endothelin release by rabbit proximal tubule cells: modulatory effects of cyclosporine a,...

VASCULAR BIOLOGY - HEMODYNAMICS - HYPERTENSION

Endothelin release by rabbit proximal tubule cells: Modulatoryeffects of cyclosporine A, tacrolimus, HGF and EGF

CORNELIA HAUG, CHRISTIAN GRILL, ALEXANDRA SCHMID-KOTSAS, ADOLF GRUENERT,and PETER MICHAEL JEHLE

Institute of Clinical Chemistry and Department of Internal Medicine II, Division of Nephrology, University Hospital Ulm, Ulm, Germany

Endothelin release by rabbit proximal tubule cells: Modulatoryeffects of cyclosporine A, tacrolimus, HGF and EGF.

Background. Previous studies have suggested that endothelins, afamily of 21 amino acid peptides with potent vasoconstrictive andmitogenic properties, are involved in the pathogenesis of acuteand chronic renal failure. In addition, endothelin seems to play animportant role in mediating the nephrotoxic side effects ofcyclosporine A (CsA) and tacrolimus. The present study investi-gated the production of endothelin-1 (ET-1) and endothelin-3(ET-3) by bipolar differentiated rabbit proximal tubule cells (PT-1cells), and the modulatory effect of CsA, tacrolimus, hepatocytegrowth factor (HGF) and epidermal growth factor (EGF) onET-1 and ET-3 release.

Methods. ET-1 and ET-3 mRNA was detected by RT-PCR,immunoreactive endothelin was localized to PT-1 cells by immu-nofluorescence staining with antibodies against ET-1 and ET-3.ET-1 and ET-3 release into the culture medium was determinedby specific radioimmunoassays after solid phase extraction.

Results. PT-1 cells exhibited a time-dependent increase of ET-1release up to an incubation period of 36 hours, whereas ET-3release already reached a steady state level after four hours. PT-1cells, cultured on filter membranes, released a significantly higheramount of immunoreactive ET-1 into the basolateral compart-ment than into the apical compartment. ET-3 release did notdiffer significantly between the basolateral and the apical com-partment. Supplementation of the cell culture medium with 10%fetal calf serum induced a marked increase of the basolateral andapical ET-1 release, whereas ET-3 release was only slightlyincreased. CsA and tacrolimus (0.5 to 5000 mg/liter) induced asignificant, dose-dependent increase of ET-1 and ET-3 release byPT-1 cells with a maximum stimulation at a CsA concentration of500 mg/liter (P , 0.001) and a tacrolimus concentration of 50mg/liter (P , 0.001). HGF and EGF (10210 to 1028 mol/liter)exerted a significant (P , 0.001) dose-dependent inhibitory effecton ET-1 release, whereas ET-3 release was not significantlyreduced. Coincubation of PT-1 cells with CsA or tacrolimus andHGF or EGF also resulted in a marked reduction of ET-1 release.

Conclusions. The present data suggest that ET-1 and ET-3release by cultured rabbit proximal tubule cells are regulated

differently, and that the stimulatory effect of CsA and tacrolimuson ET-1 release is antagonized by HGF and EGF.

The endothelins are a family of peptides with potentvasoactive and mitogenic properties. Endothelin-1 wasoriginally isolated from the supernatant of cultured porcineaortic endothelial cells [1]. To date, three isoforms havebeen identified, endothelin-1, -2 and -3 (ET-1, ET-2, ET-3),and they are encoded by three separate genes [2]. Endo-thelial cells seem to represent the main site of endothelinproduction, however, endothelin synthesis has also beenlocalized to other cell types, including renal mesangial cells[3], proximal tubule cells and collecting duct cells [4–6].Endothelial cells have been shown to secrete ET-1 predom-inantly into the basolateral compartment [7], and recentlypublished data suggest that renal proximal tubule andcollecting duct cells also exhibit a polar ET-1 secretion [8,9]. In the present study, we investigated the basolateral andapical ET-1 and ET-3 release by rabbit proximal tubulecells (PT-1 cells).

Several studies have suggested an upregulation of endo-thelin production in acute and chronic renal failure [10–12]. An enhanced endothelin production seems also to beinvolved in mediating the nephrotoxic side effects of cyclo-sporine [13–15]. Recently published studies have demon-strated a stimulatory effect of cyclosporine A (CsA) onET-1 production by cultured endothelial cells [16], culturedmesangium cells [17, 18] and renal cortical epithelial cells[19]. The present study investigated the effects of CsA andthe newer immunosuppressive agent tacrolimus on theET-1 and ET-3 release by cultured rabbit proximal tubulecells. In addition, we investigated whether the basal and theCsA- and tacrolimus-induced endothelin release is modu-lated by renotropic factors, such as hepatocyte growthfactor (HGF) and epidermal growth factor (EGF), whichhave been shown to stimulate renal tubular cell prolifera-tion and to accelerate renal regeneration after drug-in-duced and ischemia-induced renal injury [20–24].

Key words: vasoconstriction, mitogenesis, chronic renal failure, nephro-toxicity, CsA, hepatocyte growth factor, epidermal growth factor.

