crescents (Fig. 2A, g and h) (Table I). Glomerular injuryin ��/� mice was associated with CD4� T cell infiltration in adose-dependent manner (Table II). Interestingly, �A mice weredrastically protected from proteinuria (Fig. 1) and glomerular in-jury (Fig. 2Ai) with absence of T cell infiltration (Table II) in the3� NTS model, even though the heterologous (rabbit IgG), autol-ogous (mouse IgG) Ab, and C3 depositions in �A mice were sim-ilarly noted in ��/� (Fig. 2B) or WT (data not shown) mice. Thesedata indicate that tissue AT1 is responsible for T cell-associatedglomerular injury in this disease.

Cutaneous DTH response is not attenuated in AT1�/� or ��/�

mice with anti-GBM GN

To investigate whether FcR or AT1 deficiency may affect systemiccell-mediated immune responses, we induced cutaneous DTH, aclassical T cell-dependent inflammatory lesion. No difference inDTH responsiveness was noted in WT, ��/�, and AT1�/� mice(Table III). This finding is consistent with the data of the autolo-gous IgG deposition (Fig. 2B), and suggests that cutaneous DTHresponse is independent of FcR and AT1.

In certain conditions, AngII participates in the regulation of sys-temic cellular immune response (20). Therefore, to investigatewhether the systemic RAS activation in anti-GBM GN (16, 17) issufficient to nonspecifically enhance the systemic cellular immuneresponse, we simultaneously induced anti-GBM GN (3� NTSmodel) and cutaneous DTH response in ��/� mice. We failed tofind any difference in systemic DTH response in ��/� mice withor without GN (Table III), suggesting that systemic RAS activationin this disease may have no significant role in systemic T cellfunction.

AngII enhances the chemotactic activity for T cells and themRNA expression of Th1-associated chemokines in MCthrough AT1

Based on the above-mentioned in vivo findings from bonemarrow chimeras and systemic DTH responses, we next postu-lated that glomerular T cell infiltration may be regulated byRRCs activated by AngII. We especially focused on glomerular

FIGURE 1. FcR on BMCs and tissue AT1 are critical for the inductionof anti-GBM GN. Renal injury in mice was assessed by urinary proteinexcretion until day 150. Acute glomerular damage peaking at day 7 wasdependent on FcR on BMCs (A). Although higher amounts of NTS (3�NTS) were required, mice strains lacking FcR on BMCs also showed glo-merular damage in a later phase peaking at day 14, which was dependenton tissue AT1 (B). �, p � 0.01, ��, p � 0.05, (A) vs ��/�, (B) vs �A.

4139The Journal of Immunology

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

MC, because they possess both AT1 and AT2, regulate theglomerular blood flow, and release proinflammatory cytokinesin response to AngII (11, 12, 33). In addition, mesangialproliferation is abolished in �A mice with 3� NTS. In thisstudy, we examined AngII-induced chemotactic activity for Tcells, as a possible mechanism involved in the recruitment ofthose cells. In the chemotactic assays with T cells, supernatantsfrom WT MC treated with AngII had significantly higheractivity (�3.5-fold) than treatment medium alone, reaching aplateau after 12 h (Fig. 3A). By contrast, supernatants fromAngII-stimulated AT1�/� MC showed significantly less activ-ity (around basal at 12 h) (Fig. 3A), indicating that the chemo-tactic activity induced by AngII in MC occurred mainly through

AT1. We also noted that AngII by itself presented a lowchemotactic activity for T cells (average of migration in controlmedium vs medium with AngII; 354 � 90 vs 743 � 92 cells),consistently with a previous study (36).

Recent data have convincingly demonstrated that nephrito-genic T cells associated with crescent GN in this disease aremainly Th1 cells (6). Functional diversity between Th1 and Th2is partly due to the difference the chemokine receptor pheno-types (37, 38). CXCR3 and CCR5 are preferentially expressedin Th1 cells (37). Therefore, we also studied the regulation oftheir corresponding ligands (CXCR3, IP-10; CCR5, MIP1�) inMC stimulated by AngII. As noted in Fig. 3B, AngII (10�6 M)significantly up-regulated the mRNA expression of IP-10 and

FIGURE 2. FcR on BMCs and tissueAT1 are responsible for distinct glomerulardamages. A, Severe endothelial damage withfibrin deposits in acute phase was observedin mice strains having FcR on BMCs (at day7, a, WT; b, AT1�/�; c, WA), but not inmice strains lacking them (d, ��/�; e, �W; f,�A). However, higher amounts of NTS (3�NTS) induced mesangial proliferative GNdepending on tissue AT1 (at day 14, g, ��/�;h, �W; i, �A). B, Although �A did notpresent morphological lesions, no obviousdifferences in the deposition of heterologousIgG (rabbit IgG) and autologous IgG and C3(mouse IgG/C3) were noted between ��/�

and �A mice. (Original magnification ineach panel, �100).

4140 ROLE OF LOCAL RAS IN T CELL-MEDIATED IC DISEASE

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

MIP1� in WT MC, peaking at 6 h, as determined by semiquan-titative RT-PCR. These data were confirmed by Northern blotanalyses. As shown in Fig. 3C, AngII induced mRNA expres-sion of both chemokines in MC with similar kinetics. By bothmethods (Fig. 3, B and C), AT1�/� MC showed significantlyless mRNA expression in IP-10 and MIP1� than WT MC,indicating that the expressions of both chemokine genes aremainly elicited through AT1 stimulation.

