ETA and ETB receptor antagonists synergistically increase extracellular endothelin-1 levels in primary rat astrocyte cultures

Brain Research 785 Ž1998. 253–261

Research report

ETA and ETB receptor antagonists synergistically increase extracellular endothelin-1 levels in primary rat astrocyte cultures Martin Hasselblatt a , Heike Kamrowski-Kruck a , Niels Jensen a , Lothar Schilling b , Hartmut Kratzin a , Anna-Leena Siren ´ a , Hannelore Ehrenreich a, ) a

Departments of Neurology and Psychiatry, Georg-August-UniÕersity, and Max-Planck-Institute for Experimental Medicine, Gottingen, Germany ¨ b Department of Neurosurgery, UniÕersity Hospital, Mannheim, Germany Accepted 4 November 1997

Abstract Astrocytes produce and bind endothelins ŽETs., suggesting that these cells have ET autoregulatory and eliminatory functions. To further investigate these functions in primary rat astrocytes, ET-1 levels in the cell culture media ŽRIArHPLC. and intracellular content of ET-1 mRNA ŽRT PCR. were measured under basal and stimulated Žthrombin, 2.2 Urml. conditions in the presence and absence of ETA and ETB selective antagonists ŽBQ123 or LU135252, and BQ788, respectively.. Neither basal nor stimulated ET-1 levels in astrocyte media were influenced by ETA or ETB antagonists alone, but were significantly increased by a combination of both. ir ET-3 levels were not affected by antagonist treatment. Exogenous ET-1, added to the cultures, was rapidly cleared from the supernatant; this clearance was markedly inhibited by a combination of BQ123 and BQ788. ET-1 mRNA levels were not altered by any treatment. To conclude, in primary rat astrocyte cultures, extracellular ET-1 is cleared by binding to ET-receptors, apparently involving both, ETA and ETB sites. Thus, a blockade of the astrocytic ET eliminatory function as a consequence of the in vivo application of non-selective ET receptor antagonists may lead to increased extracellular ET levels in the brain. q 1998 Elsevier Science B.V. Keywords: Endothelin-1; ETA ; ETB ; Endothelin receptor antagonist; Thrombin; Hirudin; Astrocyte; Rat

1. Introduction Endothelin ŽET.-1, a potent 21-amino acid vasoconstrictor peptide, was first isolated from the supernatant of porcine aortic endothelial cells w36x. ET-2 and ET-3 were identified as isopeptides with somewhat different pharmacological properties w15x. In mammals, ETs act via at least two distinct ET receptor subtypes, ETA and ETB . While ETA displays ET-1 selective binding, the ETB receptor accepts all ET isopeptides with similar affinity w1,30x. Soon after their discovery, ETs have been shown to be produced by cells other than endothelial cells. In the brain, astrocytes were found to produce and also specifically bind these peptides w6,8,22x. The nature of the astrocytic ET binding is unclear: PCR analysis revealed the presence of mRNA for both ETA and ETB receptors in astrocytes w7x, whereas binding studies favour a single population of ET ) Corresponding author. Max-Planck-Institute for Experimental Medicine, Hermann-Rein-Str. 3, D-37075 Gottingen, Germany. Fax: ¨ q49-551-3899670; E-mail: [email protected]

0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 1 3 6 8 - 1

receptors on these cells w8,13,20x. Although simultaneous production and binding of ETs by the same cell type suggests an autoregulatory mechanism w6x, the Žpatho-. physiological significance of the astrocytic ET system is still obscure and may also vary within brain regions. However, since ETs do not cross the intact blood–brainbarrier w18x, astrocytes may be a physiological source of these peptides within the central nervous system. Here, ETs apparently influence cell proliferation and differentiation as well as contribute to the maintenance of cerebrovascular tone w19,35x. While under basal conditions, astrocytic ET secretion is low, ET-1 release can be stimulated by, e.g., thrombin w7x, a substance not only accumulated during brain injury and cerebral bleeding, but also expressed locally by glial cells and neurons w5x. Since ETs are known to influence cerebral blood vessels of all sizes and types, leading to a severe and longlasting vasospasm when applied intracisternally even at low doses w2x, an efficient ET eliminatory system would be expected to exist within the normal brain. In fact, an ET eliminatory task of astrocytic ET receptors has been sug-

