Kidney International, Vol. 49 (1996), pp. 1320—1329

Role of endothelin as a mitogen in experimentalglomerulonephritis in rats

KY0IcHI FUKUDA, TAIHET YANAGIDA, SEIYA OKUDA, KIYOSHI TAMAKI, TAKASHI ANDO,and MASATOSHI FUJTSHIMA

Second Department of Internal Medicine, Faculty of Medicine, Kyushu University, Fukuoka, Japan

Role of endothelin as a mitogen in experimental glomerulonephritis inrats. Recent studies have revealed that endothelin-1 (ET-1) is a potentmitogen for mesangial cells in vitro. To determine whether ET-1 exerts themitogenic action on mesangial cells in vivo, we examined the glomerularexpression of ET-1 and its receptors in a rat model of mesangialproliferative glomerulonephritis and assessed the effect of a specificendothelin A (ETA) receptor antagonist, FR139317, on mesangial cellproliferation in this model. The levels of preproET-i mRNA expressionand ET-1 protein production in glomeruli increased markedly on days 4and 7 after disease induction, and the levels changed in concordance withthe glomerular cell proliferation. In contrast, the level of ETA receptormRNA initially decreased on day 1, and thereafter increased on days 4 and7. Administration of FR139317 to rats with experimental glomerulone-phritis induced a significant reduction in mesangial cell proliferation. Inaddition, in situ hybridization of preproET-I mRNA and double-immu-nolabeling of ED-i and OX-7 in a mirror image section revealed that theprincipal cells expressing ET-1 in glomeruli were infiltrating macrophageson day 1, and they were replaced by mesangial cells on day 4. Thesefindings indicate that ET-i functions as a potent mitogen for mesangialcells in vivo in an autocrine or paracrine fashion.

Endothelins (ET) are a family of 21 amino acid peptides thatinclude three distinct isoforms (ET-1, ET-2, and ET-3) [1, 2].These isopeptides show different biological activities and bind toat least two subtypes of G protein-coupled receptors. One is ETAreceptor that shows a selective affinity for ET-1 and ET-2, and theother is ETB receptor that shows a non-selective affinity for threeisopeptides [3—6]. Although endothelin was originally identified asa potent vasoconstrictor peptide, subsequent studies have re-vealed that endothelin possesses a wide range of biologicalactivities in addition to vasoconstrictor action, including regula-tion of vasomotor tone, neuromodulation and neurotransmission,Na and water handling in the kidney, and release of pituitaryhormones, atrial natriuretic peptide, renin, and catecholamines [7,8].

In addition, many in vitro studies have revealed that ET-1 is apotent mitogen for a variety of cell types [9, 101. In the kidney,ET-1 is produced by glomerular endothelial cells, mesangial cells,epithelial cells, and tubular epithelial cells [11—131. ET-1 has also

Received for publication September 25, 1995and in revised form November 30, 1995Accepted for publication November 30, 1995

© 1996 by the International Society of Nephrology

been shown to exert a potent mitogenic effect on culturedmesangial cells as well as on vascular smooth muscle cells,suggesting that ET-1 acts as a growth factor in the kidney [14, 15].In contrast to the findings of in vitro studies, the mitogenic actionof ET-1 has not yet been established in vivo. Therefore, wedesigned this study to determine whether ET-1 functions as amitogen for mesangial cells in vivo and, if so, which glomerularcells mainly produce ET-1. To answer these questions, we chosean experimental model of mesangial proliferative glomerulone-phritis (anti-Thy I UN). This model is relevant to our studybecause it causes selective injury and subsequent proliferation ofmesangial cells, and moreover, its similar features, namely me-sangiolysis and mesangial proliferation, can be observed in manyhuman renal diseases such as hemolytic-uremic syndrome, renaltransplant rejection, IgA nephropathy, acute glomerulonephritis,and diabetic nephropathy [16]. In the present study, we examinedthe glomerular expression of ET-1 and its receptors in rats withanti-Thy I GN, and in addition assessed the effect of endothelinreceptor antagonist on mesangial cell proliferation in the samemodel.

Methods

Experimental protocol

Experiment I. Male Sprague-Dawley (SD) rats, 6 to 8 weeks old,were used in the present study. Experimental mesangial prolifer-ative glomerulonephritis was induced by intravenous administra-tion of anti-thymocyte serum (ATS), 1 ml/100 g body wt, in SDrats (anti-Thy 1 GN). ATS was produced by immunizing NewZealand white rabbits with I X i0 rat thymocytes in completeFreund's adjuvant, followed by booster administration of 1 >< 106thymocytes intravenously two and four weeks later [17]. The ratswere divided into the following five groups: (1) normal baselinerats (day 0) (N = 21); (2—5) rats with anti-Thy 1 GN on 1, 4, 7, and14 days after disease induction (N = 21 in each group). The ratswere sacrificed for histological examination of kidney tissue andisolation of glomeruli for both culture and RNA extraction.

Experiment II. The effect of a specific ETA receptor antagonist,FR139317 (Fujisawa Pharmaceutical Co., Ltd., Osaka, Japan)[18], on glomerular cell proliferation was investigated in rats withanti-Thy 1 GN. Anti-Thy 1 GN in SD rats was induced using thesame methods as described in the experiment I. The rats weredivided into the following two groups: (1) rats that were admin-istered FR139317 at a dose of 10 mg/lOU g body wt daily and

1320

Fukuda et al: Mitogenic action of endothelin 1321

intraperitoneally from three days before to seven days after ATSinjection (N = 16); (2) control rats that were administered anequivalent quantity of saline for the same period (N = 16). Foreach group of rats, systolic blood pressure was measured beforeand on days 4 and 7 after disease induction. The systolic bloodpressure was measured in conscious rats by the tail cuff method.The rats were sacrificed on day 4 or 7, and the kidneys wereexcised for histological examination.

Glomerular culture

The rats were anesthetized intraperitoneally with pentobarhitalsodium 5 mgIlOO g body wt. The kidneys were perfused in situ viathe aorta with phosphate-buffered saline (PBS), pH 7.4 and thenexcised. The capsules were removed and then the cortical tissuewas dissected out and minced with a razor blade. Glomeruli wereisolated using the graded sieving technique [19]. The tissue waspassed sequentially through a 150, 112, and 71 jim sieve. intactglomeruli retained on the 71 jim sieve were removed, washedthree times in PBS, pH 7.4, and centrifuged. All glomerularpreparations used consisted of more than 95% glomeruli withminimal tubular contamination. Isolated glomeruli were resus-pended at 5 X i03 glomeruli per milliliter in serum-free andantibiotic-free Dulbecco's modified Eagle's medium (DMEM)(GIBCO BRL, Grand island, NY, USA) in six-well multiwellplates [20, 21]. After a 24-hour incubation, the glomeruli sus-pended in 1.0 ml of DMEM were quickly homogenized withPolytron for 15 seconds [21]. The homogenate was centrifuged at1500 g for five minutes, and the supernatants were removed, andstored at —20°C until endothelin assay was performed.