Received for publication February 24, 1998and in revised form May 21, 1998Accepted for publication June 5, 1998

© 1998 by the International Society of Nephrology

Kidney International, Vol. 54 (1998), pp. 1626–1636

1626

METHODS

Culture of rabbit proximal tubule cells

Rabbit proximal tubule (PT-1) cells were originally de-rived from microdissected proximal tubule segments ofnormal rabbits and immunoselected into a homogeneouspopulation. Cells were transfected with plasmid pSV2-neoDNA (which contains the bacterial gene neo thatconfers resistance to the toxic antibiotic G418) plus DNAfrom the early region of SV40, containing the immortaliz-ing large T antigen. PT-1 cells exhibit multiple morpholog-ical and functional characteristics of differentiated proxi-mal tubule epithelial cells. PT-1 cells stain positively forcytokeratin and alkaline phosphatase, include a sodium-coupled hexose transport system, and have apical microvilliand lateral tight junctions as described previously [25, 26].PT-1 cells were cultured in culture flasks and six well platesin a humidified incubator (37°C, 95% air, 5% CO2).Maintenance medium contained DMEM/HAM’S F-12(1:1, Biochrom, Berlin, Germany), supplemented with ge-neticin (G418, 1 mg/ml), penicillin (500 U/ml), streptomy-cin (500 mg/ml), and 10% fetal calf serum (FCS).

Determination of endothelin-1 and endothelin-3 mRNA

ET-1 and ET-3 mRNA expression was investigated inPT-1 cells, maintained in culture medium containing 5%FCS. RNA extraction was performed using a High PureRNA Isolation Kit (Boehringer Mannheim, Mannheim,Germany) according to the manufacturer’s instructions.The concentration and purity of RNA was determined bymeasuring the absorbance at 260 nm and 280 nm. AfterDNA digestion (DNase I Amplification Grade, RNase-free; Gibco, Eggenstein, Germany), reverse transcriptionpolymerase chain reaction (RT-PCR) was performed withthe Titany One Tube RT-PCR System (Boehringer Mann-heim). Specific primers were synthesized (Interactiva, Ulm,Germany) according to known prepro-ET-1 and ET-3cDNA sequences [27, 28]. Prepro-ET-1 primer 1 (anti-sense) was defined by bases 621 to 640, sequence 59-CTGTTGCTGATGGCCTCCAA-39, primer 2 (sense) en-compassed bases 201 to 220, sequence 59-ACAGGAGCG-GAAAACCCAGG-39, ET-3 primer 1 (anti-sense) com-prised bases 835 to 855, sequence 59-TGTTGATCCA-GATGATGTCCA-39, and primer 2 (sense) was defined bybases 579 to 596, sequence 59-GTGCCTTGCCCCCA-GACT-39. The reverse transcription was performed at 50°Cfor 30 minutes using 100 ng total RNA for detection ofET-1 mRNA and 10 ng total RNA for detection of ET-3mRNA. Up to 40 amplification cycles were performedstarting with denaturation at 94°C for 30 seconds (first cycle3 min), followed by primer annealing at 54°C for 45seconds, and extension at 68°C for one minute. The ex-pected size of the RT-PCR product was 440 base pairs forET-1 and 276 base pairs for ET-3. RT-PCR products wereanalyzed on a 1% agarose gel and visualized by staining

with ethidium bromide. To confirm that RT-PCR productsrepresent rabbit prepro-ET-1 and ET-3 encoding frag-ments, PCR products were isolated from agarose gel with aQIAquick Gel Extraction Kit (Qiagen, Hilden, Germany)and sequenced (Sequiserve, Vaterstetten, Germany; ABIPRISM Dye Terminator, Perkin Elmer, Foster City, CA,USA).

Immunofluorescence staining of cultured rabbit proximaltubule cells

Cells were grown on glass slides. For detection ofendothelin-like immunoreactivity, cells were fixed withmethanol (6 min, 220°C). Unspecific staining was blockedby incubation with porcine normal serum (1:10) for 30minutes at room temperature. Cells were washed withphosphate buffered saline (PBS) and incubated with anti-bodies against ET-1 [1:100, monoclonal (mouse) anti-endothelin 1 antibody; Dianova, Hamburg, Germany; crossreactivity for ET-1 was 100%, ET-2 ,10%, and ET-3,10%] and ET-3 (1:200, rabbit anti-endothelin 3; PhoenixPharmaceuticals, Mountain View, CA, USA; cross reactiv-ity for ET-3 was 100%, ET-1 0.1%, and ET-2 0.1%) for twohours at room temperature. Cells were washed with PBSand incubated with biotinylated secondary antibodies [bio-tinylated F(ab9)2 fragment of rabbit anti-mouse immuno-globulins 1:400; biotinylated F(ab9)2 fragment of swineanti-rabbit immunoglobulins 1:500; Dako, Hamburg, Ger-many] for one hour at room temperature. After anotherwashing step, binding of the secondary antibody was de-tected with streptavidin-Cy3 (1:300 for 1 hr at roomtemperature; Rockland, Gilbertsville, PA, USA). Controlsincluded Western blots to demonstrate that anti-ET-1 andanti-ET-3 antibodies react with specific bands of the pre-dicted sizes for ET-1 and ET-3 as well as immunofluores-cence staining of PT-1 cells with (a) anti-ET-1 and anti-ET-3 antibodies, preabsorbed with the respective peptides(1026 and 1025 mol/liter, 24 hr at 4°C) and (b) equivalentconcentrations of pre-immune serum instead of the pri-mary antibody.

Investigation of the time course of endothelin-1 andendothelin-3 release

The time course of endothelin release was determined inPT-1 cells, cultured in six well plates (N 5 3) in culturemedium containing 5% FCS. Cell culture supernatantswere collected after 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 24, 36, 48 and72 hours into EDTA containing tubes (1.2 mg/ml), imme-diately frozen and stored at 280°C, until assayed. Thecorresponding cell numbers were counted in a cell countingchamber and controlled by an automatic cell counter(Cell-Dyn 3500; auxiliary mode; Abbott, Abbott Park, IL,USA).