254

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

gested, analogous to findings from lung tissue w12x. Particularly in situations with elevated cerebral ET levels, e.g., in subarachnoid haemorrhage-associated cerebrovasospasm w9x, ET production as well as elimination by astrocytes may play a pivotal role. This study has been designed to investigate a possible ET autoregulatory andror eliminatory capacity of ET receptors under both basal and stimulated conditions in primary rat astrocyte cultures using receptor subtypespecific antagonists. BQ123 and BQ788 are peptides with potent antagonistic properties: BQ123 is ETA selective, with an IC 50 of about 7 nM at classical ETA receptors, and an IC 50 of 18 m M at ETB receptors. BQ788 is ETB selective, with IC 50 values of about 300 nM and 1 nM at ETA vs. ETB receptors, respectively w14,16x. More recently, orally active, non-peptide ET receptor antagonists, effective in a variety of animal disease models, and with great potential for treating human disease w4,23x have been developed. Among them, LU135252 Žcompound ŽS.-6a in w28x. is an ETA selective antagonist of non-peptidic nature w28x. We report here for the first time that a combined blockade of astrocytic ET binding sites by ETA and ETB antagonists leads to a considerable increase in ET-1 concentration in astrocyte media while each of the antagonists alone is ineffective. Furthermore, exogenous ET-1 added to astrocyte cultures is rapidly cleared from the media, and this clearance is markedly delayed in the presence of a combination of ETA and ETB antagonists.

2. Methods 2.1. Cell culture Primary astrocytes were obtained from cortices of 1day-old Wistar rats ŽCharles River Laboratories. and cultured in neuron-free conditions, as described previously w32x. Cells were grown to confluence in plastic petri dishes ŽFalcon, Becton Dickinson, Heidelberg, Germany. or 24 well plates ŽGibco, Eggenstein, Germany. in glutamine-free Dulbecco’s modified Eagle’s medium ŽDMEM, Serva, Heidelberg, Germany. supplemented with 10% fetal calf serum ŽFCS, Sigma, Deisenhofen, Germany.. The medium was exchanged twice weekly. Cells were used for experiments after 2–3 weeks in culture. The purity of astrocyte cultures was routinely determined by immunofluorescence using a mouse monoclonal antibody against glial fibrillary acidic protein ŽGFAP, Boehringer, Mannheim, Germany., an astrocyte specific marker. Rhodamine-conjugated sheep anti-mouse antiserum ŽBoehringer. was used as secondary antibody. Subconfluent monolayers showed 98% positive staining for GFAP. Contamination of these cultures with microglia as assessed by using microglia specific antibodies ŽCD11b, Serotec, Raleigh, NC. was consistently less than 2%. Checking of our cultures for presence of endothelial cells and oligodendrocytes by using anti-von Wille-

brandt factor ŽBoehringer. and anti-CNPase ŽSerotec., staining gave negative results. For each experiment, only astrocytes from one preparation were used, and these cells Žapproximately 6 = 10 6 cellsrdish. were switched into phenol red-free DMEM ŽServa. in the absence of serum. For experiments, 2.2 Urml thrombin from human plasma ŽSigma., 10y6 M BQ123, 10y6 M BQ788 ŽAlexis, Grunberg, Germany., 10y6 M LU132525 Žkindly provided ¨ by Knoll, Ludwigshafen, Germany. or 10 ATUrml hirudin Žkindly provided by Behringwerke, Marburg, Germany. were added to the medium. Following the indicated incubation times, conditioned media were harvested for peptide analysis, while the cells were used for mRNA isolation. 2.2. Extraction of peptides, chromatographic procedures and radioimmunoassays (RIAs) The serumfree astrocyte-conditioned medium was centrifuged Ž2500 = g, 10 min, 48C. and concentrated by extraction through a Sep-Pak C18 cartridge ŽWaters Associates, Eschborn, Germany., pretreated with 60% acetonitrile ŽPromochem, Wesel, Germany.r0.1% trifluoroacetic acid ŽTFA, Merck, Darmstadt, Germany.. The retained material was washed with 0.1% TFA, eluted with 60% acetonitriler0.1% TFA, Speed-Vac-dried, and then exposed either directly to an ET-1 or ET-3 specific RIA ŽPeninsula, Belmont, CA. or to reverse-phase high performance liquid chromatography ŽHPLC.. HPLC was per˚ pore, 15 formed using a 3.9 mm = 30 cm column Ž300 A m m particle, C18 Delta-Pak. ŽWaters Associates.. Peptides were eluted with a linear gradient of 30% to 50% at 32 min, and to 100% at 42 min with B-solution Ž75% acetonitrile containing 0.1% TFA. at a flow rate of 1 mlrmin. Data were collected and analyzed with a LKB-Bromma HPLC work station. Fractions Ž500 m l. were collected and Speed-Vac-dried, redissolved in RIA buffer and subjected to an ET-1 RIA. Detection limit of both, ET-1 and ET-3 RIA, was 1 pgrtube. According to the manufacturer ŽPeninsula. crossreactivities in the ET-1 RIA are 17% for bigET-1 and 7% for ET-3. In the ET-3 RIA, they are less than 0.2% for ET-1 and big ET-1. Data are given as mean " standard error of the mean ŽSEM.. Statistical differences were analyzed by unpaired two-tailed Student’s t-test. A p value of - 0.05 was considered significant. 2.3. ReÕerse transcriptase–polymerase chain reaction (RT–PCR) Whole-cell RNA was isolated by guanidine thiocyanate, cesium chloride gradient centrifugation w31x. Total RNA Ž5 m g, determined spectrophotometrically. was used to generate first strand cDNA by random priming with reagents and protocols used as recommended by the manufacturers ŽPharmacia, Freiburg, Germany; Gibco.. The cDNA representing 50 ng input RNA was amplified by PCR using Taq polymerase ŽGibco. in a reaction volume of 50 m l. Spe-

