12 pages

Angiotensin II stimulates expression of the chemokine RANTES in rat glomerular endothelial cells. Role of the angiotensin type 2 receptor

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Angiotensin II stimulates expression of the chemokine RANTES in rat glomerular endothelial cells. Role of the angiotensin type 2 receptor
    ANG II and RANTES  1047  J. Clin. Invest.© The American Society for Clinical Investigation, Inc.0021-9738/97/09/1047/12$2.00Volume 100, Number 5, September 1997, 1047–1058  Angiotensin II Stimulates Expression of the Chemokine RANTES in Rat Glomerular Endothelial Cells  Role of the Angiotensin Type 2 Receptor   Gunter Wolf,* Fuad N. Ziyadeh,   §   Friedrich Thaiss,* John Tomaszewski,      Robert J. Caron,      Ulrich Wenzel,* Gunther Zahner,* Udo Helmchen,   ‡   and Rolf A.K. Stahl*   *   Department of Medicine, Division of Nephrology and Osteology; ‡   Department of Pathology, University of Hamburg, D-20246 Hamburg, Germany; the §   Renal-Electrolyte and Hypertension Division and the Penn Center for Molecular Studies of Kidney Diseases, and    Department of Pathology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104-6144   Abstract   Glomerular influx of monocytes/macrophages (M/M) occursin many immune- and non-immune-mediated renal dis-eases. The mechanisms targeting M/M into the glomerulusare incompletely understood, but may involve stimulatedexpression of chemokines. We investigated whether angio-tensin II (ANG II) induces the chemokine RANTES in cul-tured glomerular endothelial cells of the rat and in vivo.ANG II stimulated mRNA and protein expression of RANTES in cultured glomerular endothelial cells. TheANG II–induced RANTES protein was chemotactic forhuman monocytes. Surprisingly, the ANG II–stimulatedRANTES expression was transduced by AT   2   receptorsbecause the AT   2   receptor antagonists PD 123177 and CGP-42112A, but not an AT   1   receptor blocker, abolished the in-duced RANTES synthesis. Intraperitoneal infusion of ANG II(500 ng/h) into naive rats for 4 d significantly stimulatedglomerular RANTES mRNA and protein expression com-pared with solvent-infused controls. Immunohistochemistryrevealed induction of RANTES protein mainly in glomeru-lar endothelial cells and small capillaries. Moreover, ANG II–infused animals exhibited an increase in glomerular ED-1–positive cells compared with controls. Oral treatment withPD 123177 (50 mg/liter drinking water) attenuated theglomerular M/M influx without normalizing the slightly ele-vated systolic blood pressure caused by ANG II infusion,suggesting that the effects on blood pressure and RANTESinduction can be separated. We conclude that the vasoac-tive peptide ANG II may play an important role in glomer-ular chemotaxis of M/M through local induction of thechemokine RANTES. The observation that the ANG II–   mediated induction of RANTES is transduced by AT   2   re-ceptors may influence the decision as to which substancesmight be used for the therapeutic interference with the ac-tivity of the renin–angiotensin system. (   J. Clin. Invest.   1997.100:1047–1058.) Key words: angiotensin II • RANTES •   glomerular endothelial cells • macrophages/monocytes   Introduction  Glomerular infiltration with macrophages/monocytes (M/M)   1  is a common feature in many immune- and non-immune-medi-ated glomerular diseases of the kidney (1–5). It is believed thatthe glomerular influx of M/M is crucial in the progression of renal disease towards the irreversible structural changes of glomerulosclerosis (3, 6). Glomerular M/M are locally acti-vated mononuclear cells that produce an array of cytokines,growth factors, reactive oxygen species, proteases, eicosanoids,coagulation products, and nitric oxide, which may all inducetissue injury and stimulate resident glomerular cells to synthe-size extracellular matrix