Endothelin assayA 1 ml aliquot of supernatant acidified with trifluoroacetic acid

(TFA) was applied to Spe-C8 column (iT. Baker Inc., Phillips-burg, NJ, USA) prewashed sequentially with methanol, distilledwater, and 0.1% TFA [22]. The materials adsorbed to thecartridge were eluted with 2 ml of 60% acetonitrile and 0.1%TFA, lyophilized, and reconstituted in the assay buffer for radio-immunoassay (RIA). The assay buffer was 0.01 M phosphatebuffer, pH 7.4, Containing 0.14 M NaCl, 0.01 M EDTA, 0,02 Mglycine, 0.01 M €-aminocaproic acid, 0.001 M sodium azide, 0.1mglml heat-inactivated human serum albumin, and 0.1% TritonX-100. The incubation mixture consisted of 0.1 ml of sample orstandard, 0.1 ml of assay buffer, and 0.1 ml of anti-ET-1 serum(final dilution 1:20000; Peninsula Laboratories, Belmont, CA,USA). Incubation was carried out at 4°C for 24 hours, followed bythe late addition of 0.05 ml of 1251-ET-1 (Amersham Corp.,Arlington Heights, IL, USA) and further incubation at 4°C for 48hours. To separate bound radioactivity from free ligands, 10 jil ofgoat anti-rabbit IgG, 1 l of normal rabbit serum, and 5%polyethylene glycol were added and centrifuged at 2000 g for 30minutes. The radioactivity in the precipitate was counted in ay-spectrometer. A standard curve was constructed, and sampleET-1 values were extrapolated from the standard curve linearizedby a logit-log transformation.

Preparation of cDNA probeseDNA probes for rat preproET-1 and rat ETB receptor were

kindly provided by Dr. T. Sakurai (Tsukuba University, Ibaraki,Japan) [23]. eDNA probes for rat ETA receptor and rat glyceral-dehyde-3-phosphate-dehydrogenase (GAPDH) were prepared

using reverse transcription in combination with polymerase chainreaction (RT-PCR) as described previously [241.

For the rat ETA receptor gene, the sense primer was 5'-CAGCAGACAAAAATCACGAC-3' defined by bases 254 to273, the antisense primer was 5'-AGAAGTAGAATCCAAA-GAGC-3' defined by bases 817 to 836, according to full-length ratETA receptor eDNA [6]. For the rat GAPDH gene, the senseprimer was 5'-TCCCTCAAGATFGTCAGCAA-3' defined bybases 451 to 470, the antisense primer was 5'-AGATCCA-CAACGGATACATT-3' defined by bases 739 to 758 [25]. ThePCR products were gel purified and subeloned into the EcoRVsite of pBluescript II SK(—) using T-vector method [261. Theauthenticity of the PCR amplified products was validated by DNAsequencing using fluorescence-based cycle sequencing methodwith Taq DNA polymerase and fluorescence DNA sequencer.

RNA extraction and Northern blot analysis

Glomerular isolation was carried out as described above. TotalRNA was extracted from isolated glomeruli of four rats in eachgroup by the acid guanidinium thiocyanate-phenol-chloroformextraction method [27]. Twenty micrograms of total RNA wassubjected to electrophoresis in a 2.2 M formaldehyde-i% agarosegel and transferred to Hybond-N nylon membranes (Amersham),and then fixed by baking at 80°C for two hours. The membraneswere prehybridized for at least two hours at 37°C in hybridizationsolution (5 X SSC, 5 X Denhardt's solution, 0.1 mg/mI of salmonsperm DNA, 0.1% SDS, and 50% formamide). The eDNA probeswere labeled with 32P-dCTP by the random primer method andhybridized in the hybridization solution at 42°C overnight. Themembranes were washed twice in 2 >< SSC, 0.1% SDS, and twicein 1 X SSC, 0.1% SDS at 42°C for 15 minutes. Autoradiographywas performed by the standard methods. The autoradiographswere scanned with a Dualwavelength flying-spot CS-9000 scanner(Shimidzu, Tokyo, Japan) to quantitate the relative intensities ofthe mRNA bands. The quantitative measurements of each mRNAband were expressed as the ratio to that of GAPDH mRNAbands.

Routine histological examination

Kidney tissues for light microscopy were fixed in neutralformalin and embedded in paraffin, and 2 jim sections werestained with periodic acid-Schiff reagent.

Immunohistochemical staining for PCNA and

monocyte-macrophages

Proliferating cell nuclear antigen (PCNA) is an auxiliary pro-tein to DNA polymerase-, and it is expressed in nucleus stronglyin late G1 and S phases of the cell cycle [28, 29]. Therefore, PCNAantibodies are used as tools for detecting proliferating cells. Toevaluate cell proliferation and cell infiltration, we carried outimmunohistochemical staining for PCNA and monocyte-mac-rophages. Kidney tissues were fixed in neutral formalin andembedded in paraffin, and 3 jim sections were mounted onpoly-L-lysine-coated slides. The sections were deparaffinized,immersed in methanol containing 0.3% H202 for 30 minutes, andstained for PCNA with PCIO, a mouse monoclonal IgG antibody(Daco Corp., Santa Barbara, CA, USA) [30], or for monocyte-macrophages with ED-i, a mouse monoclonal IgG antibody(Serotec, Oxford, UK) [31]. As a pretreatment, sections for PC1O

1322 Fukuda et al: Mitogenic action of endothelin

were immersed in 2 M HC1 for 30 minutes [32], or sections forED-i were digested with proteinase K (5 pg/mi) (BoehringerMannheim GmbH Biochemica, Mannheim, Germany) at 37°C for15 minutes. Thereafter, these sections were incubated with PC10or ED-i respectively, and subsequently processed using a strepta-vidin-biotin immunoperoxidase method [33] with 3,3'-diamino-benzidine (DAB) as the chromogen.

Quantitation of histological findings

For each section, 50 glomeruli were sequentially examined andthe mean numbers of following variables per glomerular cross-section were measured: cell nuclei in periodic acid-Schiff stainedtissue sections (total cellularity), PCNA positive cells (prolifera-tion), and ED-i positive cells (monocyte-macrophage infiltration)determined by immunohistochemistry. Glomerular cross sectionsmeasuring less than 80 am diameter were excluded from theanalysis.

