Renal endothelin system in polycystic kidney disease

J Am Soc Nephrol 9: 1 169-1 177, 1998

Renal Endothelin System in Polycystic Kidney Disease

BERTHOLD HOCHER,*t RUDIGER ZART,*t ANJA SCHWARZ,*t VOLKER VOGT,�

CLAUDE BRAUN,t CHRISTA THONE�REINEKE,t NICOLE BRAUN,t

HANS-HELLMUT NEUMAYER,* �(I�AUS KOPPENHAGEN,� CHRISTIAN BAUER,t

and PETER ROHMEISS�*Department of Nephrology, Charit#{233}, Humboldt University of Berlin; tlnstitute of Molecular Biology and

Biochemistry, Free University of Berlin; �Department of Nephrology, Klinikum Mannheim, University of

Heidelberg; and �Department of Nuclear Medicine, University Hospital Benjamin Franklin, Free University of

Berlin, Germany.

Abstract. Polycystic kidney disease (PKD) is characterized by

interstitial fibrosis and formation of renal cysts. Interestingly,

interstitial fibrosis and renal cyst formation were also seen in

human endothelin- 1 (ET-l) transgenic mice. This study, there-

fore, analyzes the tissue distribution of ET- 1 , the tissue con-

centrations of ET- 1 , as well as the expression of ET receptor

subtypes in the kidneys of a rat model of PKD: Han:SPRD rats.

Six-week-old heterozygous (cy/+) and homozygous (cy/cy),

as well as 6-mo-old heterozygous (cy/+) Han:SPRD rats and

the corresponding age-matched Sprague Dawley littermates

(SD) (+1+) were analyzed. Furthermore, the acute effects of

the mixed (A/B) endothelin receptor antagonist bosentan on

hemodynamic and renal function were investigated in 6-mo-

old, conscious, chronically instrumented (cy/+) rats. The kid-

neys of affected rats showed significantly elevated tissue levels

of ET- 1 compared with age-matched controls (3.5 ± 0.3-fold

in young cy/cy rats, P < 0.01; 1.4 ± 0.2-fold in young cy/+

rats, P < 0.01; 6.2 ± 0.4-fold in old cy/+ rats, P < 0.001) due

to a highly increased ET-1 synthesis within the epithelial cells

of the cysts. Analyzing tissue sections from patients with

typical autosomal dominant PKD demonstrated a high ET-l

expression within the epithelial cells of the cysts as well.

Scatchard analysis revealed a markedly decreased ETA and

ETB receptor density in all groups of affected rats. The acute

blockade of both endothelin receptor subtypes using bosentan

in 6-mo-old heterozygous PKD rats led to a significant de-

crease in mean arterial BP (MAP) (- 19.7 ± 2. 1 mmHg, P <

0.05) and GFR (-41 ± 5%, P < 0.005). Renal blood flow

(RBF) was significantly increased (+2.1 ± 0.5 ml/min, P <

0.05) after bosentan, whereas bosentan had no effect on MAP,

GFR, and RBF in age-matched controls. These data show that

the paracrine renal endothelin system is activated in PKD and

participates in the regulation of MAP, GFR, RBF, and possibly

contributes to renal cyst formation and fibrosis.

Autosomal dominant polycystic kidney disease (ADPKD),

thought to be the most common hereditary kidney disease in

humans, affects approximately 1 in 1000 live births. This

disease accounts for up to I 0% of all patients requiring renal

replacement therapy. Cysts arise from renal tubular segments

as focal areas of dilation. They progressively enlarge with age

and may separate from the nephron of origin.

The Han:SPRD rat strain develops a form of progressive

gender-dependent disease that appears similar in many respects

to that seen in ADPKD in humans. ADPKD in humans as well

as in Han:SPRD rats is characterized by structural alterations

of the kidneys such as thickening of the tubular basement

membrane, interstitial fibrosis, and formation of cysts, leading

to end-stage kidney disease (I). The rat PKD gene was mapped

on the rat chromosome 5, a quantitative trait locus controlling

PKD, kidney mass, and plasma urea concentration. The ho-

Received October 10. 1997. Accepted January 23, 1998.Correspondence to Dr. Berthold Hocher, Universit#{228}tsklinikum Charit#{233}derHumboldt Universit#{228}tzu Berlin, Klinik f#{252}rNephrologie, Schumannstrasse20-21, 10098 Berlin, Germany.

1046-6673/0907- 1 169$03.00/0

Journal of the American Society of Nephrology

Copyright C) 1998 by the American Society of Nephrology

mology region is likely to reside on human chromosome 8. The

gene responsible for PKD in Han:SPRD cy/+ rat is neither the

PKD I gene, localized on the short arm of human chromosome

16 encoding a high molecular weight protein of approximately

500,000 kD named polycystin, nor the PKD2 gene, localized

on human chromosome 4 (2). Despite these recent advances in

the determination of the genetic basis of PKD in humans and

rats, little is known about the cell biology and underlying

mechanisms that contribute to cyst formation in genetically or

chemically induced animal models with renal cysts. Altered

composition of the extracellular matrix (3) is thought to be

implicated in cystopathogenesis. Interstitial fibrosis, glomeru-

losclerosis, and cyst formation were also seen in human endo-

thelin- 1 (ET-l) transgenic mice (4). In addition, the renal

endothelin system seems to play a major role in renal disorders

such as lupus nephritis, impaired renal function after 5/6 ne-

phrectomy, and acute renal failure (reviewed in references S

through 7). Thus, an activated paracrine renal endothelin sys-

tem may play an important role in the pathogenesis of PKD.