AngII-induced chemotactic activity involves CaN/NF-AT andNF-�B pathways

Emerging data reveal that the AngII/NF-�B pathway contrib-utes to the pathogenesis of inflammatory diseases via regulationof chemokine production (11). In contrast, although CaN/NF-AT pathways were firstly reported in T cells (39), theirimportance has been recently highlighted in other organs, suchas the heart, vascular system, neurons, and muscles (24, 25,40 – 43). Special attention has been paid to the AngII/NF-ATpathway in the pathogenesis of certain diseases (24, 25). Theactivity of NF-AT proteins is tightly regulated by the calcium/calmodulin-dependent phosphatase CaN (39). Recent studiessuggested the implication of the CaN-mediated activation ofNF-AT in chemokine production (44 – 46). Therefore, to ap-proach possible transcriptional regulations in this mechanism,we pretreated MC with inhibitors of NF-�B (parthenolide) andCaN/NF-AT (CsA and CaN autoinhibitory peptide), and thenwe analyzed the T cell chemotaxis. Surprisingly, both CaNinhibitors showed marked attenuation in WT MC (CsA, 97%inhibition at 6 h, 72% at 12 h; CaN autoinhibitory peptide, 98%at 12 h) (Fig. 4A), even though CsA itself slightly inducedchemotactic activation to MC (around 1.2- to 1.4-fold increasevs medium alone at 6 and 12 h). By contrast, parthenolideshowed only 24 and 15% inhibition at 6 and 12 h, respectively(Fig. 4A), suggesting that the CaN-dependent pathway plays apredominant role in the AngII-induced chemotaxis by MC.

Next, we examined the implication of both pathways in Th1-associated chemokine production. AngII-induced IP-10 mRNAwas markedly attenuated by pretreatment with parthenolide (75%inhibition at 6 h, 85% at 12 h), but not with CsA (24% at 6 h, 4%at 12 h) (Fig. 4B). In contrast, MIP1� mRNA expression wasinhibited around 40–50% by CsA at 6 and 12 h. The data areconsistent with previous studies that showed the presence of func-

tional NF-AT sites in the MIP1� promoter-enhancer region (47).Accordingly, these data suggest that AngII did enhance mRNAexpression of both Th1-associated chemokines (IP-10 andMIP1�), mainly via AT1 on MC, though their transcriptional reg-ulation may be different.

��/� mice with anti-GBM GN show renal NF-AT activationwhich is attenuated by an AT1 blocker

To confirm the implication of the renal CaN/NF-AT pathway inthis disease, we performed EMSA with nuclear proteins from the

Table III. DTH responsiveness of each mouse

Group n

DTH Responsivenessa

Ag challenge Control saline

WT 3 29.3 � 4.9 8.3 � 5.9AT1�/� 3 27.7 � 3.5 6.3 � 4.2��/� 3 26.6 � 2.1 4.0 � 2.0��/� � nephritis 3 27.3 � 4.9 6.7 � 2.1

a DTH responsiveness was determined 24-h postchallenge by measuring the in-crease of footpad size (�0.01 mm).

Table II. Glomerular T cell infiltration

Group n CD4� Cells (c/gcs)

��/� 1� NTS 4 0.44 � 0.11��/� 3� NTS 4 1.24 � 0.37a

�A 3� NTS 4 0.36 � 0.16

a p � 0.05, vs ��/� 1� NTS or �A 3� NTS; c/gcs: cells/30 glomerulicross-sections.

FIGURE 3. AngII enhances the chemotactic activity for T cells and themRNA expression of Th1-associated chemokines in MC through AT1.Supernatants from AngII (10�6 M) -stimulated WT MC (E) showed sig-nificantly higher chemotactic activity for T cells than those from AT1�/�

MC (F) (A). In WT MC stimulated with AngII (10�6M) (�), mRNAexpression of Th1-associated chemokines (IP-10 and MIP1�) was higherthan those in AT1�/� MC (f), as determined by RT-PCR (B) and Northernblotting (C). Data are presented as the mean � SD (n 4–5 experiments).�, p � 0.05 (B), �, p � 0.01 (C) vs AT1�/� MC.

4141The Journal of Immunology

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

renal cortex. WT mice showed an early peak of NF-AT activationat 3 h after the injection of the Ab, and reactivation at 24 h (Fig.5). ��/� mice showed basically the same kinetics of NF-AT ac-tivation. There was no significant difference in the peak amplitudeof NF-AT activation in WT and ��/� mice. Preincubation with ananti-NF-ATc4 Ab attenuated the NF-AT peak signals, indicatingthat activated renal NF-AT in this disease involves NF-ATc4(NF-AT3).

To clarify the relevant cell types of this activation, Southwesternhistochemistry with NF-AT oligo probes was done in ��/� mice.In the acute phase of this disease, NF-AT activation was observedin glomeruli, mainly at MC (Fig. 6, upper panels). Interestingly, inthe chronic phase of this disease, ��/� mice showed activationsignals not only in glomeruli, but also in tubuli and interstitialinfiltrating cells (Fig. 6, lower panels).

Next, we investigated whether RAS blockade may affect theNF-AT activation in the acute phase of this disease. Surprisingly,treatment with valsartan, an AT1 blocker, drastically attenuated

the NF-AT activation at 3 and 24 h in ��/� mice (Fig. 5), con-sistently with the decrement of glomerular activation (Fig. 6, upperpanels). These data suggest that AngII-induced NF-AT activationin RRCs may contribute to the initiation of this disease. Further-more, the NF-AT pathway in resident and infiltrating cells couldalso be involved in the development of this disease.