Haug et al: Endothelin release by rabbit proximal tubule cells 1627

Investigation of polar secretion of endothelin-1 andendothelin-3

For the investigation of polar endothelin secretion, cellswere grown in six well plates on filter membranes (pore size0.4 mm; Becton Dickinson, Basel, Switzerland). Confluentmonolayers were incubated with cell culture medium con-taining either 1% or 10% FCS (2 experiments, 3 wellseach). After an incubation period of 72 hours, cell culturemedium was collected from the basolateral and apicalcompartments.

Incubation of proximal tubule cells with cyclosporine A,tacrolimus and renotropic factors

PT-1 cells, cultured in six well plates in culture mediumsupplemented with 5% FCS, were incubated for five dayswith cyclosporine A (Sandimmun; Sandoz, Nurnberg, Ger-many) and tacrolimus (Prograf; Fujisawa, Munchen, Ger-many) at final concentrations of 0.5, 5, 50, 500 and 5000mg/liter. Cell culture medium was exchanged after two days,and the supernatants from the last three days were col-lected into EDTA containing tubes, and immediataleyfrozen. For controls, cells were cultured in culture mediumalone and in cell culture medium containing ethanol con-centrations equivalent to 5000 mg/liter CsA (ethanol0.0035%) or 5000 mg/liter tacrolimus (ethanol 0.08%).Three different experiments were done in three wells each.

To investige the effect of renotropic factors, PT-1 cellswere incubated for five days (in culture medium containing5% FCS) with recombinant human hepatocyte growthfactor or epidermal growth factor (10210, 1029, 1028 mol/liter; R&D Systems, Wiesbaden, Germany). In addition,PT-1 cells were coincubated with CsA or tacrolimus (50and 500 mg/liter) and HGF or EGF (1028 mol/liter). Twoexperiments were done in three wells each. Cell culturemedium was exchanged after two days, and the cell culturemedium from the last three days was collected for deter-mination of ET-1 and ET-3.

Measurement of immunoreactive endothelin-1 andendothelin-3

ET-1 and ET-3 concentrations in cell culture superna-tants were determined after solid phase extraction. Sep PakC18 cartridges (Waters, Eschborn, Germany) were pre-washed successively with 100% methanol, distilled water,and 4% acetic acid. Samples were acidified with 4% aceticacid and applied to the cartridges. After applying thesamples, the columns were washed with 25% ethanol, andthe absorbed peptides were eluted by a solution containing86% ethanol and 4% acetic acid. The eluate was dried in acentrifugal vacuum concentrator and suspended in radio-immunoassay buffer. The recovery was determined byaddition of the cold peptide: ET-1, 83.7 6 11.2%; ET-3,78.4 6 9.7% (N 5 10).

ET-1 and ET-3 were measured using radioimmunoassaykits purchased from Phoenix Pharmaceuticals (cross reac-tivities were, for the ET-1 assay, ET-1 100%, ET-2 7%,ET-3 7%; for the ET-3 assay, ET-3 100%, ET-1 0.1%, ET-20.1%). Standards or samples (100 ml) were incubated with100 ml of rabbit anti-ET-1 or anti-ET-3 antibody overnightat 4°C. The following day, 100 ml of 125I-labeled ET-1 orET-3 were added. After overnight incubation at 4°C, goatanti-rabbit immunoglobulin and normal rabbit serum wereadded, followed by an incubation period of 90 minutes atroom temperature. After addition of 500 ml radioimmuno-assay buffer, samples were centrifuged (3000 rpm, 20 min),supernatants were decanted, and the pellets were countedin a gamma-counter. All standards and samples weremeasured in duplicate. Intra-assay coefficients of variation(N 5 16) were 7.4% (ET-1) and 6.3% (ET-3), and inter-assay coefficients of variation (N 5 6) were 11.1% and9.7%, respectively. No immunoreactive endothelin wasdetected in cell culture medium that had not been incu-bated with the cells.

Investigation of cell viabilityThe activity of lactate dehydrogenase was determined in

cell culture supernatants of PT-1 cells, incubated for fivedays with CsA or tacrolimus (0.5 to 5000 mg/liter), by aroutine assay, measuring the decrease of absorbance at 340nm, resulting from the conversion of NADH to NAD in thepresence of pyruvate.

PT-1 cells, seeded on to glass slides, were incubated withcell culture medium containing 5% FCS and 500 or 5000mg/liter CsA or tacrolimus, and after three days, cells wereincubated with fluorescein diacetate (10 mg/ml, 1 min atroom temperature; Sigma, Deisenhofen, Germany) or pro-pidium iodide (1 mg/ml, 5 min at room temperature;Sigma). Viable cells that could hydrolyse fluorescein diac-etate exhibited a green fluorescent staining of cytoplasm.Toxically damaged cells exhibited a red fluorescent stainingof nuclei after incubation with propidium iodide.

Statistical analysisResults are expressed as means 6 SEM. ET-1 and ET-3

release are given as pg/106 cells, referred to the cell numberat the time point of supernatant collection. Differencesbetween the endothelin release into the basolateral andapical compartment in PT-1 cells, cultured with either 1%or 10% FCS as well as the influence of different concen-trations of CsA and tacrolimus and the influence of growthfactors on ET-1 and ET-3 release were evaluated by oneway analysis of variance, followed by the Newman-Keulstest.