We therefore analyzed the expression of ET- I by immuno-

histochemistry, measured tissue concentrations of ET- 1 , and

analyzed the expression of endothelin receptor subtypes by

Scatchard analysis in the kidneys of male Han:SPRD rats

compared with age-matched controls. In addition, the acute

1 170 Journal of the American Society of Nephrology I Am Soc Nephrol 9: 1 169-1 177. 1998

effect of bosentan, a combined ETAIETB receptor antagonist

(8), on renal blood flow (RBF), GFR, heart rate (HR), and BP

was analyzed in conscious chronically instrumented PKD rats.

Furthermore, we analyzed tissue sections from patients with

typical ADPKD and could demonstrate a very high ET-l

expression within the epithelial cells of the cysts.

Materials and MethodsMale 6-wk-old heterozygous (cy/+ ) and homozygous (cy/cy), as

well as 6-mo-old heterozygous (cy/+ ) Han:SPRD rats (9) and the

corresponding age-matched Sprague Dawley rats (SD) (+1+) were

analyzed. The animals (a generous gift from Dr. N. Gretz, Klinikum

Mannheim, Mannheim. Germany) were fed a commercial diet (Al-

tromin#{174},Altromin, Lange. Germany) and given water ad lihitum. All

animal experiments were conducted in accordance with local institu-tional guidelines for the care and use of laboratory animals. [I25I]�

ET-l (2000 Ci/mmol) was obtained from DuPont (Hannover, Germa-ny). Unlabeled ET- 1 was from Peninsula Laboratories (Frankfurt,

Germany). The mixed (A/B) endothelin receptor antagonist bosentan(Ro 47-0203, 4-tert-butyl-N-[6-(2-hydroxy-ethoxy)-5-(2-methoxy-phenoxy)-2,2’-bispyrimidine-4-yl}-benzenesulfonamide) was a gen-

erous gift from Dr. Martine Clozel (Pharma Division, F. Hoffmann-LaRoche Ltd., Basel, Switzerland). The selective endothelin receptorligands (BQ 123 and BQ 3020) were from California Peptides, Inc.

(Napa, CA). The polyclonal rabbit anti-ET-l antibody was from

Peninsula Laboratories. Unless otherwise stated, all reagents were of

analytical grade and were purchased from Merck (Darmstadt, Germa-

ny), Boehringer-Mannheim (Mannheim, Germany), or Sigma (Mu-

nich, Germany).

Phenotypic Determination

The carrier status of each animal was established by determination

of the kidney weight/body weight ratio, typical kidney histology, andserum and urine urea concentrations as recently described by Biho-

reau et al. (2).

ET- 1 Radioiinmunoassay

Tissue Preparation. Kidney tissue-immunoreactive endothelin

levels were measured as recently described (10). After thawing,

kidney samples were suspended in 2 ml � g � wet weight of 0.2 M

CH3COOH, 0.5 M NaCl. disrupted using a polytron and subsequently

homogenized. The homogenate was centrifuged at 4#{176}Cfor 15 mm at

25,000 x g. The supernatant was retained for ET- I RIA and the pelletwas discarded.

Radioimmunoassay. Immunoreactive ET- 1 was extracted fromtissue supernatant using AmprepTM 500 mg C2 columns (Amersham).

One-milliliter samples were acidified with 0.25 ml of 2 M HC1,centrifuged at 10,000 X g for 5 mm at room temperature, and loaded

onto the columns. The columns were washed with 5 ml of water +

0. 1 % trifluoroacetic acid, and the adsorbed peptide was eluted with 2

ml of 80% methanol in water + 0. 1% trifluoroacetic acid. The eluents

were collected in polypropylene tubes and dried under a stream of

nitrogen.

The probes were reconstituted in 250 �tl of assay buffer (0.02 Mborate buffer, pH 7.4, containing 0. 1% sodium azide), and 2 X 100 �l

were taken for analysis in a commercial ET-l [125I1 RIA kit (ET-l,2(high sensitivity) [ ‘ 251]assay system, Amersham). Separation of the

antibody-bound fraction was effected by magnetic separation using

the Amerlex-M Separator (Amersham). This assay reacts 100% with

ET- 1 and cross-reacts I 89% with Big ET- 1 . Cross-reactivity with

ET-3 was less than 0.001.

Scatchard Analysis

Membrane Preparation. Membranes were prepared according

to Nambi et al. ( 1 1 ). The rats were sacrificed and the kidneys were

immediately frozen with liquid nitrogen and stored at -80#{176}C until

further analysis. Approximately 150 mg of kidney tissue was homog-enized at 4#{176}Cin 10 ml of 20 mmollL NaHCO3, using a motor-driven

pestle homogenizer. The homogenate was centrifuged at 4#{176}Cfor 15

mm at 1000 x g. The supernatant was decanted and centrifuged at

4#{176}Cfor 30 mm at 40,000 x g. The pellet, consisting of crude plasmamembranes, was resuspended in assay buffer ( 1 mg/mI bacitracin, 100

mM Tris-HCI, 5 mM MgCl2, and 0. 1 g% bovine serum albumin[BSA], pH 7.4) at a protein concentration of 200 j�g/ml.

