UII (urotensin II) and its paralogue URP (UII-related peptide) are two vasoactive neuropeptides whose respective central actions are currently unknown. In the present study, we have compared the mechanism of action of URP and UII on cultured astrocytes. Competition experiments performed with [125I]UII showed the presence of very-high- and high-affinity binding sites for UII, and a single high-affinity site for URP. Both UII and URP provoked a membrane depolarization accompanied by a decrease in input resistance, stimulated the release of endozepines, neuropeptides specifically produced by astroglial cells, and generated an increase in [Ca2+]c (cytosolic Ca2+ concentration). The UII/URP-induced [Ca2+]c elevation was PTX (pertussis toxin)-insensitive, and was blocked by the PLC (phospholipase C) inhibitor U73122 or the InsP3 channel blocker 2-APB (2-aminoethoxydiphenylborane). The addition of the Ca2+ chelator EGTA reduced the peak and abolished the plateau phase, whereas the T-type Ca2+ channel blocker mibefradil totally inhibited the Ca2+ response evoked by both peptides. However, URP and UII induced a mono- and bi-phasic dose-dependent increase in [Ca2+]c and provoked short- and long-lasting Ca2+ mobilization respectively. Similar mono- and bi-phasic dose-dependent increases in [3H]inositol incorporation into polyphosphoinositides in astrocytes was obtained, but the effect of UII was significantly reduced by PTX, although BRET (bioluminescence resonance energy transfer) experiments revealed that both UII and URP recruited Gαo-protein. Finally, UII, but not URP, exerted a dose-dependent mitogenic activity on astrocytes. Therefore we described that URP and UII exert not only similar, but also divergent actions on astrocyte activity, with UII exhibiting a broader range of activities at physiological peptide concentrations.
UII (urotensin II) is a cyclic neuropeptide that was initially isolated from the urophysis, a neurosecretory organ located in the caudal portion of the spinal cord of teleost fish . Orthologues of UII have been subsequently characterized in the brain and spinal cord of various species of tetrapods, including frogs, mice, rats, pigs and humans . A paralogue of UII called URP (UII-related peptide) has been identified in rats and humans . UII and URP share a fully conserved C-terminal cyclic hexapeptide CFWKYC core that plays a major role in the biological activity of the peptides . The cyclic region of UII and URP exhibits some structural similarities to that of somatostatin (Figure 1) and it has been found that the genes encoding these three neuropeptides originate from a common ancestor .
Comparison of the primary structures of rat and human UII, URP and somatostatin-14
The mRNAs encoding the UII and URP precursors have been localized both in the brain and in peripheral organs . In the rat and mouse CNS (central nervous system), the genes for UII and URP are primarily expressed in subpopulations of motoneurons located in discrete brainstem nuclei and in the ventral horn of the spinal cord [6,7]. In the human brain, high levels of URP mRNA are also present in the hippocampus and thalamus . It has been shown that a mature form of UII in the rat brain could only imply atypical cleavage sites in prepro-UII, and that URP might be the sole UII-immunoreactive substance present in the rodent brain .
The biological action of UII and URP are mediated through activation of a common G-protein-coupled receptor originally called GPR14 [3,9] and now renamed UT (UII receptor) . In UT-transfected CHO (Chinese-hamster ovary) and HEK (human embryonic kidney)-293 cells, UII stimulates the PLC (phospholipase C) pathway and causes an increase in [Ca2+]c (cytosolic Ca2+ concentration) . UII also exerts mitogenic activity in UT-transfected cells and in vascular smooth muscle cells expressing native UT [10,11].
In the CNS, UT mRNA has been detected in motoneurons of the spinal cord in mice  and rats  as well as in the thalamus, superior occipital gyrus and substantia nigra in humans . Previous studies have shown that central administration of UII modulates cardiovascular, locomotor, behavioural and endocrine functions [14,15]. Although URP has been proposed as the sole endogenous ligand for urotensinergic receptors in the brain , little is known regarding the central activities of URP.
It is now clearly established that astroglial cells release various neuroactive compounds , including the gliopeptides endozepines. Endozepines were originally characterized as endogenous ligands for benzodiazepine receptors . All endozepines characterized so far derive from an 86-amino-acid precursor called DBI (diazepam-binding inhibitor) which generates several biologically active fragments, notably the ODN (octadecaneuropeptide) DBI33–50 . The release of endozepines from astrocytes is regulated by several neuropeptides such as PACAP (pituitary adenylate cyclase-activating polypeptide) and somatostatin [19,20].
The expression of UII and its receptor has been previously reported in a human glioblastoma cell line  and the presence of functional UT has been demonstrated in astrocytes [22,23]. In astroglial cells, UII specifically binds two affinity sites, stimulates PIP (polyphosphoinositide) metabolism and causes a robust increase in [Ca2+]c [22,24]. Until now, nothing is known regarding the mechanism of action of URP on native cells, and the relevance for the existence of a redundant peptide for a unique UT remains unexplored. In the present study, we have characterized functional binding sites for URP on rat cortical astrocytes and compared the effects of URP with those of UII on electrophysiological activity, endozepine release, Ca2+, PIP turnover and cell proliferation.