proteins (6, 7). Thus, it is of consider-able interest to identify the mechanism of glomerular M/M re-cruitment. Although several factors, including interleukin-1,TNF-    , platelet activating factor, leukotrienes, and comple-ment components are all chemotactic for M/M, a novel class of proinflammatory chemoattractant cytokines called chemo-kines, has been identified and characterized in the past fewyears (8–10). Interleukin-8 and monocyte chemoattractantprotein-1 (MCP-1) are probably the most thoroughly studiedmembers of this superfamily (11, 12). RANTES is a memberof the C-C-chemokine subfamily with chemoattractant proper-ties for M/M, eosinophil and basophil granulocytes, and forT lymphocytes (13, 14). TNF-    and LPS stimulate RANTESexpression in the kidney as well as in cultured renal proximaltubular and mesangial cells (15, 16). It remains unclear, how-ever, whether other cell types in the kidney may expressRANTES.We have had a long interest in the nonhemodynamic mech-anisms of how angiotensin II (ANG II) may be involved in theprogression of chronic renal disease (17). The observation thatangiotensin-converting enzyme (ACE) inhibitors prevent therenal influx of M/M in some models of chronic renal diseaseprompted us to study potential effects of ANG II on RANTESexpression in cultured glomerular endothelial cells (18–20).  Part of this study was presented at the 29th meeting of the AmericanSociety of Nephrology (November 3–6, 1996), and was published inabstract form (   J. Am. Soc. Nephrol.  7:1725, 1996).Address correspondence to Gunter Wolf, M.D., University of Hamburg, University Hospital Eppendorf, Department of Medicine,Division of Nephrology and Osteology, Pavillion 61, Martinistra    e52, D-20246 Hamburg, Germany. Phone: 49-40-4717-3667; FAX: 49-40-4717-5186.  Received for publication 20 March 1997 and accepted in revised form 20 May 1997.  1.  Abbreviations used in this paper:  ACE, angiotensin-converting en-zyme; ANG II, angiotensin II; GER, glomerular endothelial cells;MC, mesangial cells; MCP-1, monocyte chemoattractant protein-1;M/M, macrophages/monocytes; PRA, plasma renin activity.     1048  Wolf et al.  This cell type was selected since a transendothelial gradient of soluble chemokines is most likely responsible for the glomeru-lar influx of M/M (21). Moreover, haptotaxis in response toimmobilized chemoattractant molecules on the endothelialsurface may be important in this process (21). Our studiesdemonstrate that ANG II stimulates RANTES protein andmRNA expression in cultured glomerular endothelial cells iso-lated from rats (GER). Furthermore, the secreted RANTES ischemotactic for M/M as assessed in a Boyden chamber. The in-duction of RANTES by ANG II is surprisingly mediated byAT   2  receptors. Short-term ANG II infusion for 4 d in rats withosmotic minipumps stimulates RANTES protein and mRNAexpression in isolated glomeruli, but not in other renal struc-tures. These data demonstrate for the first time that the vaso-active peptide ANG II may have immunomodulatory proper-ties in the kidney through the induction of the chemokineRANTES.   Methods  Cell culture.  GER are a nontransformed glomerular endothelial cellline isolated from adult Sprague-Dawley rats. These cells have beenpreviously characterized in detail, and exhibit positive staining for theendothelial markers Factor VIII, CD 31, endothelial leukocyte adhe-sion molecule-1, and the lectin BS1 (22). Cells were routinely carriedin Dulbecco’s modified Eagle’s medium (DMEM with 450 mg/dl glu-cose; GIBCO BRL, Gaithersburg, MD) supplemented with 100 U/mlpenicillin, 100   g/ml streptomycin, 2 mM glutamine, and 10% heat-inactivated FCS at 37    C in 5% CO  2  . GER were passaged twice aweek by light trypsinization. All experiments were performed at pas-sage 15–25. For some experiments, primary culture of rat mesangialcells (MC) were used. MC were isolated from outgrowth of isolatedglomeruli obtained from Sprague-Dawley rats by differential sievingas previously described (23). MC stained positive for the Thy-1 anti-gen, but failed to bind the anti-Factor VIII antibody. MC showedcontractions after treatment with ANG II, indicating expression of functional ANG II receptors. MC were used at passage 5–8.   