RESULTSEndothelin-1 and endothelin-3 mRNA expression inproximal tubule cells

Investigation of ET-1 and ET-3 mRNA expression inPT-1 cells, maintained in cell culture medium containing

Haug et al: Endothelin release by rabbit proximal tubule cells1628

5% FCS, exhibited bands of 440 base pairs and 276 basepairs that corresponded to the predicted amplified se-quences for ET-1 and ET-3 (Fig. 1). No detectable PCRproducts were present in water controls and in controls,amplified without prior reverse transcription.

Detection of immunoreactive endothelin-1 andendothelin-3 in proximal tubule cells

Immunofluorescence staining of cultured rabbit proximaltubule cells grown on glass slides with antibodies againstET-1 and ET-3 revealed an intense perinuclear staining(Fig. 2). No significant immunoreactive staining was ob-served in controls, using anti-ET-1 and anti-ET-3 antibod-ies, prereacted with the corresponding peptides, or pre-immune serum instead of the primary antibody.

Time course of endothelin-1 and endothelin-3 release byproximal tubule cells

The time course of ET-1 and ET-3 release was investi-gated in PT-1 cells, cultured in medium containing 5%FCS. Supernatants were collected after different incuba-tion periods (N 5 3). ET-1 release increased up to anincubation period of 36 hours, whereas ET-3 release hadalready reached a steady state level after an incubationperiod of four hours, which, however, was lower than themaximum ET-1 release (Fig. 3).

Investigation of polar endothelin-1 and endothelin-3secretion by proximal tubule-1 cells

PT-1 cells, grown on filter membranes in cell culturemedium supplemented with either 1% or 10% FCS, re-leased a significantly higher amount of immunoreactiveET-1 into the basolateral compartment than into the apicalcompartment, whereas no significant difference betweenbasolateral and apical ET-3 release was observed. ET-1release into the basolateral and apical compartment wassignificantly increased in PT-1 cells grown in culture me-dium supplemented with 10% FCS compared to PT-1 cellsgrown in culture medium containing 1% FCS. In contrast,the stimulatory effect of FCS on ET-3 release was lesspronounced. Data are shown in Figure 4.

Effect of cyclosporine A and tacrolimus on endothelin-1and endothelin-3 release

Incubation of PT-1 cells with CsA and tacrolimus overfive days induced a dose-dependent stimulation of ET-1and ET-3 release into the cell culture medium with amaximum stimulation at CsA concentrations of 500 mg/literand tacrolimus concentrations of 50 mg/liter. Results areshown in Figures 5 and 6 and represent the relative ET-1and ET-3 release compared to ethanol containing controls.ET-1 and ET-3 release did not differ significantly betweennon-alcohol and alcohol controls (CsA, 0.0035%; tacroli-mus, 0.08%; ET-1 without ethanol 116.1 6 12.5; ET-1 with0.0035% ethanol, 115.7 6 14.7; ET-1 with 0.08% ethanol,114.4 6 15.8; ET-3 without ethanol, 41.9 6 3.5, ET-3 with0.0035% ethanol, 42.5 6 2.2, ET-3 with 0.08% ethanol,40.2 6 3.4 pg/106 cells).

To determine whether the decrease of ET-1 and ET-3release at a CsA concentration of 5000 mg/liter and attacrolimus concentrations of 500 and 5000 mg/liter mighthave been due to a cytotoxic effect, LDH activity wasdetermined in supernatants, and cells were stained withfluorescein diacetate and propidium iodide. LDH activitydid not differ significantly between controls and superna-tants of PT-1 cells, incubated with CsA 0.5 to 500 mg/liter ortacrolimus 0.5 to 5 mg/liter, but was significantly elevated inculture medium of PT-1 cells incubated with 5000 mg/literCsA (59.7 6 2.3 U/liter, P , 0.001) and 500 or 5000 mg/litertacrolimus (33.0 6 1.4 U/liter, 40.0 6 1.2 U/liter, P , 0.01)compared to controls (23.3 6 1.2 U/liter). A positivecytoplasmatic staining with fluorescein diacetate was ob-served in about 100% of control cells (5% FCS), and about90% and 95% of the cells incubated with CsA or tacroli-mus. Only very few of the control cells, and about 5 to 10%of PT-1 cells incubated with 500 mg/liter CsA exhibited apositive labeling with propidium iodide, whereas about 20to 40% of PT-1 cells incubated with 5000 mg/liter CSA or500 and 5000 mg/liter tacrolimus showed a nuclear immu-nofluorescence staining.

Fig. 1. Detection of endothelin (ET)-1 and ET-3 mRNA by RT-PCR inrabbit proximal tubule cells, grown in culture medium containing 5% fetalcalf serum. Ethidium bromide stained PCR products in 1% agarose gel.Arrows indicate the PCR products of the predicted size of 440 bp for ET-1mRNA (lane 2) and 276 bp for ET-3 mRNA (lane 3). The left lane is a 100bp DNA ladder.


Inhibitory action of renotropic factors on endothelin-1release

Incubation of PT-1 cells with HGF and EGF (10210 to1028 mol/liter) for five days induced a significant (P ,0.001) dose-dependent reduction of ET-1 release com-pared to controls, whereas ET-3 release was not signifi-cantly reduced (controls, ET-1 134.3 6 8.5, ET-3 47.7 65.3, pg/106 cells, N 5 6). Data are shown in Figures 7 and8.

Coincubation of PT-1 cells with CsA or tacrolimus (50and 500 mg/liter) and HGF (1028 mol/liter) for five days

induced a strong reduction of ET-1 release to values similarto those obtained with HGF alone. No significant inhibitoryeffect on ET-3 release was observed. Coincubation withCsA or tacrolimus and EGF (1028 mol/liter) also induced asignificant reduction of ET-1 release, and only a slightreduction of ET-3 release. Results are shown in Figures 9and 10.