URP (ACFWKYCV), rat UII (UII, pQHGTAPECFWKYCI) (pQ, pyroglutamic acid; underlining indicates the conserved hexapeptide core) and rat ODN DBI33–50 were synthesized using the solid-phase methodology on a Pioneer PerSeptive Biosystem peptide synthesizer using the standard manufacturer's procedures as described previously . All peptides were purified (>98%) on a 2.2 cm×25 cm Vydac C18 column and characterized by MALDI–TOF (matrix-assisted laser desorption ionization–time-of-flight) MS on a Voyager DE-PRO mass spectrometer. Glutamine, Hepes, the antibiotic/antimycotic solution and bovine γ-globulins were purchased from Invitrogen. DMEM (Dulbecco's modified Eagle's medium), Ham's F12 culture medium, insulin, D(+)-glucose, mibefradil, nifedipine, ω-conotoxin GVIA, PTX (pertussis toxin), ET-1 (endothelin-1) and BSA were obtained from Sigma–Aldrich. FBS (fetal bovine serum) and IL-1β (interleukin-1β) were from Eurobio. Fluo-4 AM (acetoxymethyl ester) and fura-2 AM were from Molecular Probes. Somatostatin-14 and [Tyr0,D-Trp8]somatostatin-14 were purchased from NeoMPs. PACAP was generously provided by Professor Alain Fournier (INRS, Pointe-Claire, Canada).
Purified cultures of astrocytes were prepared as described previously . Briefly, cerebral hemispheres from newborn Wistar rats were collected in DMEM/Ham's F12 (2:1, v/v) supplemented with 2 mM glutamine, 1% insulin, 5 mM Hepes, 0.4% glucose and 1% antibiotic/antimycotic solution. The tissues were disaggregated mechanically with a syringe equipped with a 1-mm gauge needle, and filtered through a 100-μm pore size mesh filter (Falcon, Becton Dickinson). Dissociated cells were resuspended in culture medium supplemented with 10% (v/v) heat-inactivated FBS and seeded in 150-cm2 culture flasks (Falcon) at the density of 2×107 cells per flask. The cells were incubated at 37 °C in a humidified atmosphere (5% CO2) and the medium was changed twice a week. When cultures were confluent, the flasks were gently shaken on an orbital shaker at 250 rev./min for 2 h. Dislodged cells were discarded, and a second step of purification was performed at 250 rev./min for 14–16 h. Remaining adhesive cells were collected by trypsinization, centrifuged at 100 g for 10 min and plated in 150-cm2 flasks. Suspended astrocytes were harvested and seeded in 96-well plates (for Ca2+ studies), 24-well plates (for binding studies), 14-mm glass coverslips (for electrophysiological experiments and [Ca2+]c measurements), 35-mm-diameter dishes (for measurement of PIP turnover) and 60-mm-diameter dishes (for measurement of endozepine release). The purity of the cultures was assessed by GFAP (glial fibrillary acidic protein) immunostaining which showed that enriched cultures contained >98% astrocytes .
Animal manipulations were carried out according to the recommendations of the French Ethical Committee and under the supervision of authorized investigators (P. Gandolfo; authorization no. 76.A.25 from the Ministère de l'Alimentation, de l'Agriculture et de la Pêche). All experiments were conducted according to the French and European guidelines for the care and use of laboratory animals (Council Directive 86/609/EEC; licence no. 21CAE035).
A 3 μg sample of UII or [Tyr0,D-Trp8]somatostatin-14 in phosphate buffer (0.375 mM, pH 7.4) was labelled with 0.5 mCi of Na125I (GE Healthcare) using the lactoperoxidase method as described previously . Mono-iodinated [125I]UII or [125I-Tyr0,D-Trp8]somatostatin-14 used for the radioligand-binding assays were purified by reverse-phase HPLC on an Adsorbosphere C18 column (0.46 cm×25 cm, Alltech) using a linear gradient (25–65% over 40 min) of acetonitrile/trifluoroacetic acid (99.9:0.1, v/v) at a flow rate of 1 ml/min, and stored at 4 °C. The specific radioactivity of each radioiodinated peptide was approx. 2000 Ci/mmol.
Cultured astrocytes were rinsed three times with PBS, dried under a cold air stream and stored at −80 °C until binding experiments. Frozen cells were washed twice with assay buffer (50 mM Tris/HCl buffer, 1 mM MnCl2 and 0.5% BSA, pH 7.35) and incubated at 22 °C in the same buffer in the presence of radiolabelled peptide. At the end of the incubation, cells were washed three times with assay buffer, solubilized with 1% SDS, and the radioactivity was counted in a γ counter (LKB Wallac). For competition experiments, cells were incubated for 3 h with [125I]UII (0.2 nM) or [125I-Tyr0,D-Trp8]somatostatin-14 (0.4 nM) in the presence of graded concentrations of unlabelled UII, URP or somatostatin-14. Non-specific binding was determined by addition of 1 μM unlabelled UII or [Tyr0,D-Trp8]somatostatin-14.