ANG II receptor expression.  We have previously shown that GERexpress high-affinity receptors for ANG II (22). The present receptorbinding studies were performed to characterize the expression of ANG II-receptor subtypes. Binding studies were performed on 5   10  4  cells grown to subconfluence in 24-well culture plates (Nunc Inc., Na-perville, IL) in assay buffer consisting of 150 mM NaCl, 5 mM MgCl  2  ,50 mM Tris-HCl (pH 7.1), 5 mM EDTA, 0.7% bovine serum albu-min, 0.5% aprotinin, and 1 mM phenylmethylsulfonyl fluoride on ashaking platform at 22    C. For displacement experiments, 0.5 pM[  125  I][Sar  1  , Ile  8  ]ANG II (2,000 Ci/mmol, Amersham Buchler GmbH,Braunschweig, Germany) was incubated in the presence or absenceof various concentrations of the AT  1  receptor antagonist losartan(gift of Merck, Sharp & Dohme, Munich, Germany) and the AT  2  blocker PD123177 (gift of Parke-Davis, Warner Lambert Co., AnnArbor, MI). After an incubation period of 2 h, cells were washedthree times with ice-cold PBS to remove unbound [  125  I][Sar  1  , Ile  8  ]-ANG II. The cells were dissolved in 1 N NaOH, and the amount of radioactivity was counted in a gamma scintillation counter. Nonspe-cific binding was determined in the presence of 10     3  M nonradioac-tive [Sar  1  , Ile 8  ]ANG II (Sigma Chemical Co., St. Louis, MO) and wasless than 10% of total binding. Specific binding of [  125  I][Sar  1  , Ile  8  ]-ANG II in the absence of competitors was considered 100%. Experi-ments were repeated four times with duplicate measurements foreach experiment.To gain insight into whether GER express mRNA for AT  2  recep-tors, cDNA amplification after reverse transcription was performedbecause abundance of transcripts was too low by Northern hybridiza-tion (data not shown). Total RNA was isolated by repeated phenol–chloroform extractions and isopropanol precipitation as described byChomczynski and Sacchi (24). A total of 10   g RNA was reversetranscribed using 0.7   g of poly-d(T)primer (Pharmacia DiagnosticsAB, Uppsala, Sweden) in the presence of 500 U of Maloney murineleukemia virus reverse transcriptase diluted in 50   l of a buffer con-taining 50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl  2  , 10 mMDTT, and 500   M dNTP (25). After incubation for 90 min at 37    C,the reaction was precipitated with 25   l 7.5 M ammonium acetate and50   l isopropanol, and pellets were recovered by centrifugation. Af-ter washing in 70% ethanol, pellets were resuspended in 50   l dis-tilled water. For the polymerase chain reaction, 5   l of cDNA wascombined with a total of 0.15   g of each of the following primers spe-cific for AT  2  receptor: sense 5    GGGGATGGAGCGAGCACA-GAATTG3    , antisense 5    AGTCTATCTATAAGAGTAATAGG3    (26). Amplification reactions were performed with the GeneAmp™kit (Perkin Elmer Cetus, Ueberlingen, Germany) using 200   M of each dNTP, 2.5 U of Amplitaq polymerase in 100   l PCR buffer(20 mM Tris–HCl, pH 8.3; 25 mM KCl, 2.0 mM MgCl  2  , and 0.05%Tween 20). A total of 40 cycles was performed with an annealing tem-perature of 60    C for 1.5 min, an extension step at 72    C for 1.5 min,and a denaturation step at 92    C for 1 min (25, 27). 15   l of the reac-tion product was separated in a 1.9% agarose gel containing 0.5   g/mlethidium bromide. The correct identity of the amplification productwas tested by sequencing after ligation into the TA II vector (Invitro-gen Corp., San Diego, CA) and comparison with the published se-quence. As an additional control, cDNA amplification was per-formed after reverse transcription of total RNA from isolatedglomeruli of Sprague-Dawley rats.  