Binding Assay for ETA and ET8. Binding studies were per-

formed as described previously (12). To analyze the expression of

endothelin receptor subtypes (ETA, ETB) in the kidney, binding assays

were performed in the presence or absence of the subtype-specific

endothelin receptor ligands BQI23 (3 �tM) and/or BQ3020 (5). The

assay buffer for binding studies contained I mg - ml � bacitracin, 100

mM Tris-HCI, S mM MgCl2, and 0.1 g% BSA, pH 7.4, in a totalvolume of 150 �tl. The [‘251]-ET-I tracer concentration was keptconstant at 40,000 cpmltube, whereas the concentration of unlabeled

ET-l was increased from 0 to 25 nM (competition studies with “coldsaturation”). Samples from crude plasma membranes were used at a

concentration of 0.53 mg of protein . ml�t. Binding studies were

performed at room temperature for I 20 mm. Nonspecific binding was

assessed in the presence of excess ET- 1 (5 p.M). After adding 1 ml of

cold binding buffer, free and receptor-bound radioactivity was sepa-

rated by centrifugation at 30,000 X g (4#{176}C)for I S mm, and the pellets

thus obtained were washed two additional times with I ml of cold

binding buffer. [12511 was counted in a Packard gamma counter (78%

counting efficiency for [125I]).

Im,nunohistochemistrv

Immunohistochemistry for the detection of ET- I in the kidney was

performed with minor modifications, as recently described by Schafer

et al. (3) and Bachmann and Ramasubbu (13), using a polyclonal

rabbit anti-ET- I antibody. Briefly, for antibody incubation, 5-j.tm-

thick cryostat sections were mounted on poly-L-lysine-coated glass

slides. Polyclonal rabbit anti-ET- 1 antibody (Peninsula Laboratories)

was applied at dilutions of I :50 in phosphate-buffered saline contain-

ing I % BSA. Detection of the bound antibody was performed using a

biotinylated second antibody and streptavidin-Texas red (Amersham

Buehler, Braunschweig, Germany). according to the manufacturer’s

instructions. Control experiments were performed omitting the first

antibody and using phosphate-buffered saline instead.

For immunohistochemistry, male 6-wk-old heterozygous (cy/+)and homozygous (cy/cy), as well as 6-mo-old hetetozygous (cy/+)

Han:SPRD rats and the corresponding age-matched Sprague Dawleyrats (SD) (+1+) were analyzed. In a second set of experiments, we

analyzed kidney sections from patients with typical PKD compared

with normal human kidney.

Measurement of GFR, Mean Arterial BP, HR, and RBF

in Conscious, Chronically Instrumented Rats

Surgical Procedures. One week before the acute experiments,

the rats were anesthetized with ether, and femoral artery and vein

catheters were implanted, as recently described ( 14). Three days

before starting the experiments, flow probes ( 1RB with implantable

180

I 60

z 140

:�- 120

�. 100

� 80..e: 60

� 40

20

0

r ‘ < 0.005 -�

77/..

I-’

6 weeks old 6 months old

J Am Soc Nephrol 9: 1 169-1 177, 1998 Renal Endothelin System in PKD 1 171

connector, Transsonic Systems, Inc., Ithaca, NY) were chronically

implanted around the left renal artery. The left kidney was exposed by

a retroperitoneal access. The renal artery was carefully dissected. The

flow probe was then placed around the artery and, after the best signal

had been achieved, the probe was fixed in proper position using a

small envelope of Merocel Op-Wipe#{174} (Merocel Corp., Mystic, CN)

covering the probe and the artery at the point of reflector attachment.

To improve signal conductance, the envelope was filled with ultra-

sound gel. All catheters and cables were led subcutaneously to the

rat’s neck (14).

Circulatory Measurements. Mean arterial BP (MAP) and HR

were measured via the arterial line with a Statham pressure transducer

P23Db and a Gould pressure processor coupled to a Gould Brush

2600 recorder. RBF was measured via the chronically implanted flow

probe with a transit time flowmeter (Tl06, Transsonic Systems, Inc.)

and continuously recorded on a Gould Brush 2600 recorder. The

Transsonic flowmeter system determines absolute volume flow (15).

The flow probes were precalibrated, and measured absolute blood

flow with an accuracy of ±2%.Measurement of GFR. GFR was measured using the inulin

single-shot method (16). The single-shot clearance was evaluated by

displaying a two-compartment model with resolution of the plasma

inulin concentrations into two monoexponential functions. The rats

received an intravenous bolus injection of 150 mg of inulin (Inutest#{174}).

Blood samples for determination of serum inulin concentrations weredrawn at 0, 15, 30, 90, 135, and 180 mm after injection. Inulin was

measured by a modified f3-fructosidase method (17). GFR is ex-

pressed as ml/min per 100 g body wt.

Effects of the Mixed (A/B) Endothelin Receptor Antagonist

Bosentan on MAP, HR, and RBF. To examine the endogenous

endothelin system in PKD rats, the effect of bosentan on resting MAP,

HR, and RBF in 6-mo-old heterozygous (cy/+) Han:SPRD (PKD)

rats and the corresponding age-matched Sprague Dawley littermates

(SD) was analyzed. The rats were divided into four groups: Group I

(SD; n = 6) and group 2 (PKD: n = 7) received cumulative intrave-

nous bolus injections of bosentan (10 mg/kg) every 15 mm up to a

total load of 100 mg/kg. Group 3 (SD; n = 6) and group 4 (PKD; n =

6) received intravenous bolus injections of vehicle.