DiscussionFcR on BMCs plays a crucial role in acute glomerular injury inanti-GBM disease

The role of activating FcR in providing a critical link betweenligands and effector cells in Ab/IC-mediated inflammation hasbeen well-established (1, 2), but the significance of these receptorson each effector cell type along the disease still remains unclear. Inthe present study, even though �W and �A mice have FcR onRRCs, acute lethal damage observed in WT and AT1�/� mice was

FIGURE 4. AngII-induced chemotacticactivity and chemokine expressions in MCinvolve CaN/NF-AT and NF-�B pathways.A, Chemotactic activity for T cells in AngII-induced WT MC (10�6 M, at 6 and 12 h, Fig.3A), was markedly attenuated by preincuba-tion with CaN/NF-AT inhibitors, CsA andCaN autoinhibitory peptide (CaN inhibitoryP.), whereas NF-�B inhibitors (parthenolide)had less effect. AT1�/� MC also showed asimilar response at 6 h. �, p � 0.05 vs with-out inhibitors. (M, control medium; TM, me-dium with AngII 10�6 M). B, Th1-associatedchemokines were transcriptionally regulatedin a different manner. Thus, mRNA expres-sion of IP-10 in WT MC was drastically at-tenuated by pretreatment with parthenolide,but not with CsA. In contrast, CsA attenu-ated MIP1� mRNA expression more mark-edly than parthenolide. Data are presented asthe mean � SD (n 3–4 experiments).

4142 ROLE OF LOCAL RAS IN T CELL-MEDIATED IC DISEASE

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

completely abolished in these chimeric mice. These findings fur-ther confirm the critical implication of FcR in the acute phase ofanti-GBM GN (3) and indicate that FcRs, especially on BMCs, areessentially required for an initial inflammatory response after Abdeposition. Consistently, the injury can be induced in W� chimeras(BMCs; WT, RRCs; ��/�) (48). Imasawa et al. (49) recently havesuggested that BMCs may have the potential to differentiate into

glomerular resident cells. However, in contrast, Mayadas and co-workers demonstrated that Fc�RIII on PMN is essentially requiredfor initial recruitment of PMN in anti-GBM GN (50) and, follow-ing interaction between Fc�R and CD11b/CD18 (Mac 1) on PMN,is also necessary for sufficient PMN spreading on the glomerularcapillary wall (51). In fact, the absence of acute glomerular dam-age in ��/� mice was associated with the lack of PMN influx (3).Although we need to examine the contribution of bone marrow-derived glomerular resident cells in the acute inflammatory set-tings, our present data further support the idea that Fc�R on PMN,but not on RRCs, may play a major role for acute endothelialdamage.

Persistent proliferative GN in ��/� mice is closely linked toglomerular-infiltrating CD4� T cells

Interestingly, ��/� mice developed GN persisting for �5 mo andits severity was dependent on the amount of Ab injected and thenumber of glomerular-infiltrating CD4� T cells. In addition, themorphological lesions are highly analogous to those seen in studiesdemonstrating T cell-dependent injury of this disease (6). Thesedata indicate that CD4� T cell (Th1)-dependent response is pivotalfor the development of mesangial proliferative GN in ��/� mice.This hypothesis is further supported by recent findings of ourgroup using mouse strain overexpressing Smad7 (an inhibitorymolecule of TGF-� signaling) (26), in which the CD4� T cellscannot migrate into the inflammatory sites due to the disregulationof CD62 ligand (L-selectin) expression. In anti-GBM disease, thedevelopment of GN, including macrophage infiltration, in theseanimals was drastically attenuated, suggesting that the develop-ment and the persistence of this disease essentially require glo-merular-infiltrating CD4� T cells.

Renal RAS activation conducts glomerular T cell response

T cell-dependent injury in ��/� and �W mice required three timeshigher amounts of anti-GBM Ab than FcR-mediated endothelialinjury, indicating different thresholds for their activation by thesame Ab. However, �A chimeras were protected from the glomer-ular T cell response even in the high dose model, emphasizing thatAngII action via AT1 on recipients could be responsible for thethreshold of the T cell-mediated mechanism. Besides, different Tcell responses between �W and �A chimeras indicate that their

FIGURE 5. ��/� mice with anti-GBM GN exhibit NF-AT activationthat is attenuated by an AT1 blocker. EMSA of nuclear proteins from ��/�

renal cortex showed NF-AT activation with two peaks at 3 and 24 h. Thiskinetics was similar to that of WT. Furthermore, the activation peaks in��/� mice were significantly attenuated by valsartan, an AT1 blocker(AT1A) (�, p � 0.01, ��, p � 0.05). Lane C denotes coincubation with asample of ��/� mice at 3 h and NF-AT cold oligos. Data are presented asthe mean � SD (n 4–6 experiments).