RANTES mRNA expression.  A total of 10  7  GER or MC weremade quiescent in DMEM without serum for 24 h, and were subse-quently incubated for another 24 h with 10     8  –10     6  M ANG II (SigmaChemical Co.). For time-course experiments, 10     7  M ANG II wasgiven for 6–24 h. ANG II was free of measurable endotoxin concen-trations (endotoxin   0.001 ng/ml, endotoxin kit from Sigma Chemi-cal Co.). Additional cells were treated with ANG II and either 10     6  Mlosartan or PD 123177. Some cells were also incubated with 1   g/mlLPS from Escherichia coli  , serotype 02:B6 (endotoxin   10,000 U/mgLPS; Sigma Chemical Co.). At the end of the stimulation period, celllayers were washed twice with RNAse-free PBS, and total RNA wasisolated as described above. The quantity and purity of the prepara-tions were assessed by measuring absorption at 260 and 280 nm. Totalcellular RNA (20   g/lane) was denatured in formamide-formalde-hyde, and was electrophoresed through a 1.2% agarose gel containing2.2 M formaldehyde (27). After completion of electrophoresis, RNAwas vacuum-blotted onto a nylon membrane (Zetabind; Cuno Inc.,Meriden, CT), and short-wave UV cross-linked. All hybridizationsteps were carried out in a rotating drum in a temperature-controlledoven. The Rapidhyb™ system (Amersham Buchler GmbH) was usedas recommended by the manufacturer. An EcoRI-XhoI insert fromthe murine RANTES clone pMuR3 (16) and a 2.0-kb cDNA insert of the plasmid pMC1 encoding the murine 18S ribosomal RNA bandwere radioactively labeled using hexamer primers with 5   Ci[  32  P]deoxyadenosine 5    -triphosphate (3,000 Ci/mmol; New EnglandNuclear, Boston, MA). Prehybridization was performed at 70    C for2 h. Hybridization was done with 2 ng radioactive probe per ml hy-bridization fluid for 2 h at 70    C. After hybridization, the membranewas washed at room temperature in 2    SSC (20   SSC: 3 M NaCl,0.3 M sodium citrate, pH 7.0), 0.1% SDS, followed by two washes(each wash for 15 min) at 65    C in 0.5    SSC with 0.1% SDS. Autora-diography was performed with intensifying screens for 48 h(RANTES cDNA) and 30 min (18S cDNA). Blots were stripped in 5mM Tris-HCl, 0.2 mM EDTA, 0.5% sodium pyrophosphate, and 5    Denhardt’s (50    Denhardt’s: 1% ficoll, 1% polyvinylpyrrolidone,1% BSA, 50% formamide, and 0.1% SDS) solution for 3 h at 65    C,and subsequently were rehybridized with a cDNA probe for the 18Sribosomal RNA to account for small loading and transfer variations.Exposed films were scanned with a laser densitometer (GS 300; Hoe-fer Scientific Instruments, San Francisco, CA) connected to a com-puter system, and the area under the curves was determined by    ANG II and RANTES  1049  Gaussian integration with the computer program GS 365W (Hoe-fer). Relative changes in RNA were calculated after assigning hy-bridization in control lanes a relative value of one (27). Sampleswere normalized for the signal intensity of the 18S hybridizations.All Northern blot experiments were repeated three times with simi-lar results.  RANTES ELISA.  RANTES protein production was measuredin cell culture supernatants with a commercial solid-phase ELISA as-say that uses two antibodies specific for RANTES (R&D Systems,Minneapolis, MN). For this assay, 10  6  GER or MC were made quies-cent in serum-free DMEM, and were subsequently stimulated withdifferent concentrations of ANG II (10     8  –10     6  M) in the presence orabsence of 10     6  M losartan, 10     6  M PD 123177, or 10     7  M CGP-42112A (Neosystem, Strasbourg, France) for 24 h. Additional cellswere treated with 1   g/ml LPS. At the end of the experiment, super-natants were harvested, and RANTES protein concentrations weredirectly measured in culture supernatants with the ELISA as recom-mended by the manufacturer. Cells were released from the plateswith trypsin/EDTA, and were counted in a hemocytometer. Concen-trations were expressed as fg RANTES/10  4  cells. Experiments wereindependently repeated six times with duplicate measurements.  