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

cific primer pairs, constructed from the reported rat gene sequence for ET-1 w29x ŽCCC AGC ACA TCC TGG AGArCTC CAC CAG CTG CTG ATA; expected PCR product: 378 bp. were applied as described previously w7x. As stably expressed reference gene Ž‘housekeeping gene’., rat lamin-b was used ŽATT GAG TAT GAG TAC AAG CTGrCGC ATC TCT CTC TCT TTG TC; expected PCR product: 335 bp.. Rat lamin-b has been sequenced by our laboratory in collaboration with Dr. Alfred Bach, BASF, based on the sequence information from the human lamin-b gene w27x. The rat sequence shows 87.4% homology to the respective human lamin-b gene sequence. ŽDetails will be reported elsewhere.. Both primer pairs were added simultaneously to the PCR reaction vials. Each primer pair amplified a single band of the expected size. 32 cycles were performed at 4 min 928C, 1 min 628C, 1 min 728C, 1 min 928C. Samples were analyzed by agarose gel electrophoresis Ž1% agarose ŽGibco. containing 0.4 m grml ethidium bromide and 0.5 = TAE buffer w31x.. As positive control, rat lung tissue, known to express ET-1 w10x was used. Bands were visualized with 302 nm light and photographed using a videoprocessor ŽMitsubishi, Tokyo, Japan..

255

2.4. Binding studies DMEM without bicarbonate ŽSigma., 0.2% bovine serum albumin Žwrv. ŽPaesel and Lorei, Hanau, Germany., 25 mM HEPES ŽGibco., adjusted to pH 7.4 with sodium hydroxide solution was used as binding medium. Cells Žapproximately 250,000 cellsrwell. were washed three times with binding medium at room temperature and were incubated with 0.5 ml per well of the prepared incubation solution containing 50 pM I 125 ET-1 ŽDupont NEN, Bad Homburg, Germany. and the cold ligands ŽAlexis.. Each point was determined at least in duplicate. After incubation for 2 h at 378C, cells were washed three times with ice-cold binding medium on ice. The remaining cell layer was controlled for complete adherence under the microscope and solubilized with 0.5 ml per well of a 10 mM EDTA and 1% SDS Žwrv. solution. Cell-bound radioactivity was counted with a Wallac 1470 gamma counter. Non-specific binding in the presence of 1 m M cold ligands accounted for - 1% of the total binding of I 125 ET-1. Degradation of ligands under these assay conditions did not appear to play a role since cells, incubated in the

Fig. 1. Effect of ET antagonists on immunoreactive Žir. ETs in the media of primary rat astrocyte cultures under basal conditions. ) p - 0.05; ) ) p - 0.01 as compared to control; n s 4–7 for each experiment. 1a: time-dependent increase in ir ET-1 in astrocyte media: effect of ET antagonists Žadded at a concentration of 10y6 M each.. 1b: comparison of the effect of the ETA antagonists BQ123 vs. LU135252 in combination with the ETB antagonist BQ788 Žadded at a concentration of 10y6 M each. on ir ET-1 in astrocyte media at the 24 h time point. 1c: effect of ET antagonists on ir ET-3 at the 24 h time point.

256

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

Fig. 2. Effect of ET antagonists on thrombin-stimulated ET-1 in primary rat astrocyte cultures. ) ) p - 0.01 as compared to the thrombin-stimulated control. n s 3–6 for each experiment. 2a: time-dependent increase in ir ET-1 in astrocyte media: effect of thrombin Ž2.2 Urml. and ET antagonists Žadded at a concentration of 10y6 M each.. Note the difference in scale as compared to Fig. 1. 2b: reverse-phase HPLC profiles of ir ET-1 in extracts of astrocyte media at the 48 h time point. The arrows indicate the elution positions of synthetic ET-1, ET-3, and bigET-1. The inset denotes the respective ir ET-1 levels as determined by RIA only. The data shown are from one representative experiment. 2c: effect of the specific thrombin-antagonist hirudin Ž10 ATUrml., added 2 or 12 h after thrombin, on ir ET-1 levels in astrocyte media. 2d: analysis of rat astrocyte gene transcription by RT PCR. The temporal pattern Ž3, 8, and 24 h. of ET-1 mRNA expression Ž378 bp. under basal or stimulated Žthrombinq BQ123q BQ788. conditions is compared to that of the lamin-b PCR product Ž335 bp.. Rat lung tissue and water are used for controls.