FIGURE 6. NF-AT activation inRRCs is implicated in the pathogenesisof tissue injury in anti-GBM GN.Southwestern histochemistry with NF-AT-specific oligo probes revealedNF-AT activation in glomerular resi-dent cells of ��/� mice at 3 h, whichwas drastically attenuated by valsartan,an AT1 blocker (upper panels). Inter-estingly, in the chronic phase of thisdisease, this activation was detectednot only in glomerular resident cells,but also in the tubular and infiltratingcells (lower panels). Arrows and openarrowheads denote representative acti-vated nuclei of RRCs (glomerular andtubular) and infiltrating cells, respec-tively. (Original magnification in eachpanel, �100)

4143The Journal of Immunology

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

bone marrow-derived glomerular resident cells (49) (presumablyFcR� but AT1�) may not play an important role for T cell re-cruitment. In this disease, dose-dependent activation of intrarenaland systemic RAS has been demonstrated (15–17). AngII hassome cellular effects on most tissues, mainly via AT1, that maycontribute to the disease pathogenesis (11, 19), and also regulatescellular immunity by acting on the proliferation of splenic lym-phocytes (20). However, in a cutaneous DTH, AT1 deficiency didnot alter the responsiveness, in accordance with a previous study(21), and we failed to find any difference between ��/� mice withor without nephritis, suggesting that systemic RAS activation inthis disease may not play a significant role in general T cell func-tion. Accordingly, the amplitude of intrarenal RAS activationwould dominantly conduct the glomerular DTH response in thisdisease, though we must carefully elucidate the alteration of AT1expression on T cells by elevated plasma AngII.

AngII enhances Th1-associated chemokine expression in MC,mainly via AT1

It is already known that AngII itself is chemotactic for T cells (36).Besides confirming this feature, we noted that supernatants of MCtreated with AngII elicited a marked chemotactic activity for Tcells, indicating a predominant role of second mediators (presum-ably chemokines) induced by AngII. The experiments withAT1�/� MC revealed that those AngII actions were exertedmainly via AT1. These findings are consistent with previous stud-ies showing that AngII, acting through both AT1 and AT2, inducesT cell-chemokine production (12, 33, 52).

Enhanced expressions of IP-10, MIP1�, and their receptors inkidney have been previously shown in this disease (53), as well asin human mesangial proliferative GN (e.g., IgA nephropathy) (54).The present study shows that the expression of these Th1-associ-ated chemokines in MC is up-regulated by AngII mainly throughAT1. MIP1� redundantly cross-reacts with CCR5 and CCR1 aswell as other chemokines (53), whereas IP-10 is more selective toCXCR3 (37). Moreover, CCR1 is expressed equally in Th1 andTh2, while CCR5 is not (37, 38). In this regard, interestingly,CCR1-deficient mice with anti-GBM GN showed enhanced Th1response and glomerular crescents, where not only both chemo-kines, but also CXCR3 and CCR5, were up-regulated in associa-tion with higher CD4� T cell and macrophage infiltration (53).This evidence suggests that Th1-deviated immune response of thisdisease may be partially enhanced by chemokine phenotypes pro-duced by RRCs.

Importantly, once T cells and subsequent macrophages are, evenif nonspecifically, recruited into the inflamed kidney by local RASactivation, they may orchestrate the autocrine/paracrine-accelera-tion loop accompanied by locally elevated AngII because they areequipped with all RAS components (55). In fact, significantsources of tissue ACE in human atherosclerotic plaques are re-gions of clustered macrophages (56). In addition, IL-12, a keycytokine for Th1 response, from mononuclear cells is suppressedby ACE inhibitors (57). In this regard, the role of immunocompe-tent cells in nonimmune renal diseases further supports this notion(55). Salt-sensitive hypertension after AngII infusion was associ-ated with tubulointerstitial accumulation of AngII-producing lym-phocytes and was prevented by the immunosuppressor mycophe-nolate mofetil coincidentally with a reduction of those cells (58).

AngII exerts proinflammatory effects in the kidney, partlythrough the CaN-dependent NF-AT pathway

Elevated local AngII in this disease may result from physiologicalresponses to the alterations elicited by the specific Ab deposition.Therefore, MC could be one of the major targets of the AngII

effect. Indeed, as a consequence of mesangial contraction byAngII, a significant decrease in glomerular plasma flow and singlenephron glomerular filtration rate, followed by increased renal vas-cular resistance, was observed in this model in a dose-dependentmanner (15, 17). Consequently, one can postulate that excessivelyelevated AngII may elicit increased intracellular calcium levels inMC and subsequently a wide variety of cellular responses by aCa2�-dependent pathway. Some parts of RAS influence on immu-nological function may be due to such indirect outcome (20, 36,59). In this sense, it is noteworthy that the chemotactic activity forT cells in AngII-treated MC was largely attenuated by CaN-spe-cific inhibitors in the present study. Although NF-AT3 (NF-ATc4)mRNA was previously detected in the kidney (60) and endothelin1 activates cyclooxygenase 2 expression via NF-AT in culturedMC (61), there are still no studies demonstrating the functional orpathological contribution of NF-AT during kidney disease. NF-ATactivity requires the sustained Ca2� stimulus provided by the Ca2�

release-activated Ca2� influx channel and Ca-dependent phosphataseCaN (39, 43). Therefore, there is considerable evidence that the Ca2�

release-activated Ca2� influx in MC is under the control of both pro-tein kinase C and calmodulin, and thus represents a key mechanismfor the control of Ca2�-regulated mesangial function (62).