Effects of the Mixed (A/B) Endothelin Receptor AntagonistBosentan on GFR. To analyze the endogenous renal endothelin

system in PKD rats, we tested the effect of bosentan on GFR in

6-mo-old heterozygous (cy/+) Han:SPRD (PKD) rats and the corre-sponding age-matched Sprague Dawley littermates (SD). The rats

were divided into four groups: Group I (SD, n = 6) and group 2

(PKD, n = 6) received a single injection of 100 mg/kg bosentan. and

group 3 (SD. n = 7) and group 4 (PKD. n = 7) received a single

injection of placebo, followed by an injection of 150 mg of inulin S

mm later. At 0, 15, 30, 90, 135, and 180 mm after injection of inulin,

blood samples (200 .d) were taken from the arterial catheter for

determination of serum inulin concentrations.

Effects of Exogenously Applied ET-1 on BP and RBF. Incontrast to the above-described set of experiments analyzing the

impact of the endogenous renal endothelin system in PKD rats, the

following experiments were performed to analyze the effect of exog-

enously added ET-l on BP and RBF. We tested the hemodynamic

responses to increasing doses of ET-l (20, 50, 100, and 200 ng of

ET-l , intravenously) in 6-mo-old heterozygous (cyl+) Han:SPRD(PKD) rats and corresponding age-matched Sprague Dawley litter-

mates. To avoid a possible tachyphylaxis after repeated injections of

ET- 1 , the rats received in independent experiments intravenous bolus

injections of ET-l.

Statistical Analyses

The unpaired t test was used for determination of statistical differ-

ence of group means. ANOVA followed by t test was used if appro-

priate. Results were considered statistically significant at a value of

P < 0.05.

ResultsIn 6-wk-old as well as in 6-mo-old rats with PKD, the

kidneys of affected rats showed significantly elevated tissue

concentrations of ET- I compared with age-matched litter-

mates. In 6-wk-old PKD rats, we observed a gene-dose effect,

because tissue ET- I concentrations were much higher in ho-

mozygous (cy/cy) Han:SPRD rats compared with heterozygous

(cy/+) Han:SPRD rats (Figure 1).

Immunohistochemical analysis of 6-wk-old and 6-mo-old

rats (Figure 2) with PKD revealed that the increased tissue

ET- 1 concentrations as described above are due to ET- I ex-

pression mainly within the epithelial cells of the cysts (Figure

2). A weaker specific signal was obtained within the interstitial

renal tissue, blood vessels, and glomeruli of homozygous (cy/

cy) and heterozygous (cy/+) Han:SPRD rats.

However, there are some differences with respect to the

ET- 1 fluorescence pattern of ET- 1 in 6-wk-old homozygous

and 6-mo-old heterozygous Han:SPRD rats. ET- 1 staining of

the epithelial cells of the cysts is stronger in 6-mo-old het-

erozygous PKD rats compared with 6-wk-old homozygous

PKD rats. ET-1 staining of the interstitial tissue, on the other

hand, appears stronger in homozygous PDK rats compared

with heterozygous PKD rats. Control rats (6-wk-old as well as

6-mo-old littermates) had a very low renal ET- I tissue immu-

noreactivity near the detection limit of the method used in this

study (Figure 2C).

Interestingly, a similar pattern of ET- I immunoreactivity

within the kidney was seen in kidney sections from patients

with ADPKD (Figure 3). Kidneys were removed in these

patients due to the size of the polycystic kidneys. We analyzed

sections from four different patients (38 to 56 yr old; three

men, one woman) with classic ADPKD compared with normal

E�I Sprague-Dawley littermates� PKD heterozygousE’.� PKD homozygous

Figure 1. Bar graph of renal tissue endothelin- I (ET- 1 ) concentrations

in 6-wk-old and 6-mo-old rats with polycystic kidney disease (PKD)and the corresponding age-matched littermates are shown. Data are

means ± SEM.

1 172 Journal of the American Society of Nephrology J Am Soc Nephrol 9: 1 169-1 177, 1998

� �-,�

�%

a �

-� ;�

. r

. .�

,� ) -

, �-.1’ � � 4J(.P4.

Figure 2. (A) Kidney section from a 6-mo-old male heterozygous PKD rat (cy/+) (Hematoxylin and eosin IH&E] staining). (B) Immuno-

histochemical staining using an ET-l antibody showing a highly increased ET-l expression within the epithelial cells of renal cysts in the

kidneys of a 6-mo-old male heterozygous PKD rat. (C) The corresponding age-matched littermate also stained with an ET-1 antibody showed

only a very weak fluorescence signal within the tubules, blood vessels, and glomeruli. (D) Kidney section from a 6-wk-old male homozygous

PKD rat (cy/cy) (H&E staining). (E and F) Immunohistochemical staining using an ET- I antibody showing an increased ET- I expression within

the epithelial cells of renal cysts but also of the interstitial tissue in the kidneys of a 6-wk-old male homozygous PKD rat. ET- 1 staining of the

epithelial cells of the cysts is stronger in 6-mo-old heterozygous PKD rats (B) compared with 6-wk-old homozygous PKD rats (E), whereas

staining of the interstitial tissue appears stronger in homozygous PDK rats (E and F) compared with heterozygous PKD rats (B).

human kidney tissue from nephrectomies due to kidney cancer. interstitial tissue also showed a specific ET- I fluorescence

We always detect a very high ET-l immunoreactivity in the signal. Normal kidney tissue from the nephrectomies due to

epithelial cells of the renal cysts of ADPKD patients. Again, kidney cancer showed only a very weak signal.