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

257

Fig. 2. Žcontinued.

presence of a peptidase inhibitor cocktail Žconsisting of 1 mM PMSF, 1 m M pepstatin A, 100 m M leupeptin and 100 m M E-64; all from Sigma., did not yield different binding results.

3. Results 3.1. Basal concentrations of immunoreactiÕe (ir) ET-1 and ET-3 in media from primary rat astrocytes: effect of ET antagonists Under basal conditions, confluent primary rat astrocytes secreted ir ET-1 in the cell culture media in a time-dependent manner, reaching concentrations of about 20 pgr10 7 cells at 48 h. Neither addition of the ETA specific antagonist BQ123 Ž10y6 M. alone nor addition of the ETB

specific antagonist BQ788 Ž10y6 M. alone had any significant effect on ir ET-1 concentration in the media at any time point. There was, however, a tendency towards slightly elevated basal levels upon antagonist addition ŽFig. 1a.. Simultaneous addition of the two ET receptor antagonists, BQ123 and BQ788 Žboth at a concentration of 10y6 M., caused a distinct increase in ir ET-1 in the media, detectable already at the 8 h time point, and amounting to approximately 3.5 times the respective baseline level at 48 h ŽFig. 1a.. Replacement of the ETA antagonist BQ123 by the non-peptidic ETA antagonist LU135252 Ž10y6 M. yielded ir ET-1 concentrations within the same range as after BQ123 ŽFig. 1b.. None of these antagonist treatments had any effect on ir ET-3 concentration in the cell culture media ŽFig. 1c., while basal levels of ir ET-3, similar to those of ET-1, showed a consistent increase over time, reaching levels of about 35 pgr10 7 cells at 48 h.

258

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

3.2. Effect of ETA and ETB specific antagonists on thrombin-stimulated ET-1 concentrations in astrocyte media Thrombin Ž2.2 Urml. is known to significantly stimulate ir ET-1 but not ir ET-3 secretion from primary rat astrocytes w7x. As with basal secretion of ET-1, the thrombin-induced increase in ET-1 was time-dependent. BQ123 or BQ788 alone did not amplify the thrombin-induced stimulation of ET-1 levels in astrocyte media. A combination, however, of BQ123 and BQ788 provoked a dramatic increase in ir ET-1 in the media, reaching about 4–5 times the thrombin-stimulated concentration at 48 h ŽFig. 2a.. Again, these antagonist treatments had no effect on ir ET-3 Ždata not shown.. To further characterize the nature of ir ET-1 in astrocyte media under conditions of maximum stimulation, the results from direct radioimmunological determination of ET-1 concentrations in Sep-Pak C18 extracted media were compared to those obtained by HPLCrRIA analysis of aliquots from the same media. As shown in Fig. 2b, the main peaks of ir ET-1 in HPLC fractions co-elute with bigET-1 and ET-1 standards, respectively. At the ET-3 elution position, only low levels of Žcrossreacting. ET-1 immunoreactivity were detectable. The inset in Fig. 2b denotes ir ET-1 concentrations as measured in parallel by RIA only. These results indicate that ET-1 immunoreactivity as determined from direct RIA under stimulated conditions Žthrombin plus BQ123 plus BQ788. is due to ET-1 and its precursor bigET-1. To address the question of whether a continuous presence of thrombin is required to obtain the ET-1 levels described above, or whether thrombin initiates other mechanisms leading to a thrombin-independent stimulation of ET-1 secretion, hirudin, a specific thrombin inhibitor was used. Hirudin at a concentration of 10 ATUrml was added 2 h or 12 h after addition of thrombin to astrocyte cultures. As demonstrated in Fig. 2c, hirudin, added after 2 h, is capable of almost completely antagonizing the effect of thrombin. Even when added as late as 12 h after thrombin, hirudin still remarkably reduces the thrombin effect, suggesting that for a full stimulation to occur, the presence of thrombin is required over the entire period of 48 h. 3.3. Transcription of the gene encoding ET-1 in astrocytes following addition of thrombin and ET antagonists In order to investigate whether the increase in ir ET-1 in the media might be related to an increase in ET-1 mRNA expression, RT PCR analysis has been performed. Using primers for lamin-b as the reference gene together with specific primers for ET-1 within the same PCR reaction vial, no alteration in ET-1 transcription could be detected under any of the stimulated conditions of this study. Fig. 2d shows the RT PCR results of a typical experiment. Addition of thrombin plus BQ123 plus BQ788 to primary rat astrocytes did not affect ET-1 mRNA after 3 h, 8 h, or 24 h.