DISCUSSION

In the present study, we have demonstrated that culturedrabbit proximal tubule cells express ET-1 and ET-3 mRNA,exhibit a positive immunofluorescence staining with anti-bodies against ET-1 and ET-3, and secrete ET-1 and ET-3into the cell culture medium. A time-dependent increase ofET-1 release up to an incubation period of 36 hours wasobserved, whereas the ET-3 release had already reached asteady state level after four hours. PT-1 cells releasedimmunoreactive ET-1 predominantly into the basolateralcompartment. The basolateral and the apical ET-1 releasewas significantly increased when cells were incubated in cellculture medium containing 10% FCS, compared to 1%FCS. In contrast, ET-3 release did not differ significantlybetween the basolateral and the apical compartments, andthe stimulation by FCS was less pronounced. The currentfindings of a polar ET-1 secretion by PT-1 cells are inaccordance with previous studies performed with rabbitproximal tubule cells [9], rat inner medullary collecting ductcells [29], mouse cortical and medullary collecting duct celllines [8], and a cell line originating from dog collecting duct[30]. A dose-dependent stimulatory effect of FCS on ET-1release has been observed in experiments with rat inner

Fig. 2. Immunofluorescence staining of rabbitproximal tubule cells with antibodies againstendothelin (ET)-1 (A) and ET-3 (B). Originalmagnification 31000.

Fig. 3. Endothelin-1 (f) and endothelin-3 (■) release by cultured rabbitproximal tubule cells, incubated for different time periods in cell culturemedium containing 5% FCS. Data are means 6 SEM, N 5 3.


medullary tubule suspensions [31]. In addition to previousstudies on ET-1 secretion, the present data provide supportfor a different regulation of ET-1 and ET-3 release, bydemonstrating a non-polar secretion of ET-3, a differenttime course of ET-1 and ET-3 release, and a differentstimulatory effect of FCS on ET-1 and ET-3 release byrabbit proximal tubule cells.

Several studies have suggested that ET-1 plays an impor-tant role in mediating the nephrotoxic side effects of CsAand tacrolimus. Application of CsA has been demonstratedto induce a vasoconstriction in isolated rat renal afferentarterioles, which was antagonized by the application of aselective ETA receptor antagonist [32]. In addition, severalin vivo studies have shown a protective effect of ETA

Fig. 4. Basolateral (bas.) and apical (ap.)endothelin-1 (A, dark grey columns) andendothelin-3 (B, light grey columns) release byproximal tubule (PT-1) cells, grown on filtermembranes in culture medium containing 1%or 10% FCS. Data are means 6 SEM, N 5 6.***P , 0.001 versus apical endothelin-1release, 111P , 0.001 compared to endothelin-1 release by PT-1 cells, cultured in mediumcontaining 1% FCS (basolateral vs. basolateral,apical vs. apical), 1P , 0.05 11P , 0.01compared to endothelin-3 release by PT-1 cells,cultured in medium containing 1% FCS(basolateral vs. basolateral, apical vs. apical).

Fig. 5. Effect of cyclosporine A on endothelin-1(A) and endothelin-3 (B) release by culturedrabbit proximal tubule cells. Data are means 6SEM, N 5 9, *P , 0.05 **P , 0.01 ***P ,0.001 versus control.


receptor antagonists on the cyclosporine-induced renaldysfunction [13, 33]. In cell culture experiments, cyclospor-ine has been shown to exhibit a stimulatory effect on ET-1production by human and rabbit endothelial cells [16, 34,35], rat vascular smooth muscle cells [36], rat mesangialcells [17, 18], and by the renal cell line LLC-PK1 [37].Similarly, a stimulatory effect of tacrolimus on ET-1 release

by cultured human endothelial cells [34], human and ratmesangial cells [38, 39] and by the LLC-PK1 tubular cellline [38] has been described. In the present study, adose-dependent stimulatory effect of CsA and tacrolimuson ET-1 and ET-3 release by rabbit proximal tubule cellswith a maximum stimulation at a CsA concentration of 500mg/liter and a tacrolimus concentration of 50 mg/liter was

Fig. 6. Effect of tacrolimus on endothelin-1 (A)and endothelin-3 (B) release by cultured rabbitproximal tubule cells. Data are means 6 SEM,N 5 9, *P , 0.05 ***P , 0.001 versus control.

Fig. 7. Effect of hepatocyte growth factor(HGF) on endothelin-1 (A) and endothelin-3(B) release by cultured rabbit proximal tubulecells. Data are means 6 SEM, N 5 6, ***P ,0.001 versus control.


observed. Consistent with the decrease of ET-1 and ET-3release at a CsA concentration of 5000 mg/liter and attacrolimus concentrations of 500 and 5000 mg/liter, asignificant increase of lactate dehydrogenase (LDH) re-lease and a deterioration of cell viability has been shown.

The maximum CsA concentration used in this study wasabout 25- to 50-fold higher than the therapeutic range ofcirculating concentrations in patients with chronic cyclo-sporine administration; however, it has been suggested thattissue levels of the lipophilic substances CsA or tacrolimus

Fig. 8. Effect of epidermal growth factor(EGF) on endothelin-1 (A) and endothelin-3(B) release by cultured rabbit proximal tubulecells. Data are means 6 SEM, N 5 6, **P ,0.01 ***P , 0.001 versus control.