Mirror section analysis of preproET-i mRI'JA and ED-i or OX-7

To identify the glomerular cells expressing ET-i, we carried outa mirror section analysis of preproET-1 mRNA and cell markerantigens [34, 35]. Kidney tissues were fixed with 4% paraformal-dehyde in 0.1 M phosphate buffer overnight, dehydrated withgraded ethanol series and chloroform, and embedded in paraffin.Using a mirror sectioning technique, two serial 3-.tm-thick sec-tions were obtained with the cut surfaces facing each other, andmounted on 2% 3-amino-propyl-triethoxysilane coated slides. Alltreatments were performed in a ribonuclease free condition.

Each section was then individually reacted either with a pre-proET-1 cRNA probe by in situ hybridization or with ED-iantibody and OX-7 antibody by immunohistochemistry as de-scribed below.

Double immunohistochemical staining

Kidney sections were deparaffinized, immersed in methanolcontaining 0.3% H202 for 30 minutes, and digested with protein-ase K (5 pg/mI) at 37°C for 15 minutes, and then the sections weredouble immunostained for monocyte-macrophages and mesangialcells.

First, the sections were stained for monocyte-macrophages withED-i, a mouse monoclonal IgU antibody, using biotinylatedrabbit anti-mouse IgG, streptavidin-immunoperoxidase, and 3,3'-diaminobenzidine (DAB) (Nichirei Corp., Tokyo, Japan) as thechromogen. The sections were then immersed for 1.5 hours inglycine-hydrochioric acid buffer at pH 2.2 to elute the tissuebound antibodies [36, 37]. Thereafter, the tissue sections wererinsed in PBS containing 0.005% biotin for 30 minutes to blockthe biotin-binding sites of remaining streptavidin [38].

Subsequently, the sections were stained for mesangial cells withOX-7, a mouse monoclonal IgG antibody (Accurate Chemicals,Inc., Westbury, NY, USA), using biotinylated rabbit anti-mouseIgG, streptavidin-alkaline phosphatase, and new fuchsin(Nichirei) as the chromogen. Activity of endogenous alkalinephosphatase was blocked by adding 1 mivi levamisole (Sigma, St.Louis, MO, USA) to the substrate solution [39]. The sections werecounterstained with hematoxylin. Negative controls were obtainedby omitting either of the primary antibodies.

Preparation of digoxigenin labeled riboprobes

Digoxigenin-labeled single strand RNA probes were preparedusing DIG RNA Labeling Kit (Boehringer Mannheim) accordingto the manufacturer's instructions. A 2.0 kb preproET-i cDNAfragment (provided by Dr. T. Sakurai), was subcloned into theEcoRI site of pBluescript II SK(—) transcription vector whichcontains promoters for T7 and T3 RNA polymerase (Stratagene,La Jolla, CA, USA). This plasmid was either linearized withrestriction enzyme PstI and transcribed with T7 RNA polymerase(GIBCO BRL) to generate an anti-sense probe, or linearized withHindill and transcribed with T3 RNA polymerase (BoehringerMannheim) to generate a sense probe. Subsequently, RNAprobes were ethanol-precipitated and subjected to alkaline diges-tion by incubation in 40 mrvi NaHCO3 and 60 mrvi Na2CO3 at 60°Cfor 10 minutes to obtain 500 bp RNA probes [40]. They were thenneutralized, ethanol-precipitated, and used for in situ hybridiza-tion. The specificity of the 2.0 kb cDNA was assessed by Northernblot analysis performed as described above.

In situ hybridization

Kidney sections were deparaffinized, rehydrated, and thendigested with proteinase K (5 iWml) at 37°C for 15 minutes. Thesections were refixed with 4% paraformaldehyde in 0.i M phos-phate buffer for 15 minutes, and acetylated with 0.25% aceticanhydride in 0.1 M triethanolamine pH 8.0 for 10 minutes. Theywere dehydrated with graded ethanol series and air dried. Thehybridization solution contained 50% deionized formamide, 10mM Tris-HCI pH 7.6, iXDenhardt's solution, 10% dextran sul-fate, 600 mivi NaC1, 0.25% SDS, i mM EDTA pH 8.0, 200 pg/mIof yeast tRNA, and 2.0 .tg/ml of RNA probe. Fifty microliters ofhybridization solution was applied to each section, and it wascoverslipped. Hybridization was carried out in a moist chamberhumidified by 50% formamide at 50°C for 16 hours [41]. Next, thesections were washed in 5 X SSC briefly and in 2 X SSC containing50% formamide for 30 minutes at 50°C. In some of the sections,RNase A treatment (20 j.Lg/ml RNase A in 10 mM Tris-HC1 pH8.0, 1 mi EDTA) was carried out at 37°C for 30 minutes toexclude any false-positive readings due to mishybridization. Thesections were washed in 2 x SSC and in 0.2 x SSC for 20 minutestwice at 50°C. Blocking was accomplished by 2% skim milk for onehour. The sections were incubated with alkaline phosphatase-conjugated anti-digoxigenin antibody (1:1000 dilution) (Boehr-inger Mannheim) for 30 minutes at room temperature. Afterwashing in DIG 1 buffer (100 mM Tris-HC1 pH 7.5, 150 ms NaC1)twice for 15 minutes, and in DIG 3 buffer (100 mM Tris-HC1 pH9.5, 100 mM NaCI, 50 mivi MgCl2) for five minutes, visualizationwas carried out with NBT (nitroblue tetrazolium salt), 5-bromo-4-chloro-3-indolyl phosphate, and 1 mM levamisole in DIG 3buffer. Controls were obtained by replacing the antisense probewith a sense probe under the same conditions.

Statistical analysis

Data are expressed as mean SE. The statistical analysis of themultiple groups was performed using a one-way analysis ofvariance (ANOVA) with the modified t-test by Bonferroni'smethod. Statistical significance was defined as P < 0.05.

Days

0 1 4 7 14

ppET-1

2.3 kb—-

ETAR4.2 kb—-

5.0 kb — *lbet.eulNusS

GAPDH

1.3kb— 4 t

Fukuda et al: Mitogenic action of endothelin 1323

Table 1. Total cellularity, proliferating cells, and monocyte-macrophages in normal rats day 0 (baseline), rats with anti-Thy 1 GN

on days 1, 4, 7, and 14

Day Total cellsProliferating cells

(PCNA+)Monocyte-macrophages

(ED-1+)0 47.8 1.1 0.5 0.1 1.2 0.21 41.0 0.6 2.6 0.5 11.9 0.4k'4 58.1 1.3 12.0 0.2 6.0 0.27 62.9 0.5 7.6 0.3a 5.5 o.2a

14 60.0 1.2 2.5 0.3 2.4 0.1Data are expressed as the mean number

cross section from eight rats in each group.a P < 0.01 versus day 0.