Monocyte chemotactic assay.  Monocyte chemotactic activity wasdetermined in modified Boyden chambers (NeuroProbe, Cabin John,MD) by using freshly prepared human peripheral blood mononuclearcells exactly as previously described (28–30). In brief, 10  4  GER wereplated in 24-well plates, stimulated as indicated, and medium was col-lected after 24 h, centrifuged at 10,000  g  for 5 min, and stored at 70    Cuntil assay. After a 30-min incubation at 37    C with or without 30   g/ml neutralizing polyclonal goat anti-RANTES antibody (R&D Sys-tems), different dilutions in serum-free DMEM were assayed formonocyte migration using 3.5   10  6  monocytes/ml, freshly preparedfrom human blood by ficoll gradient centrifugation (28). As addi-tional control, 10     6  M ANG II was directly added to nonconditionedDMEM, and the chemotaxis assay was performed. Chemotactic ac-tivity was expressed as the mean number of monocytes migrating perfield in 10 high-power fields (29). Background migration in responseto nonconditioned medium (only serum-free DMEM) was 1.06    0.35(  n     10) and was subtracted from all values. Chemotactic assays wereindependently performed six times with duplicates for each experi-ment.   Animal experiments.  To test the effect of ANG II delivery in vivo,male Sprague-Dawley rats (SAVO-Ivanovas, Kissleg, Germany) witha body weight of 150 g were anaesthetized by ether, and osmoticminipumps (Model 2002; Alzet, Palo Alto, CA) were intraperito-neally placed under sterile conditions. These pumps delivered 500 ngANG II dissolved in 0.9% NaCl per h with a pumping rate of 0.5   l/h.Control animals received only 0.9% NaCl. As an additional positivecontrol, a limited number of rats (  n     3) were infused with 500 ng of LPS from E. coli  , serotype 02:B6 (endotoxin   10,000 U/mg LPS;Sigma Chemical Co.) per h. Some animals were directly treated afterpump implantation with the AT  2  receptor antagonist PD 123177 inthe drinking water (50 mg/liter) ad libitum. After 4 d, systolic bloodpressures were measured by tail plethysmography under light etheranesthesia, blood was collected in EDTA tubes on ice by puncturingthe aorta, and samples were stored at   70    C after plasma separation.Plasma renin activity (PRA) was measured as the generation of ANGI/ml/h with RIA using commercially available reagents (Sorin Bio-medica, Saluggia, Italy). Kidneys were removed, and glomeruli wereisolated by differential sieving with ice-cold Krebs-Ringer buffer aspreviously described (23). Purity of the preparations was assessed bylight microscopy. Part of the glomeruli was directly lysed in 10 ml of abuffer containing 4 M guanidine thiocyanate, 25 mM sodium citrate(pH 7.0), 0.5% sodium lauroyl sarcosinate, and 0.7%   -mercaptoeth-anol. Total RNA was prepared, and Northern blots for RANTES hy-bridization were performed as described above. In addition to glom-eruli, the remnant renal tissue was also collected for RNA isolation.Some glomeruli were not lysed, but incubated in serum-free DMEMat 37    C for 2 h in a shaker. After centrifugation, the protein contentof the glomeruli was determined by the Lowry method (31), andchemotactic activity of the supernatants was measured in various di-lutions as described above. As an additional control, 30   g/ml neu-tralizing polyclonal goat anti-RANTES antibody or control IgG wereadded to some supernatants before the chemotactic assay was per-formed. Isolated glomeruli from control animals and ANG II–infusedrats were also lysed in disruption buffer (60 mM Tris-HCl, pH 6.8, 2%SDS, and 100 mM dithiothreitol) and the protein content was mea-sured by a modification of the Lowry method, which is insensitive tothe used concentrations of SDS and dithiothreitol (31). Equalamounts of protein (100   g/lane) were loaded onto a denaturing 15%SDS-polyacrylamide gel. Low molecular weight Rainbow™ markers(2,350–46,000 kD; Amersham Buchler GmbH) were used as stan-dards. Proteins were electroblotted onto nitrocellulose (Hybond-N;Amersham Buchler GmbH), and membranes were stained with 0.2%Ponceaus S (Sigma Chemical Co.) to test for complete protein transfer.Detection of RANTES was performed exactly as previously de-scribed (16) using a 1:500 dilution of a mouse monoclonal anti–human RANTES antibody (R & D Systems). The secondary anti-body was a rabbit horseradish peroxidase-conjugated anti–mouseIgG in a 1:1000 dilution (Transduction Laboratories, Lexington, KY).Peroxidase labeling was detected with luminescence immunodetec-tion (ECL; Amersham Buchler GmbH) according to the manufac-turer’s recommendations.   Immunohistochemistry for M/M.  To analyze the infiltration of M/Minto glomeruli, a separate series of animals consisting of five rats pergroup were treated as described above. Kidneys were perfused in situwith ice-cold 0.9% NaCl, and tissues were fixed in Carnoy’s solution.Kidney tissue was cut and stained with a monoclonal antibody di-rected against the rat monocyte-specific marker ED-1 (Chemicon In-ternational, Inc., Temecula, CA). The stainings were developed usingthe alkaline phosphatase anti-alkaline phosphatase technique. Allquantitative morphologic analyses were performed in a blinded fash-ion. Evaluation of ED-1–positive cells was performed by countingpositive cells in 30 glomeruli in each section from five (three LPS-infused rats) individual animals.  RANTES immunohistochemistry.  Pilot experiments revealed thatfrozen sections of rat kidney may be successfully used to identifyRANTES protein expression in ANG II–infused animals, but thatthis procedure is suboptimal for delineation of structural details. Ad-ditional pilot studies demonstrated that fixation in Carnoy’s solutionor buffered formaldehyde destroyed RANTES epitopes (data notshown). A separate series of animals consisting of three rats pergroup were infused with ANG II or vehicle as described above. Kid-neys were in situ perfused with ice-cold 0.9% NaCl followed by 2%glutaraldehyde in cacodylate buffer for 2 min. Tissues were stainedusing a modified avidin–biotin complex technique with Tyramide Sig-nal Amplification system (Renaissance; DuPont-NEN, Boston, MA).Before staining, 5-    m sections were post-fixed with 1% glutaralde-hyde for 15 min at room temperature. Steam antigen retrieval pre-treatment was performed using 1    H.E.I.R. 10 mM citrate buffer(Ventana-Bioteck; Tucson, AZ), pH 6.0 at 90    C for 20 min. Afterpretreatment, sections were reacted with a murine monoclonal anti-human RANTES antibody (Biosource International, Camarillo, CA)at 0.2–0.5   g/ml. After primary antibody reaction, the sections weretreated according to a modification of the Ventana-Biotek protocol,which included staining enhancement using biotin–tyramide. Slideswere counterstained with hematoxylin. For quantification of positivestaining, an investigator unaware of the srcin of slides counted 100glomeruli and classified each glomerulus as negative or capillary lu-men positive. As positive control, NIH-3T3 fibroblasts were treatedfor 24 h with TNF-    (400 U/ml; Sigma Chemical Co.) in culture tostimulate RANTES expression. Cytospin sections were then stainedas described for kidney tissue. To ensure that the anti-RANTES anti-body specifically recognizes RANTES, the antibody was combinedwith recombinant human RANTES protein (R&D Systems) at a 1:2molar ratio with continous shaking for 24 h at 4    C. The mixture wasultracentrifuged and the supernatant was used for staining.   