Because a study with synthetic peptides blocking NF-AT acti-vation by CaN postulates CsA-sensitive (presumably CaN-depen-dent) gene expressions that are not controlled by NF-AT (47), wemust carefully elucidate the mesangial CaN/NF-AT pathway withAngII stimulation in future studies. However, our present data inEMSA and Southwestern histochemistry strongly support the no-tion that AT1-stimulated NF-AT activation may be involved in thepathogenesis of this immune-mediated disease, as it occurs inmyocardial hypertrophy (24). It is interesting to note that CsAcould directly prevent mesangial proliferative GN independentlyof its immunosuppressive action (63). AngII regulates cellular im-mune responses through the CaN-dependent pathway within thelymphoid tissue (20). The present data show for the first time theactivation of the local CaN/NF-AT pathway early in this renaldisease, and its attenuation by valsartan, an AT1 blocker, suggest-ing that locally elevated AngII, probably together with inflamma-tory cytokines (25), may regulate cellular immune responses partlyvia the local CaN/NF-AT pathway. Furthermore, the distributionof the activated NF-AT, changing from glomeruli to tubulo-inter-stitium and infiltrating cells along the disease course, suggests itsimplication in the different stages of the disease and supports theidea of the potential interest in targeting this pathway (24, 47).

Although we and others have already reported the contributionof AngII/NF-�B (11, 33) and IC/NF-�B pathways in the patho-genesis of GN (27, 64) through the chemokine release, NF-�Binhibitors had less effects on the chemotactic activity. Our dataindicate that IP-10 expression was mainly regulated by the AngII/NF-�B pathway, while AngII-enhanced MIP1� expression wasmainly through CaN/NF-AT. Interestingly, although MIP 1� canbe active to resting T cells (65), IP-10 mainly influences on acti-vated T cells (66). Tight regulation of the chemokine receptor ex-pression in T cells can be the reason for the difference (38, 67).Because this cell clone behaves more like “naive” than “activated”T cells (68, 69), it could be one of the reasons why the goodinhibitor of IP-10, parthenolide, had less effects on chemotaxis. Infact, CaN/NF-AT inhibitors showed marked attenuation of chemo-taxis, suggesting that AngII-induced NF-AT activation may pref-erentially contribute to the chemotaxis of inactivated T cells. How-ever, in vivo T cell chemotaxis may be regulated in a morecomplicated manner (38). Because chemokine/chemokine receptorinteraction, for example, contributes to position effector T cells(38, 67), the infiltrating T cell population would shift from more

4144 ROLE OF LOCAL RAS IN T CELL-MEDIATED IC DISEASE

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

like “naive” to “activated” cells along the disease course, as wellit occurs in other immune diseases. Indeed, acute activation ofNF-AT, peaking at 3 h, and delayed NF-�B activation at 24 h in��/� mice (our unpublished data) may support this idea. There-fore, our present data indicate that AngII-activated transcriptionalfactors and subsequent chemotactic mediators, including MIP1�and IP-10, play roles in a process of multistep navigation to T cellsin this disease.

Conclusion

In conclusion, the current studies provide evidence for AngII-de-pendent CD4� T cell-directed injury in ��/� mice with anti-GBMGN. In addition, they demonstrate that local RAS activation, viaAT1, facilitates glomerular T cell recruitment by Th1 chemokinerelease from RRCs activated with AngII. They also provide thefirst demonstration of the potential implication of the local AngII-dependent NF-AT pathway in the pathogenesis of IC nephritis.Finally, these results afford a rational basis for the use of RASantagonists in patients with renal immune diseases.

AcknowledgmentsWe thank Drs. M. V. Alvarez Arroyo and Y. Hisada for helpful discussion,Drs. K. Hattori and Y. Kanamaru for their help with BMT, Dr. Juan JoseGranizo Martınez for helping us with statistical analyses, T. Shibata forexcellent technical assistance, and Lise Lotte Gulliksen for secretarialassistance.

References1. Ravetch, J. V. 1994. Fc receptors: rubor redux. Cell 78:553.2. Ravetch, J. V., and S. Bolland. 2001. IgG Fc receptors. Annu. Rev. Immunol.

19:275.3. Suzuki, Y., I. Shirato, K. Okumura, J. V. Ravetch, T. Takai, Y. Tomino, and

C. Ra. 1998. Distinct contribution of Fc receptors and angiotensin II-dependentpathways in anti-GBM glomerulonephritis. Kidney Int. 54:1166.

4. Park, S. Y., S. Ueda, H. Ohno, Y. Hamano, M. Tanaka, T. Shiratori, T. Yamazaki,H. Arase, N. Arase, A. Karasawa, et al. 1998. Resistance of Fc receptor-deficientmice to fatal glomerulonephritis. J. Clin. Invest. 102:1229.

5. Wilson, C. B. 1996. Renal response to immunologic glomerular injury. In TheKidney, Vol. 1. B. M. Brenner, ed. W. B. Saunders, Philadelphia, p. 1253.

6. Kitching, A. R., S. R. Holdsworth, and P. G. Tipping. 2000. Crescentic glomer-ulonephritis: a manifestation of a nephritogenic Th1 response? Histol. His-topathol. 15:993.

7. Li, S., S. R. Holdsworth, and P. G. Tipping. 1997. Antibody independent cres-centic glomerulonephritis in � chain deficient mice. Kidney Int. 51:672.

8. Li, S., C. Kurts, F. Kontgen, S. R. Holdsworth, and P. G. Tipping. 1998. Majorhistocompatibility complex class II expression by intrinsic renal cells is requiredfor crescentic glomerulonephritis. J. Exp. Med. 188:597.

9. Egido, J. 1996. Vasoactive hormones and renal sclerosis. Kidney Int. 49:578.10. Ichikawa, I., and R. C. Harris. 1991. Angiotensin actions in the kidney: renewed

insight into the old hormone. Kidney Int. 40:583.11. Ruiz-Ortega, M., O. Lorenzo, Y. Suzuki, M. Ruperez, and J. Egido. 2001. Proin-

flammatory actions of angiotensins. Curr. Opin. Nephrol. Hypertens. 10:321.12. Suzuki, Y., M. Ruiz-Ortega, and J. Egido. 2000. Angiotensin II: a double-edged

sword in inflammation. J. Nephrol. 13(Suppl. 3):S101.13. The GISEN Group. 1997. Randomised placebo-controlled trial of effect of

ramipril on decline in glomerular filtration rate and risk of terminal renal failurein proteinuric, non-diabetic nephropathy. Lancet 349:1857.