SE of cells per glomerular

Days

Results

Experimental glomerulonephritisThe administration of ATS to rats produced an acute form of

mesangial proliferative glomerulonephritis as described previ-ously [17, 421. The glomerular lesions in the early stage (on day 1)were characterized by a decrease in total glomerular cellularitydue to complement-mediated mesangiolysis, although it waspartially offset by markedly infiltrating glomerular monocyte-macrophages which were ED-i positive (Table 1). Such lesionswere followed by an increase in total glomerular cellularity in thelater stages (on days 4 and 7). The glomerular hypercellularity inthe later stages was considered to be mainly due to proliferatingcells which were PCNA positive (Table 1).

Glomerular production of ET-1 proteinAfter a 24-hour incubation, ET-1 production in isolated gb-

meruli from normal rats (day 0) was invariably low. ET-1 produc-tion in isolated glomeruli from rats with anti-Thy 1 GN increasedslightly on day 1, and further increased significantly on days 4 and7, being maximal on day 7 (5.3-fold greater than the baseline

level), and then decreased on day 14 after disease induction (Fig.1). The level of ET-1 production in glomeruli from rats withanti-Thy 1 GN changed in concordance with the glomerular cellproliferation.

Glomerular expression of preproET-1 and ET receptors mRNAExpression of the 2.3 kb preproET-1, 4.2 kb ETA receptor, and

5.0 kb ET receptor mRNA transcripts could be detected inglomerular RNA from normal rats (day 0) (Fig. 2). The level ofpreproET-1 mRNA increased significantly on days 1, 4, 7, and 14after ATS administration, being maximal on day 4 (4.5-foldgreater than the baseline level; Fig. 3A). The level changed inconcordance with the glomerular cell proliferation. The level ofETA receptor mRNA decreased significantly on day 1, thenincreased significantly on days 4 and 7, being maximal on day 7(5.2-fold greater than the baseline level), and then returned to thebaseline level on day 14 (Fig. 3B). The level of ETB receptormRNA decreased significantly on days 1, 4, and 7, and thenreturned to the baseline level on day 14 (Fig. 3C). The level ofGAPDH mRNA did not change significantly during the diseaseprocess.

Effect of ETA receptor antagonist on glomerular cell proliferationThe glomeruli of saline-treated rats with anti-Thy 1 GN showed

a marked increase in total cellularity on days 4 and 7 (Fig. 4A). Incontrast, the glomeruli of FR139317-treated rats showed a signif-icantly smaller increase in total cellularity than that of saline-treated rats on days 4 and 7, respectively (Fig. 4B) (Table 2). Thenumber of proliferating cells (that is, PCNA positive cells) inglomeruli of FR139317-treated rats was significantly smaller thanthat of saline-treated rats on days 4 and 7, respectively (Table 2).Therefore, the inhibition of increase in total glomerular cellularitycould have been caused by the reduction in glomerular cellproliferation. No significant differences in body weight or kidneyweight were observed between saline- and FR139317-treated rats

Fig. 2. Northern blot analysis of preproET-1 (ppET-1), ETA receptor(ETAR), and ETB receptor (ETBR) mRNA in glomendi of normal rats day 0(baseline), rats with anti-Thy 1 GN on days 1, 4, 7, and 14. Total RNA wasseparated on a 1% agarose gel (20 .rg/1ane) and probed with 32P-labeledeDNA. The same results were obtained in four different experiments.

Time, days

Fig. 1. ET-1 production in isolated glomeruli of normal rats day 0 (base-line), rats with anti-Thy 1 GNon days 1, 4, 7, and 14. Data are expressed asmeans SE for five separate experiments. *P < 0.05 versus day 0. <0.01 versus day 0.

0 1 4 7 14

Cl) 4

.0

Fig. 3. Quantitative analysis ofpreproET-1 (A), ET4 receptor (B), and ETBreceptor (C) mRNA in glomeruli of normal rats day 0 (baseline), rats withanti-Thy 1 GN on days 1, 4, 7, and 14. The relative intensity of theautoradiographic signals was quantitated and calculated as the ratio ofeach mRNA to GAPDH mRNA, and then expressed as degree of changerelative to baseline levels. 'P < 0.05 versus day 0. * *p < 0.01 versus day 0.

Fig. 4. Comparison of histological findings in glomeruli of rats with mesan-gial proliferative glomerulonephritis on day 7. A. The glomeruli of saline-treated rats show marked hypercellularity. B. The glomeruli of FR139317-treated rats show significantly less cellularity than that of saline-treatedrats. (Periodic acid-Schiff reagent with hematoxylin counterstain, X400).

A

1324 Fukuda et al: Mitogenic action of endothelin

6

5

3

2

0

0 1 4 7 14

BTime, days

Cl)

C

.0

(I,C

.0

C

Time, days

11

0.5

0-

on the day of sacrifice. Systolic blood pressure values did notchange significantly from the baseline levels in either saline- orFR139317-treated rats, and also showed no significant differencesbetween either group on days 0, 4, and 7 (Table 3).

Detection of preproET-1 mRNA by in situ hybridization

In normal rats (day 0), preproET-1 mRNA expression was

0 1 4 7 14detected in glomeruli, endothelial cells of vessels, and collectingducts. Glomerular preproET-1 mRNA expression was detectable

Time, days mainly in endothelial and mesangial regions, but the level ofmRNA was low (Fig. 5A). In rats with anti-Thy 1 GN, an increasein glomerular expression of preproET-1 mRNA was observed.

Fukuda et al: Mitogenic action of endothelin 1325

Table 2. Effect of FR139317 on total cellularity and proliferating cellsio rats with anti-Thy 1 UN

cellsProliferating cells

(PCNA+)

Day 0 47.8 1.1 0.5 0.1Day 4

Cootrol 58.0 1.0 12.8 0.3FR139317 49.6 1.? 5.9 0.4

Day 7Control 63.8 1.2 8.3 0.6FR139317 52.4 l.0' 4.2 o.3Data are expressed as the mean number SE of cells per glomerular

cross section from eight rats in each group. For comparison, data fromnormal rats day 0 are shown as the baseline level.

ap < 0.01 vs. control (saline-treated rats)hP C 0.01 vs. day 0

The level of preproET-1 mRNA on day 4 (Fig. 5 C, D) was muchhigher than that on day I in glomeruli (Fig. SB). The cellsexpressing strongly preproET-1 mRNA localized to hypercellularregions in glomeruli. These results were consistent with the resultsof Northern blot analysis in glomeruli. A control experiment witha sense probe did not show any staining of tissue (Fig. SE).

Identification of the cell type expressing ET-1To determine the cellular localization of glomerular pre-

proET-i mRNA, we carried out double immunolabeling formonocyte-macrophages with ED-i and for mesangial cells withOX-7 in the mirror image sections of in situ hybridization.