Fig. 3. Clearance of ir ET-1 from astrocyte media after addition of 250 pg ET-1: effect of ET receptor antagonists. The data shown are from one representative out of three experiments. Ž‘Clearance’sdifference between the 250 pg ET-1 added at time point 0 and the amount of ir ET-1 measured in the media at the indicated time points..

3.4. Rapid clearance of exogenous ET-1 from astrocyte culture media: effect of ET antagonists As presented in Fig. 3, ET-1 added to astrocyte cultures Ž250 pgr6 = 10 6 cells. rapidly disappeared from the media with less than 5% remaining detectable after 1 h. In fact, at ‘time point 0’, i.e., time of addition and immediate removal of ET-1, only 160 pg of the 250 pg were recovered. Addition of BQ123 to the media did not measurably alter these kinetics. Addition of BQ788 appeared to delay the clearance of ET-1, whereas the combination of BQ123 plus BQ788 resulted in a distinct inhibition of ET-1 clearance from the media ŽFig. 3.. Recovery of ET-1 from cell-free Žastrocyte-conditioned and plain. media was constantly over 85%. 3.5. Competition of ET antagonists with I binding to rat astrocytes

12 5

ET-1 for

In order to retain ‘physiological’ conditions for bindingrcompetition studies, living adherent cells were incubated for 2 h at 378C with I 125 ET-1 and various concentrations of the respective cold ligands. As shown in Fig. 4, BQ123 up to a concentration of 10 m M did not compete, and BQ788 showed weak competition Ž K i f 2.5 m M. with

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

259

Fig. 4. Competition of ET-1 and ET antagonists with I 125 ET-1 for specific binding to primary rat astrocytes. Cells were incubated for 2 h at 378C with I 125 ET-1 and the respective cold ligands. In the combined competition of antagonists for I 125 ET-1 binding, BQ788 was added at increasing concentrations with fixed concentrations of BQ123 Ž10y7 M.. Shown is one representative out of three experiments.

I 125 ET-1 for binding to astrocytic ET receptors. In contrast, the stable addition of 10y7 M BQ123 to varying concentrations of BQ788 resulted in a distinct shift of the competition curve to the left Ž K i f 250 nM.. A comparable shift was observed when 10y7 M BQ788 was stably added to varying concentrations of BQ123 Ždata not shown.. The K D value for ET-1 under these assay conditions was f 1.6 nM.

4. Discussion This study demonstrates for the first time a synergistic action of ETA and ETB antagonists in elevating ET-1r bigET-1 levels in cell culture media of primary rat astrocytes both under basal and stimulated conditions. Neither one of the antagonists alone was capable of inducing a comparable ET-1 elevation. This elevation was selective, affecting only ir ET-1 but not ir ET-3. The increase in ET-1 caused by the combined action of ETA and ETB antagonists reached four times the control level. Such an increase may be due to an increased production or to a reduction in cellular bindingruptake or in degradation. An increase in production could be the result of a desinhibitory effect of ET antagonists on an autocrine negative feedback action of ET-1 on its own production. However, an increase in ET-1 mRNA was not detectable by RT PCR under these conditions. These results are consistent with recent work by others showing in endothelial cells in vitro w24x as well as in human plasma in vivo

w26x that ET antagonist-induced increases in ir ET-1 were not accompanied by a rise in peptide synthesis. Nevertheless, our HPLC data, yielding a considerable increase both in the precursor peptide, bigET-1, as well as in ET-1, suggest that the effect of the antagonist combination may be partially attributable to an increase in ET-1 synthesis. Such an increase in synthesis could well occur at the posttranscriptional level. The results of our ET-1 challenge experiment, however, favour a competition of antagonists with ET-1 for binding and uptake as the predominant mechanism leading to high ET-1 levels in the media. Here, again, a combination of ETA and ETB antagonists is required to achieve a maximum delay in ET-1 clearance from the media. The observation that, in combination with BQ788, peptidic and non-peptidic ETA antagonists had the same effect on ir ET-1 levels makes a competition of the peptides BQ123 and BQ788 with endogenous ET-1 for degradation by endopeptidases w33x as a further cause of ET-1 elevation unlikely. The lack of potentiation of ir ET-3 levels by the combined antagonists is not explained at this point, but could be due to the predominant presence of bigET-3 in astrocyte cultures accounting for ET-3 immunoreactivity. In this case, ET antagonists would not show any significant competition Žand therefore not increase ir ET-3., since bigET-3 has much less affinity for ET receptors. The question also remains, why in primary rat astrocytes increased ir ET-1 concentrations were observed only in the presence of a combination of ETA and ETB antagonists. In endothelial cells, known to express predominantly recep-