Fig. 9. Endothelin-1 (A) and endothelin-3 (B)release by proximal tubule (PT-1) cells,coincubated with cyclosporine A and hepatocytegrowth factor (HGF, 1028 mol/liter) orepidermal growth factor (EGF, 1028 mol/liter).Data are means 6 SEM, N 5 6, ***P , 0.001versus cyclosporine 50 mg/liter, 11P , 0.01 and111P , 0.001 versus cyclosporine 500 mg/liter.


might be up to about 50-fold higher than blood concentra-tions [40, 41]. To compare the effects of CsA and tacroli-mus, tacrolimus was also used in the same concentrations,although the therapeutic range of tacrolimus is abouttenfold lower than the therapeutic range of CsA. In arecently published study, a five day incubation of culturedhuman renal cortical epithelial cells with 2,000 mg/liter CsAinduced an increase of ET-1 release, whereas at a CsAconcentration of 10,000 mg/liter no more stimulatory effecton ET-1 release was observed [19]. In another studyinvestigating the effect of tacrolimus on ET-1 release by theLLC-PK1 tubular cell line, after incubation periods of 20and 30 hours the stimulatory effect of 1 mM tacrolimus (800mg/liter) was less pronounced than the stimulatory effect of0.1 mM tacrolimus (80 mg/liter) [38].

At present, the physiological or pathophysiological roleof tubular endothelin synthesis is not yet fully understood.An increased proximal tubular endothelin productionmight by paracrine mechanisms be involved in mediatingthe CsA- and tacrolimus-induced renal hypoperfusion. Thiseffect might be enhanced by an autocrine stimulation ofendothelin production. An autoinduction of ET-1 has beenobeserved in human proximal tubule cells [42] and ratmesangial cells [3]. It has also been speculated that anenhanced endothelin production by autocrine mechanismsmight be involved in mediating cyclosporine-associatedtubular structural damage. However, until now this remainsunclear, and previous animal studies have yielded contra-dictory results [43, 44].

Renotropic factors, such as hepatocyte growth factor andepidermal growth factor, have been shown to act as mito-gens for proximal tubule cells and accelerate tubular regen-eration after drug-induced and ischemic injury [20–24, 26].In the current study, HGF and EGF exhibited a significantdose-dependent inhibitory effect on ET-1 release by PT-1cells, whereas ET-3 release was not significantly reduced.The CsA- and tacrolimus-induced ET-1 release was alsosignificantly reduced by HGF and EGF. Interestingly, HGFhas recently been shown to exert a preventive effect on thedegenerative changes in renal tubular cells of cyclosporine-treated mice [45] and the tacrolimus-induced nephrotoxic-ity in rats [46]. Thus, it is speculated that the protectiveeffect of renotropic factors might be mediated in part bytheir inhibitory effect on endothelin release.

In summary, the present data suggest that ET-1 andET-3 release by rabbit proximal tubule cells is regulated bydifferent mechanisms. CsA and tacrolimus have beenshown to exert a dose-dependent effect on ET-1 and ET-3release by rabbit proximal tubule cells. The present findingsof an inhibitory effect of HGF and EGF on the basal andCsA- and tacrolimus-induced ET-1 release suggest that therenoprotective effects of HGF and EGF might be linked totheir modulatory effect on endothelin production.

ACKNOWLEDGMENTSWe thank Dr. R.W. Grunewald, Department of Nephrology, University

Hospital Gottingen, Germany, for the kind gift of proximal tubule-1(PT-1) cells used in this study. This work was supported by a grant fromthe University of Ulm and by the Landesforschungsschwerpunkt Baden-Wurtemberg “Modulation von Wachstumsfaktoren als Therapieprinzip”to P.M.J. (Baustein P. 221 and P. 397).

Fig. 10. Endothelin-1 (A) and endothelin-3 (B)release by proximal tubule (PT-1) cells,coincubated with tacrolimus and hepatocytegrowth factor (HGF, 1028 mol/liter) orepidermal growth factor (EGF, 1028 mol/liter).Data are means 6 SEM, N 5 6, **P , 0.01***P , 0.001 versus tacrolimus 50 mg/liter,111P , 0.001 versus tacrolimus 500 mg/liter.


Reprint requests to Dr. Cornelia Haug, Institute of Clinical Chemistry,University Hospital Ulm, Robert-Koch-Strasse 8, D-89070 Ulm, Germany.E-mail: [email protected]

APPENDIX

Abbreviations used in this article are: CsA, cyclosporine A; EGF,epidermal growth factor; ET, endothelin; FCS, fetal calf serum; HGF,hepatocyte growth factor; LDH, lactate dehydrogenase; RT-PCR, reversetranscription-polymerase chain reaction.

REFERENCES

1. YANAGISAWA M, KURIHARA H, KIMURA S, TOMOBE Y, KOBAYASHI M,MITSUI Y, YAZAKI Y, GOTO K, MASAKI T: A novel potent vasocon-strictor peptide produced by vascular endothelial cells. Nature 332:411–415, 1988

2. INOUE A, YANAGISAWA M, KIMURA S, KASUYA Y, MIYAUCHI T, GOTOK, MASAKI T: The human endothelin family: Three structurally andpharmacologically distinct isopeptides predicted by three separategenes. Proc Natl Acad Sci USA 86:2863–2867, 1989

3. IWASAKI S, HOMMA T, MATSUDA Y, KON V: Endothelin receptorsubtype B mediates autoinduction of endothelin-1 in rat mesangialcells. J Biol Chem 270:6997–7003, 1995

4. KOHAN DE: Endothelin synthesis by rabbit renal tubule cells. Am JPhysiol 261:F221–F226, 1991

5. CHEN M, TODD-TURLA K, WANG WH, CAO X, SMART A, BROSIUS FC,KILLEN PD, KEISER JA, BRIGGS JP, SCHNERMANN J: Endothelin-1mRNA in glomerular and epithelial cells of kidney. Am J Physiol265:F542–F550, 1993