On day 1, marked glomerular monocyte-macrophage infiltra-tion, which was scattered in normal glomeruli, was noted, butmesangial cell staining was weak. In this early stage, mirrorsection analysis revealed that the cells expressing preproET-ImRNA were mainly infiltrating monocyte-macrophages, andpartly mesangial cells (Fig. SF).

On day 4, glomerular monocyte-macrophage infiltration de-creased, on the other hand, marked mesangial cell proliferationwas noted. In this later stage, mirror section analysis revealed thatmesangial cells replaced monocyte-macrophages as the main cellsexpressing preproET-i mRNA (Fig. SG). The hypercellular re-gions were occupied by mesangial cells expressing preproET-1mRNA, while monocyte-macrophages continued to express pre-proET-i mRNA (Fig. SH).

Discussion

The main question addressed by the present study was whetherET-i functions as a mitogen for mesangial cells in vivo. Thepresent study shows that ET-1 is a potent mitogen for mesangialcells in the experimental mesangial proliferative glomerulonephri-tis. Northern blot analysis of glomerular RNA demonstrated amarked increase in preproET-1 mRNA transcript in rats withexperimental glomerulonephritis. In addition, ET-1 production inisolated glomeruli also increased markedly in rats with glomeru-lonephritis. Furthermore, these increases were concordant withthe glomerular cell proliferation. Since most of the proliferatingcells have been shown to be mesangial cells in this model [43—45],ET-i gene and protein expression are considered to change inconcordance with the mesangial cell proliferation.

Recent studies have revealed that ET-i is a potent mitogen formesangial cells in vitro [14, 15], and this mitogenesis is thought to

Table 3. Effect of FR139317 on body weight (wt), kidney weight (wt),and the time course of systolic blood pressure (SBP)

Day Control FR139317

Body wt g 7 257.5 9.0 250.0 8.2Kidney wt g 7 1.02 0.08 0.96 0.07SBP mm Hg 0

47

146.5 1.1150.6 4.6146.6 4.1

144.0 1.1149.5 3.3145.1 1.4

Data are expressed as means 5E from eight rats in each group. Nosignificant differences were observed between control and FR139317groups.

be mediated by ETA receptor [46], while mesangial cells have bothETA and ETB receptor [47—49]. Therefore, to confirm the mito-genie function of ET-1 in vivo, we administered a specific ETAreceptor antagonist, FR139317, to rats with experimental glomer-ulonephritis, and performed histological evaluation of renal tis-sues after disease induction. As a result, FR139317 significantlyreduced glomerular cell proliferation as compared to control.Previous studies demonstrated that another ETA receptor antag-onist, B0123, inhibits ET-1-induced 3H-thymidine uptake invascular smooth muscle cells [50, 51], and FR139317 reducesproliferation of mesangial cells in a model of progressive nephronloss in rats [52] or lupus nephritis in mice [53]. These data stronglysupport our results. FR139317 did not affect the nutritionalcondition or systolic blood pressure during the disease process. Inthe previous report, B0123 did not affect systemic or renalhemodynamics in normotensive rats [54]. Since rats with anti-Thy1 UN are usually normotensive, the inhibitory effects of FR139317on cell proliferation would probably not be due to any indirecteffects through antihypertensive actions but be due to the directeffects of antagonizing ETA receptor. Taken together with thefindings of ET-1 gene and protein expression, our data on ETAreceptor blocking provide convincing evidence that ET-1 func-tions as a potent mitogen for mesangial cells in experimentalglomerulonephritis and the mitogenesis is mediated by ETAreceptor at least in part.

The other question addressed by the present study was whichglomerular cells mainly produce ET-1 in this model. To answerthis question, we carried out a mirror section analysis of pre-proET-1 mRNA and cell marker antigens. Our results show thatthe principal cells expressing preproET-1 mRNA are infiltratingmonocyte-macrophages in the early stage, and they are replacedby mesangial cells in the later stage. These results could beinterpreted as follows. In the early destructive stage (1 hr to 2days), most mesangial cells are injured and functionally damaged[17j. Therefore, infiltrating monocyte-macrophages, a rich sourceof cytokines, might play an important role for initiating mesangialcell proliferation in a paracrine fashion by releasing ET-1. It issupported by recent in vivo and in vitro reports that monocyte-macrophages could produce and release ET-1 [55, 56]. In the laterproliferative stage (4 to 14 days), ET-1 released from monocyte-macrophages might decrease as they left from glomeruli, and thenmesangial cells might become a major source of ET-1 production,which could induce a subsequent proliferation of mesangial cellsin an autocrine fashion.

Although the present data demonstrated an increase in glomer-ular ET-1 expression and detected the source of ET-1 in experi-mental glomerulonephritis, the mechanism of ET-1 overexpres-sion remains unclear. Recent studies have demonstrated the

E

'

'I- -; 4

- S• S

a,-15St

a

F,—-.'It—

—

a

V.--S ,I_"4

C—.

4'--4,.

0I—

• 4.•

'I

t.4

—

-I,•0'

4,

pSa

4 , -.- '-1.;

t.a

1326 Fukuda er al: Mitogenic action of endothelin

Fukuda et at: Mitogenic action of endothelin 1327

Fig. S. In situ hybridization for the preproET- I mRNA (A-E) and double-immunolabeling with ED-i and OX-7 (F-H) in normal rats and rats with mesangialproliferative glomerulonephritis. (B) and (F) arc mirror image sections on day 1, while (C) and (G), (D) and (H) arc mirror image sections on day 4. Innormal rats, preproET-1 mRNA is expressed in endothelial cells and mesangial cells in glomcruli, endothelial cells of small arteries (arrow), andendothelial cells in arterioles (arrowhead) (A). On day 1 after disease induction, increased expression of glomerular preproET-1 mRNA is observed (B).On day 4, expression of glumerular preproET-i mRNA becomes strongly positive and localizes to hyperccllular regions (C, D). Hybridization with asense probe shows no staining of glomeruli or vessels on day 4 (E). Double-immunolabcling in the mirror image section on day 1 shows that the cellsexpressing preproET-i mRNA are mainly ED-i positive (brown; arrows) and partly OX-7 positive (bright red; arrowheads) (F). On day 4, the cellsexpressing prepruET-1 mRNA are mainly OX-7 positive (bright red) (0), while some of the cells expressing preproET-1 mRNA are ED-i positive(brown; arrows) (H) (X400).

importance of mediators derived from platelets or monocyte-macrophages for glomerular cell proliferation in renal diseases[57, 58]. Like monocyte-macrophages, platelets contain manymediators including thrombin, interleukin-1, TGF-f3, and PDGF[59, 60], which could be released in glomeruli after glomerularinjury. Indeed, increased glomerular expression of TGF-f3 [61] orPDGF [62] was shown in rats with experimental glomerulonephri-tis. In addition, recent in vitro studies have revealed that ET-1production in mesangial cells is triggered by these mediators [63,64]. Thus, these mediators derived from platelets or monocyte-macrophages could contribute to the increase of ET-1 expressionin mesangial cells in the later stage. Finally, released ET-i couldexert a proliferative effect on mesangial cells in an autocrinefashion in combination with thcse mediators.