260

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

tors of the ETB subtype w11x, incubation with BQ788 was followed by significant increases of ir ET-1 in the supernatant, while the ETA antagonist BQ123 had no effect. A combination of the two antagonists has not been tested in this study w24x. There are several possible explanations for the phenomenon observed in astrocytes: Ž1. The concentration of antagonists of 10y6 M as used in this study may not have been sufficient. This concentration, however, has been shown to be effective in other systems whereas increasing the concentration over 10y6 M was found to produce ET antagonistic effects at the respective other ET receptor subtype w14,16,17x. Ž2. The fact that neither one of the ET antagonists by itself, although present in sufficient concentration, has a significant effect on extracellular ET-1 levels, suggests that the respective unblocked receptor subtype can compensate for the other. Indeed, we have previously described the expression of mRNA for ETA as well as ETB receptors in primary rat astrocyte cultures w7x. Ž3. Assuming that in primary rat astrocytes functional receptors of both the ETA and ETB subtype are present, they could be expressed simultaneously by the same cells or belong to distinct subpopulations of cells within our primary cultures. None of these possibilities can be entirely excluded based on the results of the present study. However, our binding data, in agreement with previous findings w8,13,20x, fit best with the assumption of a single binding site. Moreover, while BQ788 weakly competed with I 125 ET-1 for specific binding, BQ123 did not compete at all. This latter finding would be unexpected if functional ETA receptors were present. Surprisingly, the weak competition of BQ788 was augmented in the presence of a fixed concentration of BQ123 Ž10y7 M.. Taken together, the synergism of ET receptor antagonists on binding as well as on the concentration of ir ET-1 in the supernatant of primary rat astrocyte cultures appears to be due to an unusual ET binding outside the classical ETA rETB scheme. Systemic administration of combined ET antagonists has been shown to lead to remarkable increases of plasma ir ET-1, and this phenomenon has been interpreted as the result of a competition of antagonists for ET binding sites w21,26x. Correspondingly, non-selective blockade of ETA and ETB receptors may provoke a drastic elevation of intracerebral ET-1 concentrations, thereby increasing the vasoconstrictor tone of the cerebral vasculature. In animal models of cerebral ischemia, ET-1 levels in brain tissue and CSF increase considerably w3,34x, heavily challenging the brain ET eliminatory system. Hereby, a blockade of the ET eliminatory capacity of astrocytes may not be desirable. In fact, it may explain the failure of a non-selective ET antagonist, bosentan, to modify hypoperfusion and reduce neuronal damage after transient ischemia in the rat w25x. To conclude, this study suggests an advantage of ETA selective antagonists, that would block ETA mediated vasoconstriction without affecting the ET-1 eliminatory capacity of astrocytes, in future treatment of ischemic brain disease.

Acknowledgements Mrs. D. Hesse’s help with the HPLC analysis is gratefully acknowledged. We thank Knoll for providing LU135252 and Behringwerke for providing hirudin. This work has been supported by a grant from the German Research Foundation ŽDFG; Eh133r1-2, 3.. A.-L. Siren ´ is supported by the Alexander-von-Humboldt-Foundation.