6. TERADA Y, TOMITA K, NONOGUCHI H, YANG T, MARUMO F: Expres-sion of endothelin-3 mRNA along rat nephron segments usingpolymerase chain reaction. Kidney Int 44:1273–1280, 1993

7. WAGNER OF, CHRIST G, WOJTA J, VIERHAPPER H, PARZER S,NOWOTNY PJ, SCHNEIDER B, WALDHAUSL W, BINDER BR: Polarsecretion of endothelin-1 by cultured endothelial cells. J Biol Chem267:16066–16068, 1992

8. ZOJA C, MORIGI M, FIGLIUZZI M, BRUZZI I, OLDROYD S, BENIGNI A,RONCO P, REMUZZI G: Proximal tubular cell synthesis and secretion ofendothelin-1 on challenge with albumin and other proteins. Am JKidney Dis 26:934–941, 1995

9. TODD-TURLA KM, ZHU XL, SHU X, CHEN M, YU T, SMART A, KILLENPD, FEJES-TOTH G, BRIGGS JP, SCHNERMANN JB: Synthesis andsecretion of endothelin in a cortical collecting duct cell line. Am JPhysiol 271:F330–F339, 1996

10. MINO N, KOBAYASHI M, NAKAJIMA A, AMANO H, SHIMAMOTO K,ISHIKAWA K, WATANABE K, NISHIKIBE M, YANO M, IKEMOTO F:Protective effect of a selective endothelin receptor antagonist, BQ-123, in ischemic acute renal failure in rats. Eur J Pharmacol 221:77–83,1992

11. OHTA K, HIRATA Y, SHICHIRI M, KANNO K, EMORI T, TOMITA K,MARUMO F: Urinary excretion of endothelin-1 in normal subjects andpatients with renal disease. Kidney Int 39:307–311, 1991

12. YOSHIMURA A, IWASAKI S, INUI K, IDEURA T, KOSHIKAWA S, YANAGI-SAWA M, MASAKI T: Endothelin-1 and endothelin B type receptor areinduced in mesangial proliferative nephritis in the rat. Kidney Int48:1290–1297, 1995

13. KON V, SUGIURA M, INAGAMI T, HARVIE BR, ICHIKAWA I, HOOVERRL: Role of endothelin in cyclosporine-induced glomerular dysfunc-tion. Kidney Int 37:1487–1491, 1990

14. GRIEFF M, LOERTSCHER R, AL SHOHAIB S, STEWART DJ: Cyclospor-ine-induced elevation in circulating endothelin-1 in patients withsolid-organ transplants. Transplantation 56:880–884, 1993

15. IWASAKI S, HOMMA T, KON V: Site specific regulation in the kidney ofendothelin and its receptor subtypes by cyclosporine. Kidney Int45:592–597, 1994

16. BUNCHMAN TE, BROOKSHIRE CA: Cyclosporine-induced synthesis ofendothelin by cultured human endothelial cells. J Clin Invest 88:310–314, 1991

17. TAKEDA M, IWASAKI S, HELLINGS SE, YOSHIDA H, HOMMA T, KON V:Divergent expression of ETA and ETB receptors in response tocyclosporine in mesangial cells. Am J Pathol 144:473–479, 1994

18. KOHNO M, YOKOKAWA K, MANDAL AK, HORIO T, YASUNARI K,TAKEDA T: Cardiac natriuretic peptides inhibit cyclosporine-inducedproduction of endothelin in cultured rat mesangial cells. Metabolism44:404–409, 1995

19. ONG ACM, JOWETT TP, SCOBLE JE, O’SHEA JA, VARGHESE Z,MOORHEAD JF: Effect of cyclosporin A on endothelin synthesis bycultured human renal cortical epithelial cells. Nephrol Dial Transplant8:748–753, 1993

20. IGAWA T, KANDA S, KANETAKE H, SAITOH Y, ICHIHARA A, TOMITA Y,NAKAMURA T: Hepatocyte growth factor is a potent mitogen forcultured rabbit renal tubular epithelial cells. Biochem Biophys ResCommun 174:831–838, 1991

21. KAWAIDA K, MATSUMOTO K, SHIMAZU H, NAKAMURA T: Hepatocytegrowth factor prevents acute renal failure and accelerates renalregeneration in mice. Proc Natl Acad Sci USA 91:4357–4361, 1994

22. MILLER SB, MARTIN DR, KISSANE J, HAMMERMAN MR: Hepatocytegrowth factor accelerates recovery from acute ischemic renal injury inrats. Am J Physiol 266:F129–F134, 1994

23. NORMAN J, BADIE-DEZFOOLY B, NORD EP, KURTZ I, SCHLOSSER J,CHAUDHARI A, FINE LG: EGF-induced mitogenesis in proximaltubular cells: Potentiation by angiotensin II. Am J Physiol 253:F299–F309, 1987

24. JENNISCHE E, ANDERSSON G, HANSSON HA: Epidermal growth factoris expressed by cells in the distal tubules during postnephrectomyrenal growth. Acta Physiol Scand 129:449–450, 1987

25. JEHLE PM, FUßGANGER RD, STRACKE S, GRUNEWALD RW, KELLER F:Different binding and degradation of proinsulin, insulin and insulin-like growth factor-1 (IGF-1) in cultured renal proximal tubular cells.Acta Diabetol 33:159–165, 1996