The results of mirror section analysis that the main cellsexpressing ET-1 varied with the different stages might explain thediscrepancy in the expression level between preproET-1 and ETreceptors mRNA. It was previously shown that ETA receptorexpression was localized to mesangial cells, and that ETB receptorexpression was localized to endothelial cells and mesangial cells innormal glomeruli [5]. In the early stage (day 1), most mesangialcells are damaged as described above, so that the decreasedexpression of ETA and ETB receptor in this stage could be relatedto the damage of receptor source cells, that is, mesangial cells. Onthe other hand, the increased expression of ET-i could beexplained by infiltrating monocyte-macrophages instead of dam-aged mesangial cells as shown by mirror section analysis. Incontrast, it is still controversial whether monocyte-macrophageshave the ET receptors. There have been a few previous reportsabout ET receptors of monocyte-macrophages [55, 65, 66], but thefindings of these reports differ greatly regarding the existence orthe character of this receptor. At least according to our results, itappears that glomerular infiltrating monocyte-macrophages didnot express ET receptors, because the decreased expression of ETreceptors should be offset by monocyte-macrophages to someextent if they expressed either ETA or ET receptor. In short, theearly accumulation of monocyte-macrophages in glomeruli couldexplain the increased expression of preproET-i mRNA in spite ofthe decreased expression of ET receptor mRNA in the early stage.

In the later stages (days 4 and 7), ETA receptor expressionincreased, while ETB receptor expression remained decreased,and then both receptors expression returned to each baseline levelon day 14. The mechanisms how both receptors showed differentexpression patterns in this stage are still unclear, but the antipro-liferative effects of ETA receptor antagonist indicate that theincreased ETA receptor expression contributes to the amplifica-tion of mesangial cell response to ET-1, and finally to subsequentmesangial cell proliferation. In contrast, the consistent decreaseof ETB receptor expression may prevent overexpression of ET-1,

and thus might possibly contribute to the self-limiting course ofanti-Thy 1 ON, because ET-1 increases its own synthesis throughETB receptor [67]. Very little remains known about the regulationof ET receptor expression in any tissue, and further studies arestill needed to clarify the mechanisms.

In conclusion, the present study provides direct in vivo evidencethat ET-1 functions as a mitogen in mesangial proliferativeglomerulonephritis, and that ET-1 acts for mesangial cells in aparacrine fashion during the early stage and in an autocrinefashion during the later stage.

Acknowledgments

This study was supported by a Research Grant from the GlomerularInjury Research Committee of the Intractable Disease Division, PublicHealth Bureau, Ministry of Health and Welfare, Japan. We thank Dr. T.Sakurai, Tsukuba University, for providing the preproET-i and ETareceptor eDNA, and Miss H. Nuguchi for her helpful technical assistance.We also thank Prof. B.T. Quinn for comments on the manuscript.

Reprint requests to Kyoichi Fukuda, Second Department of InternalMedicine, Faculty of Medicine, Kyushu University, Maidashi 3-i-i, Higashi-ku, Fukuoka 8i2-82, Japan.

References

1. YoiswAM,KURIHARA H, KJMURA 8, ToMoan Y, KoaAYAsm M,MIT5UJ Y, YAZAKI Y, GoTo K, MA5AKI T: A novel potent vasocun-strictor peptide produced by vascular endothelial cells. Nature (Lond)332:411—415, i988

2. INoUB A, YANAGI5AwA M, KIMURA S, KA5UvA Y, MIYAucHI T, GoToK, MAs&IcI T: The human endothelin family: three structurally andpharmacologically distinct isopeptides predicted by three separategenes. Proc Natl Acad Sci USA 86:2863—2867, 1989

3. ARAt H, H0RI S, ARAMORI I, OHKuao H, NMCn'asHl 8: Cloning andexpression of a eDNA encoding an endothelin receptor. Nature(Lend) 348:730—732, 1990

4. SAKURAI T, YANAGI5AWA M, TAKUwA Y, MIYAZAKI H, KJMuRA S,GoTo K, MA5AJU T: Cloning of a eDNA encoding a non-isopeptide-selective subtype of the endothelin receptor. Nature (Lend) 348:732—735, 1990

5. Horn 8, K0MAT5u Y, SHIGEMOTO R, MIZUNO N, NAK.kr'Isrn S:Distinct tissue distribution and cellular localization of two messengerribunueleic acids encoding different subtypes of rat endothelin recep-tors. Endocrinology 130:1885—1895, 1992

6. LIN HY, KAJI EH, WINKEL OK, IvEs HE, Louisa HF: Cloning andfunctional expression of a vascular smooth muscle endothelin 1receptor. Proc Natl Acad Sci USA 88:3185—3189, 1991

7. FUKuuA Y, HtRATA Y, Y05HIMI H, K0JIMA T, KOBAYA5HI Y, YANAGI-5AWA M, MA5AKI T: Enduthelin is a potent secretagogue for atrialnatriuretie peptide in cultured rat atrial myocytes. Biochem BiophysRes Common 155:167—172, 1988

8. RAKUGI H, NAKAMARU M, SAITO H, HIGAKI J, OGIHAItk T: Endothe-lin inhibits renin release from isolated rat glomeruli. Biochem BiophysRes Common 155:1244—1247, 1988

1328 Fukuda et al: Mitogenic action of endothelin

9. HIRATA Y, TAKAGI Y, FUKUDA Y, MARUMO F: Endothelin is a potentmitogen for rat vascular smooth muscle cells. Atherosclerosis 78:225—228, 1989

10. TAKAGI Y, FUKASE M, TAKATA S, Y0sHIMI H, TOKUNAGA 0, FUJITAT: Autocrine effect of endothelin on DNA synthesis in human vascularendothelial cells. Biochem Biophys Res Commun 168:537—543, 1990

11. SAKAMOTO H, SASAKI S, HIRATA Y, IMAI T, ANDO K, IDA T, SAKURAIT, YANAGISAWA M, MASAKI T, MARUMO F: Production of endothe-un-i by rat cultured mesangial cells. Biochem Biophys Res Commun169:462—468, 1990

12. MARSDEN PA, DORFMAN DM, COLLINS T, BRENNER BM, ORKIN SH,BALLERMANN BJ: Regulated expression of endothelin 1 in glomerularcapillary endothelial cells. Am J Physiol 261:F117—F125, 1991