References w1x H. Arai, S. Hori, I. Aramori, H. Ohkubo, S. Nakanishi, Cloning and expression of a cDNA encoding an endothelin receptor, Nature 348 Ž1990. 730–732. w2x T. Asano, I. Ikegaki, Y. Suzuki, S. Satoh, M. Shibuya, Endothelin and the production of cerebral vasospasm in dogs, Biochem. Biophys. Res. Commun. 159 Ž1989. 1345–1351. w3x F.C. Barone, M.Y. Globus, W.J. Price, R.F. White, B.L. Storer, G.Z. Feuerstein, R. Busto, E.H. Ohlstein, Endothelin levels increase in rat focal and global ischemia, J. Cereb. Blood Flow Metab. 14 Ž1994. 337–342. w4x M. Clozel, V. Breu, G.A. Gray, B. Kalina, B.M. Loffler, K. Burri, ¨ J.M. Cassal, G. Hirth, M. Muller, W. Neidhart et al., Pharmacologi¨ cal characterization of bosentan, a new potent orally active nonpeptide endothelin receptor antagonist, J. Pharmacol. Exp. Ther. 270 Ž1994. 228–235. w5x M. Dihanich, M. Kaser, E. Reinhard, D. Cunningham, D. Monard, Prothrombin mRNA is expressed by cells of the nervous system, Neuron 6 Ž1991. 575–581. w6x H. Ehrenreich, R.W. Anderson, Y. Ogino, P. Rieckmann, T. Costa, G.P. Wood, J.E. Coligan, J.H. Kehrl, A.S. Fauci, Selective autoregulation of endothelins in primary astrocyte cultures: endothelin receptor-mediated potentiation of endothelin-1 secretion, New Biol. 3 Ž1991. 135–141. w7x H. Ehrenreich, T. Costa, K.A. Clouse, R.M. Pluta, Y. Ogino, J.E. Coligan, P.R. Burd, Thrombin is a regulator of astrocytic endothelin1, Brain Res. 600 Ž1993. 201–207. w8x H. Ehrenreich, J.H. Kehrl, R.W. Anderson, P. Rieckmann, L. Vitkovic, J.E. Coligan, A.S. Fauci, A vasoactive peptide, endothelin3, is produced by and specifically binds to primary astrocytes, Brain Res. 538 Ž1991. 54–58. w9x H. Ehrenreich, M. Lange, K.A. Near, F. Anneser, L.A. Schoeller, R. Schmid, P.A. Winkler, J.H. Kehrl, P. Schmiedek, F.D. Goebel, Long term monitoring of immunoreactive endothelin-1 and endothelin-3 in ventricular cerebrospinal fluid, plasma, and 24-h urine of patients with subarachnoid hemorrhage, Res. Exp. Med. Berl. 192 Ž1992. 257–268. w10x J.D. Firth, P.J. Ratcliffe, Organ distribution of the three rat endothelin messenger RNAs and the effects of ischemia on renal gene expression, J. Clin. Invest. 90 Ž1992. 1023–1031. w11x Y. Fujitani, K. Oda, M. Takimoto, T. Inui, T. Okada, Y. Urade, Autocrine receptors for endothelins in the primary culture of endothelial cells of human umbilical vein, FEBS Lett. 298 Ž1992. 79–83. w12x T. Fukuroda, T. Fujikawa, S. Ozaki, K. Ishikawa, M. Yano, M. Nishikibe, Clearance of circulating endothelin-1 by ETB receptors in rats, Biochem. Biophys. Res. Commun. 199 Ž1994. 1461–1465. w13x H. Hama, T. Sakurai, Y. Kasuya, M. Fujiki, T. Masaki, K. Goto, Action of endothelin-1 on rat astrocytes through the ETB receptor, Biochem. Biophys. Res. Commun. 186 Ž1992. 355–362. w14x M. Ihara, K. Noguchi, T. Saeki, T. Fukuroda, S. Tsuchida, S.

M. Hasselblatt et al.r Brain Research 785 (1998) 253–261

w15x

w16x

w17x

w18x

w19x

w20x w21x

w22x

w23x

w24x

w25x

Kimura, T. Fukami, K. Ishikawa, M. Nishikibe, M. Yano, Biological profiles of highly potent novel endothelin antagonists selective for the ETA receptor, Life Sci. 50 Ž1992. 247–255. A. Inoue, M. Yanagisawa, S. Kimura, Y. Kasuya, T. Miyauchi, K. Goto, T. Masaki, The human endothelin family: three structurally and pharmacologically distinct isopeptides predicted by three separate genes, Proc. Natl. Acad. Sci. USA 86 Ž1989. 2863–2867. K. Ishikawa, M. Ihara, K. Noguchi, T. Mase, N. Mino, T. Saeki, T. Fukuroda, T. Fukami, S. Ozaki, T. Nagase et al., Biochemical and pharmacological profile of a potent and selective endothelin B-receptor antagonist, BQ-788, Proc. Natl. Acad. Sci. USA 91 Ž1994. 4892–4896. H. Karaki, S.A. Sudjarwo, M. Hori, Novel antagonist of endothelin ETB1 and ETB2 receptors, BQ-788: effects on blood vessel and small intestine, Biochem. Biophys. Res. Commun. 205 Ž1994. 168– 173. C. Koseki, M. Imai, Y. Hirata, M. Yanagisawa, T. Masaki, Autoradiographic distribution in rat tissues of binding sites for endothelin: a neuropeptide?, Am. J. Physiol. 256 Ž1989. R858–R866. Y. Koyama, A. Baba, Endothelins are extracellular signals modulating cytoskeletal actin organization in rat cultured astrocytes, Neuroscience 61 Ž1994. 1007–1016. E.R. Levin, H.J. Frank, A. Pedram, Endothelin receptors on cultured fetal rat diencephalic glia, J. Neurochem. 58 Ž1992. 659–666. B.M. Loffler, V. Breu, M. Clozel, Effect of different endothelin ¨ receptor antagonists and of the novel non-peptide antagonist Ro 46-2005 on endothelin levels in rat plasma, FEBS Lett. 333 Ž1993. 108–110. M.W. MacCumber, C.A. Ross, S.H. Snyder, Endothelin in brain: receptors, mitogenesis, and biosynthesis in glial cells, Proc. Natl. Acad. Sci. USA 87 Ž1990. 2359–2363. E.H. Ohlstein, P. Nambi, S.A. Douglas, R.M. Edwards, M. Gellai, A. Lago, J.D. Leber, R.D. Cousins, A. Gao, J.S. Frazee et al., SB 209670, a rationally designed potent nonpeptide endothelin receptor antagonist, Proc. Natl. Acad. Sci. USA 91 Ž1994. 8052–8056. S. Ozaki, K. Ohwaki, M. Ihara, T. Fukuroda, K. Ishikawa, M. Yano, ETB-mediated regulation of extracellular levels of endothelin-1 in cultured human endothelial cells, Biochem. Biophys. Res. Commun. 209 Ž1995. 483–489. T.R. Patel, J. McCulloch, Failure of an endothelin antagonist to modify hypoperfusion after transient global ischaemia in the rat, J. Cereb. Blood Flow Metab. 16 Ž1996. 490–499.