14. Stigant, C. E., J. Cohen, M. Vivera, and J. S. Zaltzman. 2000. ACE inhibitors andangiotensin II antagonists in renal transplantation: an analysis of safety and ef-ficacy. Am. J. Kidney Dis. 35:154.

15. Boyce, N. W., and S. R. Holdsworth. 1987. Intrarenal hemodynamic alterationsinduced by anti-GBM antibody. Kidney Int. 31:8.

16. Hisada, Y., T. Sugaya, M. Yamanouchi, H. Uchida, H. Fujimura, H. Sakurai,A. Fukamizu, and K. Murakami. 1999. Angiotensin II plays a pathogenic role inimmune-mediated renal injury in mice. J. Clin. Invest. 103:627.

17. Timmermans, V., P. W. Peake, J. A. Charlesworth, G. J. Macdonald, andM. A. Pawlak. 1990. Angiotensin II receptor regulation in anti-glomerular base-ment membrane nephritis. Kidney Int. 38:518.

18. Zhuo, J., K. Song, P. J. Harris, and F. A. Mendelsohn. 1992. In vitro autoradiog-raphy reveals predominantly AT1 angiotensin II receptors in rat kidney. Ren.Physiol. Biochem. 15:231.

19. Llorens-Cortes, C., B. Greenberg, H. Huang, and P. Corvol. 1994. Tissular ex-pression and regulation of type 1 angiotensin II receptor subtypes by quantitativereverse transcriptase-polymerase chain reaction analysis. Hypertension 24:538.

20. Nataraj, C., M. I. Oliverio, R. B. Mannon, P. J. Mannon, L. P. Audoly,C. S. Amuchastegui, P. Ruiz, O. Smithies, and T. M. Coffman. 1999. AngiotensinII regulates cellular immune responses through a calcineurin-dependent pathway.J. Clin. Invest. 104:1693.

21. Hisada, Y., T. Sugaya, S. Tanaka, Y. Suzuki, C. Ra, K. Kimura, andA. Fukamizu. 2001. An essential role of angiotensin II receptor type 1a in re-cipient kidney, not in transplanted peripheral blood leukocytes, in progressiveimmune-mediated renal injury. Lab. Invest. 81:1243.

22. Hernandez-Presa, M., C. Bustos, M. Ortego, J. Tunon, G. Renedo,M. Ruiz-Ortega, and J. Egido. 1997. Angiotensin-converting enzyme inhibitionprevents arterial nuclear factor-�B activation, monocyte chemoattractant pro-tein-1 expression, and macrophage infiltration in a rabbit model of early accel-erated atherosclerosis. Circulation 95:1532.

23. Ruiz-Ortega, M., C. Bustos, M. A. Hernandez-Presa, O. Lorenzo, J. J. Plaza, andJ. Egido. 1998. Angiotensin II participates in mononuclear cell recruitment inexperimental immune complex nephritis through nuclear factor-�B activation andmonocyte chemoattractant protein-1 synthesis. J. Immunol. 161:430.

24. Molkentin, J. D., J. R. Lu, C. L. Antos, B. Markham, J. Richardson, J. Robbins,S. R. Grant, and E. N. Olson. 1998. A calcineurin-dependent transcriptional path-way for cardiac hypertrophy. Cell 93:215.

25. Boss, V., K. L. Abbott, X. F. Wang, G. K. Pavlath, and T. J. Murphy. 1998. Thecyclosporin A-sensitive nuclear factor of activated T cells (NFAT) proteins areexpressed in vascular smooth muscle cells: differential localization of NFATisoforms and induction of NFAT-mediated transcription by phospholipase C-coupled cell surface receptors. J. Biol. Chem. 273:19664.

26. Kanamaru, Y., A. Nakao, M. Mamura, Y. Suzuki, I. Shirato, K. Okumura,Y. Tomino, and C. Ra. 2001. Blockade of TGF-� signaling in T cells prevents thedevelopment of experimental glomerulonephritis. J. Immunol. 166:2818.

27. Duque, N., C. Gomez-Guerrero, and J. Egido. 1997. Interaction of IgA with Fc�receptors of human mesangial cells activates transcription factor nuclear fac-tor-�B and induces expression and synthesis of monocyte chemoattractant pro-tein-1, IL-8, and IFN-inducible protein 10. J. Immunol. 159:3474.

28. Hehner, S. P., T. G. Hofmann, W. Droge, and M. L. Schmitz. 1999. The anti-inflammatory sesquiterpene lactone parthenolide inhibits NF-�B by targeting theI�B kinase complex. J. Immunol. 163:5617.

29. Perrino, B. A., L. Y. Ng, and T. R. Soderling. 1995. Calcium regulation ofcalcineurin phosphatase activity by its B subunit and calmodulin: role of theautoinhibitory domain. J. Biol. Chem. 270:7012.

30. Falk, W., R. H. Goodwin Jr., and J. Leonard. 1980. A 48-well micro chemotaxisassembly for rapid and accurate measurement of leukocyte migration. J. Immu-nol. Methods. 33:239.