1000

800

600

400

200

0

800

600

400

200

06 weeks old 6 months old

Sprague-Dawley littermatesPKD heterozygous

PKD homozygous

Figure 4. Bar graphs showing the density of ETA and ETH receptors

in the kidneys of 6-wk-old as well as 6-mo-old rats with PKD and the

corresponding age-matched littermates. Data are means ± SEM.

J Am Soc Nephrol 9: 1 169-1 177. 1998 Renal Endothelin System in PKD I 173

Figure 3. (A) Kidney section from a 39-yr-old patient with typicalautosomal dominant PKD (H&E staining). (B and C) Immunohisto-

chemical staining using an ET-l antibody showing a highly increased

ET- 1 expression within the epithelial cells of renal cysts in thekidneys of this patient. Kidneys were removed due to the size of the

polycystic kidneys.

Scatchard analysis, on the other hand, revealed a markedly

decreased ETA, as well as ETB receptor density (Bmax), �fl

6-wk-old and 6-mo-old affected PKD rats (Figure 4). In this

case, an inverse gene-dose effect was observed. The receptor

density was much more reduced in homozygous (cy/cy) Han:

a,0

0.a)E

0

E

C

SPRD rats compared with heterozygous (cy/+ ) Han:SPRD rats

(Figure 4).

The binding affinity of the ETA and the ETB receptor was

slightly reduced in 6-wk-old PKD rats compared with control

rats (Table 1), whereas the binding affinity in 6-mo-old PKD

rats was only reduced for the ETB receptor compared with

controls (Table 1 ). ET- 1 tissue concentrations and the expres-

sion of endothelin receptor subtypes in 6-mo-old homozygous

(cy/cy) Han:SPRD rats could not be determined, because ho-

mozygous (cy/cy) Han:SPRD rats usually died at the age of

approximately 10 wk due to end-stage renal disease.

Measurement of GFR, MAP, HR, and RBF in conscious,

chronically instrumented rats was performed only in the 6-mo-

old rats, because the methods used in our study are not suitable

for very small animals such as 6-wk-old rats. Measurement of

MAP, HR. and RBF was started 90 mm before the functional

experiments were performed and was completed 90 mm afterthe last injection of bosentan. Mean basal MAP before the

functional experiments were begun was 1 18.4 ± 9.3 mmHg in

Han:SPRD (cy/+) rats and 109.8 ± 8.7 mmHg in the corre-

sponding littermates. The basal MAP differences between Han:

SPRD (cy/+) rats and the corresponding controls were not

significant. The acute blockade of both endothelin receptor

subtypes using bosentan (10 mg/kg intravenously every 15 mm

for 2.5 h up to a total load of 100 mg/kg) in 6-mo-old het-

erozygous PKD rats led to a significant decrease in MAP

(Figure 5). No significant effect of bosentan on BP was seen in

age-matched Sprague Dawley littermates, and no effect of

bosentan was seen in the control group (Figure 5). RBF,

140 -

130 -

x -

E.� 110 -

0�

< 100 -

90 -

80 -

25 -

20 -*

a)

� -

CE

E

U-

0-

I 174 Journal of the American Society of Nephrology J Am Soc Nephrol 9: 1 169-1 177. 1998

Table I. ETA and ETB receptor binding affinities (Kd) in the kidneys of 6-wk-old and 6-mo-old Han:SPRD and

corresponding control rats�’

Category

6-wk-old 6-mo-old

Control SDRats (n = 6)

HeterozygousPKD Rats

(P1 = 6)

HomozygousPKD Rats

(ii = S to 6)

Control SDRats (ii = 6)

HeterozygousPKD Rats

(ii = 6)

ETA-receptor binding affinity

(nmol/L)

ETB-receptor binding affinity

(nmol/L)

0.24 ± 0.08

0.3 1 ± 0.0 I

0.60 ± 001b

0.68 ± 002b

oso ± 005b

0.66 ± 0.03k’

o.so ± 0.34

0.48 ± 0.05

0.70 ± 0.32

1 .50 ± 009L�

U Values are mean ± SEM. SD, Sprague Dawley; PKD, polycystic kidney disease.b p < 0.01 compared with age-matched control rats (Sprague Dawley littermates).

however, was significantly increased in 6-mo-old heterozygous

PKD rats. HR was not affected by bosentan in PKD or in

Sprague Dawley rats (data not shown).

GFR was markedly reduced in 6-mo-old heterozygous PKD

rats compared with control rats. A single bolus injection of

bosentan led to a further significant (P < 0.005) decrease of

GFR in PKD rats only (0.42 ± 0.06 ml/min per 100 g body wt

in PKD rats treated with placebo and 0.29 ± 0.06 mI/mm per

100 g body wt in PKD rats treated with bosentan) (Figure 5).

Intravenous bolus injections of 20, 50, 100, and 200 ng of

ET- 1 produced a dose-dependent biphasic blood pressor re-

sponse: An initial, short-lasting depressor effect was followed

by a long-lasting BP elevation. The changes in BP were ac-

companied by reciprocal alterations in HR. The observed ef-

fects of ET- I -induced alterations of the pressor as well as the

depressor response were dose-dependent, and qualitatively and

quantitatively similar in 6-mo-old rats with PKD and cone-

sponding controls. RBF showed a dose-dependent monophasic

response pattern to exogenously applied ET- 1 . RBF decreased

to a similar extent in the Han:SPRD (cy/+) rats and the

corresponding controls (data not shown).