1050  Wolf et al. Statistical analysis.  All values are presented as means  SEM. Sta-tistical significance among multiple groups was tested with nonparamet-ric Kruskal-Wallis test. A P value of   0.05 was considered significant. Results  ANG II receptor subtypes. We have previously shown thatGER expressed high affinity receptors for ANG II (22). Thepresent studies were performed to characterize possible ANGII receptor subtypes. Fig. 1  A  demonstrates that competitionexperiments with the AT 1  antagonist losartan replaces   80%of the tracer [ 125 I][Sar 1 , Ile 8 ] ANG II. The AT 2  receptorblocker PD 123177 appears to have at lower concentrations aslightly higher affinity to the ANG II receptor compared withlosartan, and replaces up to 30% of the radioactive tracer (Fig.1  A ). The complex competition characteristics are compatiblewith expression of both AT 1  and AT 2  receptors on GER. Thepresence of specific transcripts for the AT 2  receptor was de-tected by cDNA amplification after reverse transcription (Fig.1 B ). In addition to the predicted size and selecting primersthat span introns, the identity of the band was confirmed bydideoxynucleotide chain termination sequencing after cloningin the TA II vector, and comparison with the published se-quence (26). As an additional control, cDNA amplification forAT 2  receptor mRNA was performed with RNA from isolatedglomeruli obtained from normal male Sprague-Dawley rats(Fig. 1 B ). Northern blots with 20  g total RNA, however,failed to detect a discrete signal, probably due to the low abun-dance of AT 2  receptor transcripts. RANTES production in GER. We first used a sandwichELISA to quantify RANTES protein in cell culture superna-tants of GER after stimulation with ANG II. To control forany growth stimulatory effects of ANG II on GER, RANTESprotein secretion was normalized to total cell counts. Asshown in Fig. 2, a single dose of 10  8 –10  6  M ANG II for 24 hsignificantly stimulated secretion of RANTES from GER.This effect was specific for these cells because syngeneic mes-angial cells failed to secrete RANTES after ANG II treatment(Fig. 2). No endotoxin could be detected in the used ANG IIpreparation, indicating that the induced RANTES secretion isnot due to contamination with endotoxins. 1  g/ml LPS for24 h, however, strongly stimulates RANTES secretion in bothglomerular cell types in accordance with previous observations(15, 32). This response appears to be specific for RANTES be-cause ANG II failed to stimulate the release of MCP-1, an-other chemokine of the C-C family (data not shown). To char-acterize the subtype of the ANG II receptor which isresponsible for signal transduction of this response, GER werestimulated in the presence of 10  6  M of the AT 1  antagonistlosartan, or 10  6  M of the AT 2  blocker PD 123177. In addition,10  6  M of the peptide AT 2  antagonist CGP-42112A with someagonistic properties was also used. ANG II–stimulatedRANTES secretion was almost completely abolished in thepresence of the AT 2  antagonists PD 123177 or CGP-42112A,whereas the AT 1  blocker losartan had no significant effect(Fig. 3).We next tested whether secreted RANTES induced byANG II treatment is chemotactic for human monocytes. Hu-man monocytes were used because they can be easily isolatedin large quantities, and samples from the same donor are a reli-able system for multiple analysis (29). As shown in Fig. 4  A ,conditioned supernatant from ANG II–treated GER in vari- Figure 1. ANG II-receptor expression in GER. (  A ) Competition of the AT 1  receptor antagonist losartan and the AT 2  blocker PD 123177 for [ 125 I][Sar 1 -Ile 8 ] ANG II binding sites on GER. GER express AT 1  and AT 2  receptors with complex binding characteristics. Mean values of four independent binding experiments with duplicates for each measurement. ( B ) cDNA amplification of reverse-transcribed RNA isolated from GER and normal rat glomeruli using specific primers for the rat AT 2  receptor. The predicted band at 476 bp indicates the presence of AT 2  recep-tor transcripts in GER and whole glomeruli. The identity of the bands was confirmed by sequencing the cDNA amplification products. A total of 40 cycles was performed with an annealing temperature of 60  C for 1.5 min, an extension step at 72  C for 1.5 min, and a denaturation step at 92  C for 1 min.   