31. Glabinski, A. R., M. Tani, R. M. Strieter, V. K. Tuohy, and R. M. Ransohoff.1997. Synchronous synthesis of �- and �-chemokines by cells of diverse lineagein the central nervous system of mice with relapses of chronic experimentalautoimmune encephalomyelitis. Am. J. Pathol. 150:617.

32. Stafford, S., H. Li, P. A. Forsythe, M. Ryan, R. Bravo, and R. Alam. 1997.Monocyte chemotactic protein-3 (MCP-3)/fibroblast-induced cytokine (FIC) ineosinophilic inflammation of the airways and the inhibitory effects of an anti-MCP-3/FIC antibody. J. Immunol. 158:4953.

33. Ruiz-Ortega, M., O. Lorenzo, M. Ruperez, J. Blanco, and J. Egido. 2001. An-giotensin II activates nuclear factor-�B via AT1 and AT2 receptors in the kidney.Am. J. Pathol. 158:1743.

34. Hernandez-Presa, M. A., C. Gomez-Guerrero, and J. Egido. 1999. In situ non-radioactive detection of nuclear factors in paraffin sections by Southwestern his-tochemistry. Kidney Int. 55:209.

35. Cua, D. J., D. R. Hinton, L. Kirkman, and S. A. Stohlman. 1995. Macrophagesregulate induction of delayed-type hypersensitivity and experimental allergic en-cephalomyelitis in SJL mice. Eur. J. Immunol. 25:2318.

36. Weinstock, J. V., and J. Kassab. 1986. Chemotactic response of splenic mono-nuclear cells to angiotensin II in murine schistosomiasis. J. Immunol. 137:2020.

37. Bonecchi, R., G. Bianchi, P. P. Bordignon, D. D’Ambrosio, R. Lang, A. Borsatti,S. Sozzani, P. Allavena, P. A. Gray, A. Mantovani, and F. Sinigaglia. 1998.Differential expression of chemokine receptors and chemotactic responsivenessof type 1 T helper cells (Th1s) and Th2s. J. Exp. Med. 187:129.

38. Kim, C. H., L. Rott, E. J. Kunkel, M. C. Genovese, D. P. Andrew, L. Wu, andE. C. Butcher. 2001. Rules of chemokine receptor association with T cell polar-ization in vivo. J. Immunol. 108:1331.

39. Rao, A., C. Luo, and P. G. Hogan. 1997. Transcription factors of the NFATfamily: regulation and function. Annu. Rev. Immunol. 15:707.

40. Ranger, A. M., M. J. Grusby, M. R. Hodge, E. M. Gravallese, F. C. de la Brousse,T. Hoey, C. Mickanin, H. S. Baldwin, and L. H. Glimcher. 1998. The transcrip-tion factor NF-ATc is essential for cardiac valve formation. Nature 392:186.

41. Graef, I. A., P. G. Mermelstein, K. Stankunas, J. R. Neilson, K. Deisseroth,R. W. Tsien, and G. R. Crabtree. 1999. L-type calcium channels and GSK-3regulate the activity of NF-ATc4 in hippocampal neurons. Nature 401:703.

42. Olson, E. N., and R. S. Williams. 2000. Remodeling muscles with calcineurin.BioEssays 22:510.

43. Crabtree, G. R. 1999. Generic signals and specific outcomes: signaling throughCa2�, calcineurin, and NF-AT. Cell 96:611.

44. Ali, H., J. C. Ahamed Hernandez-Munain, J. L. Baron, M. S. Krangel, andD. D. Patel. 2000. Chemokine production by G protein-coupled receptor activa-tion in a human mast cell line: roles of extracellular signal-regulated kinase andNFAT. J. Immunol. 165:7215.

45. Staruch, M. J., R. Camacho, and F. J. Dumont. 1998. Distinctive calcineurin-dependent (FK506-sensitive) mechanisms regulate the production of the CC che-mokines macrophage inflammatory protein (MIP)-1�, MIP-1�, and RANTES vsIL-2 and TNF-� by activated human T cells. Cell. Immunol. 190:121.

46. Kuhns, D. B., H. A. Young, E. K. Gallin, and J. I. Gallin. 1998. Ca2�-dependentproduction and release of IL-8 in human neutrophils. J. Immunol. 161:4332.

4145The Journal of Immunology

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

47. Aramburu, J., M. B. Yaffe, C. Lopez-Rodriguez, L. C. Cantley, P. G. Hogan, andA. Rao. 1999. Affinity-driven peptide selection of an NFAT inhibitor more se-lective than cyclosporin A. Science 285:2129.

48. Suzuki, Y., S. Kodera, I. Shirato, Y. Kanamaru, C. Ra, and Y. Tomino. 1999.Quantitative threshold of Fc receptor expression may affect the severity of glo-merulonephritis. J. Am. Soc. Nephrol. 10:522.

49. Imasawa, T., Y. Utsunomiya, T. Kawamura, Y. Zhong, R. Nagasawa, M. Okabe,N. Maruyama, T. Hosoya, and T. Ohno. 2001. The potential of bone marrow-derived cells to differentiate to glomerular mesangial cells. J. Am. Soc. Nephrol.12:1401.

50. Coxon, A., X. Cullere, S. Knight, S. Sethi, M. W. Wakelin, G. Stavrakis,F. W. Luscinskas, and T. N. Mayadas. 2001. Fc�RIII mediates neutrophil re-cruitment to immune complexes: a mechanism for neutrophil accumulation inimmune-mediated inflammation. Immunity 14:693.