DiscussionThe paracrine endothelin system is highly activated in the

kidneys of Han:SPRD-PKD rats. Renal tissue concentrations

of ET- 1 were 3.5 ± 0.3-fold increased in young cy/cy rats and

even more in old cy/+ rats (6.2 ± 0.4-fold). This finding is

probably due to a highly increased ET- 1 synthesis within the

epithelial cells of the cysts in Han:SPRD-PKD rats, as shown

by immunohistochemistry. Patients with ADPKD are also

characterized by a highly increased ET- 1 synthesis within the

epithelial cells of the kidney cysts.

Endothelin System in PKD Compared with Other

Models of Chronic Renal Failure with Involvement ofthe Renal Endothelin System

It is important to note that the increase in renal tissue ET-1

immunoreactivity is much higher in rats with PKD than in any

other animal model of chronic kidney disease reported thus far

(4-7,10,12). ET-l transgenic mice and ET-2 transgenic rats,

C

.g 15-

E

� 10-

0-�

3-

I I Vehicle

I:��1 Bosentan

Sprague-Dawley littermates PKD (cy/+)

Figure 5. Effects of intravenous administration of 100 mg/kg bosentan(�) or placebo (El) on mean arterial BP (MAP; mmHg), renal bloodflow (RBF; mL/min), and GFR (ml/min per 100 g body wt) in

6-mo-old conscious, chronically instrumented (cy/+) Han:SPRD rats

and corresponding control rats (Sprague Dawley littermates). Data aremeans ± SEM. *JJ < 0.05; **� < 0.01.

J Am Soc Nephrol 9: 1 169-1 177. 1998 Renal Endothelin System in PKD I I 75

for example, are characterized by the development of a patho-

logic renal phenotype (4, 1 0). This pathology developed in spite

of only slightly elevated tissue ET-1 or ET-2 concentrations.

Impaired renal function in rats with liver cirrhosis (12) is also

associated with a significant but mild increase in renal tissue

ET- I concentrations. However, the pathophysiologic relevance

of these findings in cirrhotic rats is clearly demonstrated,

because the blockade of the endothelin system in these cir-

rhotic animals with bosentan resulted in a decreased water

excretion and increased formation of ascites (12). Thus, our

data strongly suggest that the stronger activation of the renal

endothelin system-compared with the above-mentioned

pathophysiologic conditions-in PKD (Han:SPRD rats as well

as humans) may play a major role in the pathogenesis of PKD.

The finding of an increased ET- 1 tissue concentration in PKD

is in agreement with a recent report ( 18) showing increased

ET- 1 mRNA expression in the kidneys of a mouse PKD model

(cpk mice). The role of the renal endothelin system in human

PKD is supported by the finding of an inverse relationship

between ET- I concentrations and sodium concentration in the

cyst fluid of patients with ADPKD (19).

Renal Endothelin System as a Cofactor in the

Development of Renal Cysts

Overexpression of the human ET-l gene in the kidneys of

male ET- 1 transgenic NMRI mice promoted renal cyst growth

(4). Nontransgenic NMRI mice develop only a small number

of small renal cysts. Therefore, we propose that primary ge-

netic alterations such as mutations in the PKD1 or PKD2 gene

in humans or mutations within a yet unknown gene on rat

chromosome 5, leading to the development of PKD in rats (2),

might cause a downstream secondary activation of the renal

endothelin system, thus promoting growth and formation of

renal cysts. The hypothesis that ET- 1 seems to be an important

cofactor in the pathogenesis of renal cysts requiring additional

primary genetic or environmental stimuli is supported by: (1)

the finding that ET-2 transgenic rats (10) and their correspond-

ing nontransgenic littermates (rats without genetic predisposi-

tion for renal cysts) did not develop renal cysts (B. Hocher,

unpublished observation); and (2) the finding that patients with

typical ADPKD are also characterized by a highly increased

ET- I synthesis within the epithelial cells of the cysts (Figure

3), as seen in PKD rats (Figures 1 and 2).

The finding of an activated renal endothelin system in pa-

tients with ADPKD is of clinical relevance, because the ther-

apeutic strategies currently available are limited (e.g. , lowering

of BP and low protein diet).

Kidney cysts in PKD arise from renal tubular segments as

focal areas of dilation. Tubular epithelial cells mainly express

ETB receptors (10, 12,20), and the ETB receptor is involved in

the growth of tubular cells, as demonstrated recently by Ong et

a!. (2 1 ). Thus, ETB blockade may reduce cyst formation in

PKD.

However, there have been no studies analyzing long-term

effects of ETA Of ETB receptor blockade in humans or in

animal models of PKD that might prove the hypothesis that the

renal endothelin system is a major progression factor of

chronic renal failure in PKD, as suggested by the present study

and the findings in ET-l transgenic mice (4).

Downregulation of Endothelin Receptors in

Han:SPRD RatsThe data presented in this study suggest that the highly

increased ET-1 synthesis in Han:SPRD rats resulted in a reac-

tive downregulation of the receptor density of both endothelin

receptor subtypes. These results are in agreement with recent in

vitro experiments showing a downregulation of endothelin

receptors in response to increased autocrine production of ET- I

(22).

Furthermore, the binding constants of both endothelin re-

ceptor subtypes in the kidney of PKD rats are, with the excep-

tion of the ETA receptor in 6-mo-old PKD rats, approximately

two times higher compared with Sprague Dawley littermates.

Thus far there are no reports showing that such a slight alter-

ation of the binding affinity is of pathophysiologic relevance.