13. KASINATH BS, FRIED TA, DAVALATH S, MARSDEN PA: Glomerularepithelial cells synthesize endothelin peptides. Am J Pathol 141:279—283, 1992

14. BADR KF, MURRAY JJ, BREYER MD, TAKAHASHI K, INAGAMI T,HARRIS RC: Mesangial cell, glomerular and renal vascular responsesto endothelin in the rat kidney. J Clin Invest 83:336—342, 1989

15. SIMONSON MS, WANN S, MENE P, DUBYAK GR, KESTER M, NAKA-zATO Y, SEDOR JR, DUNN MJ: Endothelin stimulates phospholipaseC, Na/H exchange, c-fos expression, and mitogenesis in rat mesan-gial cells. J Clin Invest 83:708—712, 1989

16. MORITA T, CHURG J: Mesangiolysis. Kidney mt 24:1—9, 198317. YAMAMOTO T, WILSON CB: Quantitative and qualitative studies of

antibody-induced mesangial cell damage in the rat. Kidney mt 32:5 14—525, 1987

18. ARAMORI I, NIREI H, SHOUBO M, SOGABE K, NAKAMURA K, KoJo H,NOTSU Y, ONo T, NAKANISHI S: Subtype selectivity of a novelendothelin antagonist, FR139317, for the two endothelin receptors intransfected Chinese hamster ovary cells. Mol Pharmacol 43:127—131,1993

19. STRIKER GE, STRIKER U: Glomerular cell culture. Lab Invest 53:122—131, 1985

20. BENIGNI A, PERICO N, GASPARI F, ZOJA C, BELLIZZI L, GABANELLI M,REMUZZI G: Increased renal endothelin production in rats withreduced renal mass. Am J Physiol 260:F331—F339, 1991

21. NAKAMURA T, EBIHARA I, FUKUI M, OSADA 5, TOMINO Y, MASAKI T,GoTo K, FuRUICI-lI Y, KOIDE H: Modulation of glomerular endothelinand endothelin receptor gene expression in aminonucleoside-inducednephritis. JAm Soc Nephrol 5:1585—1590, 1995

22. ANDO K, HIRATA Y, SHICHIRI M, EMORI T, MARUMO F: Presence ofimmunoreactive endothelin in human plasma. FEBS Lett 245:164—166, 1989

23. SAKURAI T, YANAGISAWA M, IN0UE A, RYAN US, KIMURA S, MITSUIY, Go-ro K, MASAKI T: eDNA cloning, sequence analysis and tissuedistribution of rat preproendothelin-1 mRNA. Biochem Biophys ResCommun 175:44—47, 1991

24. ANDO T, OKUDA S, TAMAKI K, Y0SHIT0MI K, FUJISHIMA M: Local-ization of transforming growth factor-13 and latent transforminggrowth factor-13 binding protein in rat kidney. Kidney mt 47:733—739,1995

25. TERADA Y, TOMITA K, NONOGUCHI H, MARUMO F: Different local-ization of two types of endothelin receptor mRNA in microdissectedrat nephron segments using reverse transcription and polymerasechain reaction assay. J Clin Invest 90:107—112, 1992

26. MARCHUK D, DRUMM M, SAULINO A, COLLINS FS: Construction ofT-vectors, a rapid and general system for direct cloning of unmodifiedPCR products. NuclAcids Res 19:1154, 1991

27. CHOMCZYNSKI F, SACCHI N: Single-step method of RNA isolation byacid guanidinium thiocyanate-phenol-chloroform extraction. AnalBiochem 162:156—159, 1987

28. BRAVO R, FRANK R, BLUNDELL PA, MACDONALD BH: CyclinIPCNAis the auxiliary protein of DNA polymerase-b. Nature (Lond) 326:5 15—517, 1987

29. KURKI P, VANDERLAAN M, DOLBEARE F, GRAY J, T EM: Expres-sion of proliferating cell nuclear antigen (PCNA)/cyclin during the cellcycle. Exp Cell Res 166:209—219, 1986

30. WASEEM NH, LANE DP: Monoclonal antibody analysis of the prolif-erating cell nuclear antigen (PCNA). J Cell Sci 96:121—129, 1990

31. DIJKSTRA CD, DovP EA, JOLING P, KRAAL G: The heterogeneity ofmononuclear phagocytes in lymphoid organs: Distinct macrophage

subpopulations in the rat recognized by monoclonal antibodies ED!,ED2 and ED3. Immunology 54:589—599, 1985

32. SUZUKI K, KATOH R, KAWAOI A: Immunohistochemical demonstra-tion of proliferating cell nuclear antigen (PCNA) in formalin-fixed,paraffin-embedded sections from rat and human tissues. Acta Histo-chem Cytochem 25:13—21, 1992

33. TAMAKI K, OKUDA S, ANDO T, IWAMOTO T, NAKAYAMA M, FUJISHIMAM: TGF-131 in glomerulosclerosis and interstitial fibrosis of adriamy-cm nephropathy. Kidney Int 45:525—536, 1994

34. FURIHATA M, INOUE K, Ol-ITSUKI Y, HASHIMOTO H, TERAO N, FUJITAY: High-risk human papillomavirus infections and overexpression ofp53 protein as prognostic indicators in transitional cell carcinoma ofthe urinary bladder. Cancer Res 53:4823—4827, 1993

35. IWAMURA M, WU G, ABRAHAMSSON PA, DI SANT' AGNESE PA,COCKETI-r AT, DEFTOS U: Parathyroid hormone-related protein isexpressed by prostatic neuroendocrine cells. Urology 43:667—674, 1994

36. NAKANE PK: Simultaneous localization of multiple tissue antigensusing the peroxidase-labeled antibody method: A study on pituitaryglands of the rat. J Histochem Cytochem 16:557—560, 1968

37. VANDESANDE F, DIERICKX K: Identification of the vasopressin pro-ducing and of the oxytocin producing neurons in the hypothalamicmagnocellular neurosecretroy system of the rat. Cell Tissue Res164:153—162, 1975

38. WooD GS, WARNKE R: Suppression of endogenous avidin-bindingactivity in tissues and its relevance to biotin-avidin detection systems.JHistochem Cytochem 29:1196—1204, 1981

39. GOLDBERG A, BARKA T: Acid phosphatase activity in human bloodcells. Nature (Lond) 195:297, 1962

40. TOKUNAGA 0, FAN J, WATANABE T, KOBAYASHI M, KUMAZAKI T,MITSUI Y: Endothelin. Immunohistologic localization in aorta andbiosynthesis by cultured human aortic endothelial cells. Lab Invest67:210—217, 1992