ANG II and RANTES 1051 ous dilutions has significantly more chemotactic activities forhuman monocytes than culture supernatant from control GERthat were not treated with ANG II. A maximal monocyte mi-gration was observed at a 1:2 dilution of the conditioned super-natant. A bell-shape activity curve with an optimal monocytemigration at a distinct dilution is typical for chemotactic as-says, and represents a genuine chemotactic response (33). A Figure 2. RANTES secretion in culture supernatants. A single dose of 10  8 –10  6  M ANG II for 24 h significantly stimulated RANTES protein secretion in GER, but not in syngeneic MC as measured by a sandwich ELISA. LPS stimulated RANTES secretion in both cell cultures. n     6 independent stimulation experiments with duplicate measurements; * P   0.01 versus control; ** P   0.001 versus controls. White bars , MC;  striped bars , GER. Figure 3. Effect of ANG II receptor blocker on RANTES secretion. Quiescent GER were stimulated for 24 h with 10  6  M ANG II in the presence or absence of 10  6  M of the receptor antagonists. The ANG II–stimulated RANTES secretion was attenuated by the AT 2  recep-tor antagonists PD 1231777 ( PD ) or CGP-42112A (CGP), but not by the AT 1  antagonist losartan ( los ). n     6 independent stimulation ex-periments with duplicate measurements; ** P   0.01 versus unstimu-lated controls; * P   0.05 versus cells treated with ANG II only. Figure 4. Chemotactic assay for human monocytes. Conditioned me-dium from GER treated for 24 h with a single dose of 10  6  M ANG II or supernatant from GER incubated with solvent (0.9% NaCl) were diluted with endotoxin-free sterile PBS and used for the chemotactic assay in a modified Boyden chamber. Migrated monocytes were quantitated by counting after staining of filter. (  A ) 1:2 dilution of con-ditioned medium from ANG II–treated cells showed the maximal chemotactic activity. All other dilutions of conditioned supernatants from ANG II–incubated GER, however, exhibited more chemotactic activity than control supernatants. Such a bell-shape dilution curve is typical for chemotactic assay. n     6 separate GER stimulation exper-iments with subsequent chemotactic assays in duplicate. * P   0.05 versus control medium; ** P   0.001 versus control medium. White  squares , control medium; black circles , ANG II medium. ( B ) 1:2 dilu-tion of conditioned supernatant from ANG II–treated GER was supplemented with 30  g/ml of a neutralizing goat anti-RANTES antibody ( anti-RANTES Ab ) or normal goat IgG. The neutralizing anti-RANTES antibody almost completely abolished the chemotactic activity in conditioned supernatant from ANG II–treated cells, indi-cating that the chemotactic factor for human monocytes is RANTES. Control goat IgG did not influence the chemotactic activity of condi-tioned supernatant obtained from ANG II–treated GER. Direct addi-tion of 10  6  M ANG II in serum-free DMEM into the Boyden cham-ber exhibited no significant chemotactic activity, suggesting that ANG II alone is not chemotactic. n     6 separate GER stimulation experiments with subsequent chemotactic assays in duplicate. * P   0.01 versus control medium; * # P   0.05 versus ANG II medium without antibody.
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!