26. STRACKE S, ERNST F, JEHLE DR, GRUNEWALD RW, HALLER H,KELLER F, JEHLE PM: Differentiating and proliferative effects of HGFin renal proximal tubular cells are mediated via different signallingpathways. Nephrol Dial Transplant 13:1398–1405, 1998

27. MARSDEN PA, SULTAN P, CYBULSKY M, GIMBRONE MA, BRENNERBM, COLLINS T: Nucleotide sequence of endothelin-1 cDNA fromrabbit endothelial cells. Biochim Biophys Acta 1129:249–250, 1992

28. OHKUBO S, ITOH Y, KIMURA C, ONDA H, FUJINO M: Nucleotidesequence of a rabbit genomic DNA encoding mature endothelin-3.Nucleic Acids Res 18:374, 1990

29. KOHAN DE, PADILLA E: Endothelin-1 is an autocrine factor in ratinner medullary collecting ducts. Am J Physiol 263:F607–F612, 1992

30. UCHIDA S, HORIE M, YANAGISAWA M, MATSUSHITA Y, KUROKAWA K,OGATA E: Polarized secretion of endothelin-1 and big ET-1 in MDCKcells is inhibited by cell Na1 flux and disrupted by NH4Cl. J CardiovascPharmacol 17(Suppl 7):S226–S228, 1991

31. UJIIE K, TERADA Y, NONOGUCHI H, SHINOHARA M, TOMITA K,MARUMO F: Messenger RNA expression and synthesis of endothelin-1along rat nephron segments. J Clin Invest 90:1043–1048, 1992

32. LANESE DM, CONGER JD: Effects of endothelin receptor antagoniston cyclosporine-induced vasoconstriction in isolated rat renal arte-rioles. J Clin Invest 91:2144–2149, 1993

33. KIVLIGHN SD, GABEL RA, SIEGL PKS: Effects of BQ-123 on renalfunction and acute cyclosporine-induced renal dysfunction. Kidney Int45:131–136, 1994

34. TAKEDA Y, YONEDA T, ITO Y, MIYAMORI I, TAKEDA R: Stimulation ofendothelin mRNA and secretion in human endothelial cells by FK506. J Cardiovasc Pharmacol 22(Suppl 8):S310–S312, 1993

35. LANGMAN LJ, YATSCOFF RW: Comparison of the effects of cyclo-sporin G (OG37–325), cyclosporin A, and their metabolites on therelease of endothelin and prostacyclin from primary renal and aorticendothelial cell lines. Ther Drug Monit 16:450–457, 1994

36. TAKEDA Y, ITOH Y, YONEDA T, MIYAMORI I, TAKEDA R: CyclosporineA induces endothelin-1 release from cultured rat vascular smoothmuscle cells. Eur J Pharmacol 233:299–301, 1993

37. NAKAHAMA H, FUKUNAGA M, KAKIHARA M, HORIO M, FUJIWARA Y,FUKUHARA Y, UEDA N, ORITA Y, KAMADA T: Comparative effects ofcyclosporine A and FK-506 on endothelin secretion by a culturedrenal cell line, LLC-PK1. J Cardiovasc Pharmacol 17(Suppl 7):S172–S173, 1991

38. MOUTABARRIK A, ISHIBASHI M, FUKUNAGA M, KAMEOKA H, TAKANOY, KOKADO Y, TAKAHARA S, JIANG H, SONODA T, OKUYAMA A: FK506 mechanism of nephrotoxicity: Stimulatory effect on endothelin


secretion by cultured kidney cells and tubular cell toxicity in vitro.Transplant Proc 23:3133–3136, 1991

39. GOODALL T, KIND CN, HAMMOND TG: FK506-induced endothelinrelease by cultured rat mesangial cells. J Cardiovasc Pharmacol26(Suppl 3):S482–S485, 1995

40. CHRISTIANS U, SEWING KF: Cyclosporin metabolism in transplantpatients. Pharmac Ther 57:291–345, 1993

41. VENKATARAMANAN R, JAIN A, CADOFF E, WARTY V, IWASAKI K,NAGASE K, KRAJACK A, IMVENTARZA O, TODO S, FUNG JJ, STARZLTE: Pharmacokinetics of FK 506: Preclinical and clinical studies.Transplant Proc 22(Suppl 1):52–56, 1990

42. ONG ACM, JOWETT TP, FIRTH JD, BURTON S, KARET FE, FINE LG:An endothelin-1 mediated autocrine growth loop involved in humanrenal tubular regeneration. Kidney Int 48:390–401, 1995

43. HUNLEY TE, FOGO S, IWASAKI S, KON V: Endothelin A receptormediates functional but not structural damage in chronic cyclosporinenephrotoxicity. J Am Soc Nephrol 5:1718–1723, 1995

44. KON V, HUNLEY TE, FOGO A: Combined antagonism of endothelinA/B receptors links endothelin to vasoconstriction whereas angioten-sin II effects fibrosis. Transplantation 60:89–95, 1995

45. AMAIKE H, MATSUMOTO K, OKA T, NAKAMURA T: Preventive effect ofhepatocyte growth factor on acute side effects of cyclosporin A inmice. Cytokine 8:387–394, 1996

46. TAKADA S, NAMIKI M, TAKAHARA S, MATSUMIYA K, KONDOH N,KITAMURA M, UCHIDA K, KOGA M, JIANG H, KOKADO Y, KAMEOKAH, MIKI T, MATSUMOTO K, NAKAMURA T, OKUYAMA A: HGFprevents tacrolimus (FK 506)-induced nephrotoxicity in SHR rats.Transplant Proc 28:1089–1090, 1996


endothelin release by rabbit proximal tubule cells: modulatory effects of cyclosporine a,...

Documents