Posttranslational structural alterations of endothelin receptors

(phosphorylation or glycosylation) may explain these findings.

N-glycosylation sites were identified in the outer cell domain

of both endothelin receptor subtypes (23,24). In cpk mice with

PKD, however, a yet unknown factor/mechanism increases

both the expression of the ET- 1 mRNA and the expression of

the ETA and ETB receptor mRNA (18).

To analyze the pathophysiologic consequences of a simul-

taneous upregulation of tissue ET- I and downregulation of

both endothelin receptor subtypes in Han:SPRD (cy/+ ) rats

(Figures 1 and 4), we blocked the endogenous endothelin

system using bosentan. These experiments demonstrate that the

renal endothelin system in Han:SPRD (cyl+ ) rats is involved

in the regulation of BP, GFR, and RBF despite the downregu-

lation of endothelin receptors, because MAP, GFR, and RBF

were significantly altered after acute blockade of both endo-

thelin receptors in PKD rats only (Figure 5). In the correspond-

ing Sprague Dawley littermates, none of these parameters was

significantly modified. In addition, the response to increasing

doses of exogenous/v applied ET-1 was similar in PKD and

control rats.

Both the blockade of the endogenous endothelin system

using bosentan and the results after exogenous application of

ET- 1 indicate that the downregulation of endothelin receptors

in PKD rats does not reduce or even abolish the biological

effects of the high endogenous renal ET- I concentrations and

of exogenously applied ET-l in PKD rats. Thus, postreceptor

mechanisms are obviously counteracting the downregulation of

the endothelin receptors in rats with PKD.

Effects of Bosentan on BP, RBF, and GFR

A major finding after acute blockade of the highly activated

endogenous endothelin system in PKD rats using bosentan was

the reduction of BP (- I 9.7 ± 2. 1 mmHg), whereas the same

dose of bosentan had no significant effect on MAP in Sprague

Dawley littermates (Figure 5). These data indicate that the

paracrine endothelin system in Han:SPRD (cy/+) rats contrib-

utes substantially to the regulation of BP in PKD rats. The

BP-lowering effect of the combined ETA/ETB receptor antag-

1 176 Journal of the American Society of Nephrology I Am Soc Nephrol 9: 1 169-1 177. 1998

onist in Han:SPRD (cy/+ ) rats is remarkable, because an

activated endothelin system causes or contributes in general to

structural alterations in cardiovascular (25) and kidney tissue

(e.g., glomerulosclerosis and interstitial fibrosis), but does not

affect BP as seen in ET- 1 transgenic mice (4) and ET-2

transgenic rats (10). The much higher ET-l tissue concentra-

tions in PKD rats compared with the above-mentioned trans-

genie animal models of the endothelin system (4, 10) may

explain the additional hemodynamic effects of the endothelin

system in PKD rats. Using immunohistologic techniques, we

could demonstrate that the major sites of ET- 1 expression in

6-mo-old PKD rats are the epithelial cells of the renal cysts.

We propose that the ET-l synthesized in the epithelial cells of

the cysts migrates/diffuses in a passive manner to the intersti-

tial tissue, blood vessels, and glomeruli, thus contributing to

the regulation of BP and GFR.

GFR and RBF were also significantly altered after acute

blockade of the renal endothelin system using bosentan in PKD

rats. However, these alterations were less pronounced corn-

pared with the BP-lowering effect of bosentan in PKD rats.

The reduction of MAP after a single injection of bosentan in

PKD rats may explain the bosentan-induced reduction of GFR,

probably by reducing BP in PKD rats below the setpoint of

GFR autoregulation.

In conclusion, the present study shows that the paracrine

renal endothelin system is activated in rats with PKD, may

contribute to renal cyst formation and renal fibrosis, and is also

involved in the regulation of BP, GFR, and RBF in PKD. The

finding that patients with ADPKD are also characterized by a

highly increased ET- 1 synthesis within the epithelial cells of

the kidney cysts suggests that the renal endothelin system

might be involved in human ADPKD as well.

AcknowledgmentsThis study was supported by grants from the Deutsche Forschungs-

gemeinschaft (Ho 1665/2-I), Fonds der Chemischen Industrie (to Dr.Hocher), and Zentrum f#{252}rMedizinische Forschung, Mannheim (to Dr.Rohmeiss). The technical assistance of 0. Chung and S. Schiller isgreatly appreciated.

References1 . Grantham ii: The etiology, pathogenesis, and treatment of auto-

somal dominant polycystic kidney disease: Recent advances.Am J Kidney Dis 28: 788-803, 1996

2. Bihoreau MT. Ceccherini I, Browne I, Kranzlin B, Romeo G,

Lathrop GM, James MR. Gretz N: Location of the first geneticlocus, PKDrI, controlling autosomal dominant polycystic kidneydisease in Han:SPRD cy/+ rat. Hum Mol Genet 6: 609-613,1997

3. Schafer K, Bader M, Gretz N, Oberb#{228}umer I, Bachmann 5: Focal

overexpression of collagen 4 characterizes the initiation of epi-

thelial changes in polycystic kidney disease. Exp Nephrol 2:

190-195, 1994

4. Hocher B, Thdne-Reineke C, Rohmeiss P, Schmager F, Slowin-

ski T, Burst V. Siegmund F, Quertermous T, Neumayer HH,

Bauer C, Schleuning WD, Theuring F: Endothelin-l transgenic

mice develop renal cysts, interstitial fibrosis and glomeruloscle-

rosis but not hypertension. J Cli,z invest 99: 1380-1389, 1997S. Benigni A: Defining the role of endothelins in renal pathophys-

iology on the basis of selective and unselective endothelin re-

ceptor antagonist studies. Curr Opin Nephrol Hypertens 4: 349-

353, 1995

6. Benigni A, Zoja C, Remuzzi G: The renal toxicity of sustainedglomerular protein traffic. Lab Invest 73: 461-468, 1995

7. Remuzzi G, Benigni A: Progression of proteinuric diabetec and

nondiabetic renal disease: A possible role for renal endothelin.