261

w26x C. Plumpton, C.J. Ferro, W.G. Haynes, D.J. Webb, A.P. Davenport, The increase in human plasma immunoreactive endothelin but not big endothelin-1 or its C-terminal fragment induced by systemic administration of the endothelin antagonist TAK-044, Br. J. Pharmacol. 119 Ž1996. 311–314. w27x K.M. Pollard, E.K.L. Chan, B.J. Grant, K.F. Sullivan, E.M. Tan, C.A. Glass, In vitro posttranslational modification of lamin B cloned from a human T-cell line, Mol. Cell. Biol. 10 Ž1990. 2164–2175. w28x H. Riechers, H.P. Albrecht, W. Amberg, E. Baumann, H. Bernard, H.J. Bohm, D. Klinge, A. Kling, S. Muller, M. Raschack, L. Unger, ¨ N. Walker, W. Wernet, Discovery and optimization of a novel class of orally active nonpeptidic endothelin-A receptor antagonists, J. Med. Chem. 39 Ž1996. 2123–2128. w29x T. Sakurai, M. Yanagisawa, A. Inoue, U.S. Ryan, S. Kimura, Y. Mitsui, K. Goto, T. Masaki, cDNA cloning, sequence analysis and tissue distribution of rat preproendothelin-1 mRNA, Biochem. Biophys. Res. Commun. 175 Ž1991. 44–47. w30x T. Sakurai, M. Yanagisawa, Y. Takuwa, H. Miyazaki, S. Kimura, K. Goto, T. Masaki, Cloning of a cDNA encoding a non-isopeptideselective subtype of the endothelin receptor, Nature 348 Ž1990. 732–735. w31x J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning—a Laboratory Manual, 2nd edn., Cold Spring Harbor Laboratory Press, New York, 1989, pp. 7.19–17.22. w32x M. Sensenbrenner, G. Devilliers, E. Bock, A. Porte, Biochemical and ultra-structural studies of cultured rat astroglial cells: effect of brain extract and dibutyryl cyclic AMP on glial fibrillary acidic protein and glial filaments, Differentiation 17 Ž1980. 51–61. w33x M. Sokolovsky, R. Galron, Y. Kloog, A. Bdolah, F.E. Indig, S. Blumberg, G. Fleminger, Endothelins are more sensitive than sarafotoxins to neutral endopeptidase: possible physiological significance, Proc. Natl. Acad. Sci. USA 87 Ž1990. 4702–4706. w34x M. Spatz, D. Stanimirovic, A. Strasser, R.M. McCarron, Nitro-Larginine augments the endothelin-1 content of cerebrospinal fluid induced by cerebral ischemia, Brain Res. 684 Ž1995. 99–102. w35x S. Supattapone, A.W. Simpson, C.C. Ashley, Free calcium rise and mitogenesis in glial cells caused by endothelin, Biochem. Biophys. Res. Commun. 165 Ž1989. 1115–1122. w36x M. Yanagisawa, H. Kurihara, S. Kimura, Y. Tomobe, M. Kobayashi, Y. Mitsui, Y. Yazaki, K. Goto, T. Masaki, A novel potent vasoconstrictor peptide produced by vascular endothelial cells, Nature 332 Ž1988. 411–415.