41. HIROTA 5, I-co A, MoRn E, WANAKA A, TOHYAMA M, KITAMURA Y,NOMURA 5: Localization of mRNA for c-kit receptor and its ligand inthe brain of adult rats: An analysis using in situ hybridizationhistochemistry. Mol Brain Res 15:47—54, 1992

42. BAGCHUS W, HOEDEMAEKER P, ROZING J, BAKKER W: Glomerulone-phritis induced by monoclonal anti-thy 1.1 antibodies. Lab Invest55:680—687, 1986

43. FLOEGE J, ENG E, LINDNER V, ALPERS CE, YOUNG BA, REIDY MA,JOHNSON RJ: Rat glomerular mesangial cells synthesize basic fibro-blast growth factor. J Clin Invest 90:2362—2369, 1992

44. JOHNSON RJ, ILDA H, ALPERS CE, MAJESKY MW, SCHWARTZ SM,PRITZI P, GORDON K, GOWN AM: Expression of smooth muscle cellphenotype by rat mesangial cells in immune complex nephritis. J ClinInvest 87:847—858, 1991

45. ISHIZAKI M, MASUDA Y, FUKUDA Y, SUGISAKI Y, YAMANAKA N,MASUGI Y: Experimental mesangioproliferative glomerulonephritis inrats induced by intravenous administration of anti-thymocyte serum.Acta PatholJpn 36:1191—1203, 1986

46. SIM0NSON MS, HERMAN WH: Protein kinase C and protein tyrosinekinase activity contribute to mitogenic signaling by endothelin-1. J BiolChem 268:9347—9357, 1993

47. BADR KF, MUNGER KA, SUGIURA M, SNAJDAR RM, SCHWARTZBERGM, INAGAMI T: High and low affinity binding sites for endothelin oncultured rat glomerular mesangial cells. Biochem Biophys Res Com-mun 161:776—781, 1989

48. SUGIURA M, SNAJDAR RM, SCHWARTZBERG M, BADR KF, INAGAMI T:Identification of two types of specific endothelin receptors in ratmesangial cell. Biochem Biophys Res Commun 162:1396—1401, 1989

49. MARTIN ER, BRENNER BM, BALLERMANN BJ: Heterogeneity of cellsurface endothelin receptors. J Biol Chem 265:14044—14049, 1990

50. OHLSTEIN EH, ARLETH A, BRYAN H, ELLIOYF JD, SUNG CP: Theselective endothelin ETA receptor antagonist B0123 antagonizesendothelin-l-mediated mitogenesis. Eur J Pharmacol 225:347—350,1992

51. EGUCHI S, HIRATA Y, IHARA M, YANG M, MARUMO F: A novel ETAantagonist (BQ-123) inhibits endothelin-1-induced phosphoinositidebreakdown and DNA synthesis in rat vascular smooth muscle cells.FEBS Lett 302:243—246, 1992

52. BENIGNI A, ZOJA C, CORNA D, ORIslo 5, LONGARE-I-I-I L, BERTANI T,REMUZZI G: A specific endothelin subtype A receptor antagonist

Fukuda et al. Mitogenic action of endothelin 1329

protects against injury in renal disease progression. Kidney mt 44:440—444, 1993

53. NAKAMURA T, EBIHARA I, Toiio Y, K0IDE H: Effect of a specificendothelin A receptor antagonist on murine lupus nephritis. Kidneymt 47:481—489, 1995

54. GELLAI M, DEW0LF R, PULLEN M, NAMBI P: Distribution andfunctional role of renal ET receptor subtypes in normotensive andhypertensive rats. Kidney mt 46:1287—1294, 1994

55. EFIRENREICH Fl, ANDERSON RW, Fox CH, RIECKMANN P, HOFFMANGS, TRAVIS WD, COLIGAN JE, KEHRL JH, FAUCI AS: Endothelins,peptides with potent vasoactive properties, are produced by humanmacrophages. J Exp Med 172:1741—1748, 1990

56. ZEIHER AM, GOEBEL H, SCHACFIINGER V, IHLING C: Tissue endothe-lin-1 immunoreactivity in the active coronary atherosclerotic plaque.Circulation 91:941—947, 1995

57. Vx GH, DING G, KEES FD, GROND J, SCHREINER GF, DIAMOND JR:Macrophages and renal disease. Lab Invest 71:456—464, 1994

58. JOHNSON RJ, GARCIA RL, PRITZL P, ALPERS CE: Platelets mediateglomerular cell proliferation in immune complex nephritis induced byanti-mesangial cell antibodies in the rat. Am J Pathol 136:369—374,1990

59. KLAI-IR S, SCHREINER G, ICHIKAWA I: The progression of renal disease.NEnglJMed 318:1657—1666, 1988

60. SIMONSON MS, DUNN MJ: Endothelin peptides: A possible role inglomerular inflammation. Lab Invest 64:1—4, 1991

61. OKUDA S, LANGUINO LR, RUOSLAHTI E, BORDER WA: Elevated

expression of transforming growth factor-a and proteoglycan produc-tion in experimental glomerulonephritis. J C/in Invest 86:453—462,1990

62. IIDA H, SEIFERT R, ALPERS CE, GRONWALD RG, PHILLIPS PE, PRITZLP, GORDON K, GOWN AM, ROSS R, BOWEN PD, JOHNSON RI:Platelet-derived growth factor (PDGF) and PDGF receptor areinduced in mesangial proliferative nephritis in the rat. Proc NatlAcadSci USA 88:6560—6564, 1991

63. ZOJA C, ORIslo S, PERICO N, BENIGN! A, M0RIGI M, BENA-I-I'I L,RAMBALDI A, REMUZZI G: Constitutive expression of endothelin genein cultured human mesangial cells and its modulation by transforminggrowth factor-/3, thrombin, and a thromboxane A2 analogue. LabInvest 64:16—20, 1991

64. KOHNO M, 1IzEDA M, J0HcIII M, HoRlo T, YASUNARI K, KURIHABA N,TAKEDA T: Interaction of PDGF and natriuretic peptides on mesan-gial cell proliferation and endothelin secretion. Am J Physiol 265:E673—E679, 1993

65. KIsHIN0 J, HANASAKI K, KATO T, ARITA H: Endothelin-inducedintracellular Ca2 mobilization through its specific receptors in mu-rifle peritoneal macrophages. FEBS Lett 280:103—106, 1991

66. MAGAZINE HI, ANDERSEN TI', BRUNER CA, MALIK AB: Vascularcontractile potency of endothelin-1 is increased in the presence ofmonocytes or macrophages. Am J Physiol 266:H1620—H1625, 1994

67. SAIJONMAA 0, NYMAN T, FYHRQUIST F: Endothelin-1 stimulates itsown synthesis in human endothelial cells. Biochem Biophys ResCommun 188:286—291, 1992