51. Tang, T., A. Rosenkranz, K. J. Assmann, M. J. Goodman, J. C. Gutierrez-Ramos,M. C. Carroll, R. S. Cotran, and T. N. Mayadas. 1997. A role for Mac-1 (CD11b/CD18) in immune complex-stimulated neutrophil function in vivo: Mac-1 defi-ciency abrogates sustained Fc� receptor-dependent neutrophil adhesion and com-plement-dependent proteinuria in acute glomerulonephritis. J. Exp. Med. 186:1853.

52. Wolf, G., F. N. Ziyadeh, F. Thaiss, J. Tomaszewski, R. J. Caron, U. Wenzel,G. Zahner, U. Helmchen, and R. A. Stahl. 1997. Angiotensin II stimulates ex-pression of the chemokine RANTES in rat glomerular endothelial cells: role ofthe angiotensin type 2 receptor. J. Clin. Invest. 100:1047.

53. Topham, P. S., V. Csizmadia, D. Soler, D. Hines, C. J. Gerard, D. J. Salant, andW. W. Hancock. 1999. Lack of chemokine receptor CCR1 enhances Th1 re-sponses and glomerular injury during nephrotoxic nephritis. J. Clin. Invest. 104:1549.

54. Romagnani, P., C. Beltrame, F. Annunziato, L. Lasagni, M. Luconi, G. Galli,L. Cosmi, E. Maggi, M. Salvadori, C. Pupilli, and M. Serio. 1999. Role forinteractions between IP-10/Mig and CXCR3 in proliferative glomerulonephritis.J. Am. Soc. Nephrol. 10:2518.

55. Rodriguez-Iturbe, B., H. Pons, and J. Herrera-Acosta, R. J. Johnson. 2001. Roleof immunocompetent cells in nonimmune renal diseases. Kidney Int. 59:1626.

56. Diet, F., R. E. Pratt, G. J. Berry, N. Momose, G. H. Gibbons, and V. J. Dzau.1996. Increased accumulation of tissue ACE in human atherosclerotic coronaryartery disease. Circulation 94:2756.

57. Constantinescu, C. S., D. B. Goodman, and E. S. Ventura. 1998. Captopril andlisinopril suppress production of interleukin-12 by human peripheral blood mono-nuclear cells. Immunol. Lett. 62:25.

58. Rodriguez-Iturbe, B., H. Pons, Y. Quiroz, K. Gordon, J. Rincon, M. Chavez,G. Parra, J. Herrera-Acosta, D. Gomez-Garre, R. Largo, et al. 2001. Mycophe-nolate mofetil prevents salt-sensitive hypertension resulting from angiotensin IIexposure. Kidney Int. 59:2222.

59. Foris, G., B. Dezso, G. A. Medgyesi, and G. Fust. 1983. Effect of angiotensin IIon macrophage functions. Immunology 48:529.

60. Hoey, T., Y. L. Sun, K. Williamson, and X. Xu. 1995. Isolation of two newmembers of the NF-AT gene family and functional characterization of the NF-ATproteins. Immunity 2:461.

61. Sugimoto, T., M. Haneda, H. Sawano, K. Isshiki, S. Maeda, D. Koya, K. Inoki,H. Yasuda, A. Kashiwagi, and R. Kikkawa. 2001. Endothelin-1 induces cyclo-oxygenase-2 expression via nuclear factor of activated T-cell transcription factorin glomerular mesangial cells. J. Am. Soc. Nephrol. 12:1359.

62. Mene, P., F. Pugliese, and G. A. Cinotti. 1996. Regulation of capacitative calciuminflux in cultured human mesangial cells: roles of protein kinase C and calmod-ulin. J. Am. Soc. Nephrol. 7:983.

63. Ren, K., J. B. Van Liew, and B. Noble. 1993. The effect of cyclosporin A ondisease progression in proliferative immune complex glomerulonephritis. Clin.Immunol. Immunopathol. 66:107.

64. Gomez-Guerrero, C., N. Duque, M. T. Casado, C. Pastor, J. Blanco, F. Mampaso,F. Vivanco, and J. Egido. 2000. Administration of IgG Fc fragments preventsglomerular injury in experimental immune complex nephritis. J. Immunol. 164:2092.

65. Avalos, B. R., K. J. Bartynski, P. J. Elder, M. S. Kotur, W. G. Burton, andN. M. Wilkie. 1994. The active monomeric form of macrophage inflammatoryprotein-1 � interacts with high- and low-affinity classes of receptors on humanhematopoietic cells. Blood 84:1790.

66. Loetscher M, B. Gerber, P. Loetscher, S. A. Jones, L. Piali, I. Clark-Lewis,M. Baggiolini, and B. Moser. 1996. Chemokine receptor specific for IP10 andmig: structure, function, and expression in activated T-lymphocytes. J. Exp. Med.184:963.

67. Mackay, C. R. 1996. Chemokine receptors and T cell chemotaxis. J. Exp. Med.184:799.

68. Gillis, S., and J. Watson. 1980. Biochemical and biological characterization oflymphocyte regulatory molecules. V. Identification of an interleukin 2-producinghuman leukemia T cell line. J. Exp. Med. 152:1709.

69. Kasibhatla, S., L. Genestier, and D. R. Green. 1999. Regulation of fas-ligandexpression during activation-induced cell death in T lymphocytes via nuclearfactor �B. J. Biol. Chem. 274:987.

4146 ROLE OF LOCAL RAS IN T CELL-MEDIATED IC DISEASE

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from