Kidney In! Suppl 58: S66-S68, 1997

8. Clozel M, Breu V, Gray GA, Kalna B, L#{246}ffierBM, Burn K,Cassal iM, Hirth G, Muller M, Neidhart W, Ramuz H: Pharma-

cological. characterization of bosentan, a new potent orally ac-

tive non-peptide endothelin receptor antagonist. J Pharmacol

Exp Ther 270: 228-235, 1994

9. Gretz N, Ceccherini I, Kranzlin B, Kloting I, Devoto M, Rohm-

eiss P. Hocher B, Waldherr R, Romeo G: Gender-dependent

disease severity in autosomal polycystic kidney disease of rats.

Kidney mt 48: 496-500, 1995

10. Hocher B, Liefeldt L, Th#{246}ne-Reineke C, Orzekowski HD, Distler

A, Bauer C, Paul M: Characterization of the renal phenotype of

transgenic rats expressing the human endothelin-2 gene. Hyper-

tension 28: 196-201, 1996

1 1 . Nambi P, Pullen M, Contino LC: Upregulation of renal endothe-

lin receptors in rats with cyclosporine A-induced nephrotoxicity.

Eur J Pharmacol 187: 1 13-1 16, 1990

12. Hocher B, Zart R, Diekmann F, Rohmeiss P, Distler A, Neu-

mayer HH, Bauer C, Gross P: Paracrine renal endothelin system

in rats with liver cirrhosis. Br J Pharmacol I 18: 220-227, 1996

1 3. Bachmann 5, Ramasubbu K: Immunohistochemical

tion of the alpha-subunit of neutrophil NADPH oxidase and

ecto-S’-nucleotidase in kidney and liver. Kidney mt S 1 : 479-

482, 1997

14. Hocher B, Rohmeiss P. Zart R, Diekmann F, Vogt V. Metz D,

Fakhury M, Koppenhagen K, Gretz N, Bauer C, Neumayer HH,

Distler A: Function and expression of endothelin receptor sub-

types in the kidneys of spontaneously hypertensive rats. C’ardio-

vase Res 31: 499-510, 1996

I S. Drost C: Vessel-diameter independent volume flow measure-ments using ultrasound. Proc San Diego Biomed Symp 17: 299-

302, 1978

16. Gretz N, Ecker-Tschirner KH, KUhnle HF, Dahl KV, KirschfinkM, Drescher P, Lasser ii, Strauch M: Practicability of the inulinplasma single-shot clearance. Contrib Nephrol 8 1 : 220-228,

I 990

17. Ktihnle HF, Dahl D, Schmidt FH: Fully enzymatic determinationin small volume samples without deproteinization. Nephron 62:

104-107, 1992

18. Nakamura T, Ebihara I, Fukui M, Osada 5, Tomino Y, Masaki T,

Goto K, Furuichi Y, Koide H: Increased endothelin and endo-thelin receptor mRNA expression in polycystic kidneys of cpk

mice. J Am Soc Nephrol 4: 1064-1072, 1993

19. Munemura C, Uemasu J, Kawasaki H: Epidermal growth factor

and endothelin in cyst fluid from autosomal dominant polycystickidney disease cases: Possible evidence of heterogeneity in cys-togenesis. Am J Kidney Dis 24: 561-568, 1994

20. Hocher B, Rohmeiss P, Zart R, Diekmann F, Braun C, Kop-

penhagen K, Bauer C, Distler A, Gretz N: Distribution ofendothelin receptor subtypes in the rat kidney: Renal and

haemodynamic effects of the mixed (A/B) endothelin receptor

I Am Soc Nephrol 9: 1 169-1 177. 1998

antagonist bosentan. Eur J Clin Che,n Clin Biochem 33:

463-472, 1995

21 . Ong ACM, Jowett TP, Firth JD, Burton 5, Karet FE, Fine LG:

An endothelin-1 mediated autocrine growth loop involved in

human renal tubular regeneration. Kidney Ira’ 48: 390-401, 1995

22. Clozel M, Ldffler BM, Breu V, Hilfiger L, Maire JP, Butscha B:

Downregulation of endothelin receptors by autocrine production

of endothelin-1 . Am J Physiol 265: Cl88-C192, 1993

23. Arai H, Hori 5, Aramori L, Ohkubo H, Nakanishi 5: Cloning and

Renal Endothelin System in PKD 1 177

expression of a eDNA encoding an endothelin receptor. Nature

348: 730-732, 1990

24. Haendler B, Hechler U, Schleuning WD: Molecular cloning of

human endothelin (ET) receptors ETA and ETB. J CardiovascPharmacol 20[Suppl 12]: 51-54, 1992

25. Hocher B, Thdne-Reineke C, Bauer C, Raschack M, NeumayerHH: The paracrine endothelin system: Pathophysiology and im-plications in clinical medicine. Eur J Clin Chem Clin Biochem

35: 175-189, 1997