Measurement of [Ca2+]c
Purified astrocytes were plated into 96-well plates at 30000 cells/well or on 14-mm glass coverslips in complete medium supplemented with FBS (10%). After 2 days in culture, astrocytes were rinsed twice with a modified HBSS (Hanks balanced salt solution) containing 135 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 10 mM Hepes, 3 mM glucose and 2.5 mM probenicid (pH 7.4). Cells were incubated at 37 °C with 40 μl of 2 μM fluo-4 AM or fura-2 AM dye containing 20% pluronic acid for 40 min in a 5% CO2 atmosphere. Cells were washed twice with modified HBSS, and the effects of graded concentrations of URP and UII on [Ca2+]c were measured with a fluorimetric imaging plate reader FlexStation II (Molecular Devices). Agonist-induced changes in [Ca2+]c were also measured (fluorescence at 510 nm following excitation at 340/380 nm from a 300 W xenon lamp) in individual cells from a single visual field using an integrating CCD (charge-coupled device) camera coupled to a Leica DMLFSA microscope. Cell Ca2+ imaging and fluorescence quantification were carried out using the Metafluor Imaging System (Molecular Devices).
Measurement of PIP metabolism
Cultured astrocytes [2–5 DIV (days in vitro)] were incubated at 37 °C with 10 μCi/ml myo-[3H]inositol in glucose-, insulin- and serum-free medium for 80–120 min. The incubation was stopped by removing the medium and adding ice-cold 10% trichloroacetic acid. The cells were homogenized and centrifuged at 15000 g for 10 min at 4 °C. The pellet was washed with 500 μl of chilled water and [3H]PIPs were extracted with 500 μl of chloroform/ethanol (2:1, v/v). The radioactivity was counted in a β counter (LKB 1217 Rack Beta; EG&G Wallac).
Electrophysiological recordings were performed at room temperature (22 °C) on 2–4 DIV-cultured astrocytes seeded on glass coverslips using the standard amphotericin B-perforated configuration of the patch-clamp technique. The patch pipettes were fabricated from 1.5 μm (outer diameter) soft glass tubes on a two-step vertical pipette puller (List-Medical, L/M-3P-A). Patch electrodes had a final resistance of 5–7 MΩ. The internal pipette solution contained 130 mM potassium methylsulfate, 20 mM KCl, 5 mM MgCl2, 1 mM EGTA and 10 mM Hepes (adjusted to pH 7.4 with KOH). Before each experiment, the medium was replaced by a bath solution of 145 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl2, 0.8 mM MgCl2, 10 mM Hepes and 10 mM glucose (pH 7.4 adjusted with NaOH). Cells were transferred to the stage of an upright microscope (Leica DMLFSA). Voltage signals were recorded from an Axopatch 200A amplifier (Axon Instruments) and filtered at 2 kHz (3 db, four-pole, low-pass Bessel filter). Cell input resistance was monitored by applying hyperpolarizing pulses (2 pA, 10 ms, 0.125 Hz) to the cell under study. Data acquisition and analysis were performed through a digidata 1200 interface using the pClamp 8 suite of programs (Axon Instruments) and the Origin 4.1 analysis software (Microcal Software).
Measurement of endozepine release
Astrocytes (5 DIV) were incubated for 60 min at 37 °C with fresh serum-free medium in the absence or presence of test substances. Peptides contained in culture medium were concentrated on Sep-Pak C18 cartridges (Waters). Bound material was eluted with 50% (v/v) acetonitrile/water containing 0.1% TFA and were dried by vacuum centrifugation (SpeedVac concentrator, Savant) until RIA (radioimmunoassay).
The concentrations of ODN-like immunoreactivity (ODN-LI) in the Sep-Pak-pre-purified samples were quantified by RIA using an antiserum raised against synthetic rat ODN . The analogue [Tyr0]ODN was iodinated using the chloramine-T procedure and purified on a Sep-Pak C18 cartridge. The final dilution of the ODN antiserum was 1:30000 and the total amount of tracer was 6000 c.p.m./tube. The antibody-bound ODN fraction was precipitated by addition of 100 μl of bovine γ-globulins (1%, w/w) and 2 ml of poly(ethylene glycol) 8000 (20%, w/v). After centrifugation at 3000 g for 30 min, the pellet containing the bound fraction was counted in a γ counter.
Plasmid construction, HEK-293 transfection and BRET (bioluminescence resonance energy transfer) assay
hUT (human UT) and PAR-1 (protease-activated receptor 1) in pcDNA3.1 were purchased from the cDNA Resources Center. The ability of receptors to activate Gi/o-proteins was controlled by measuring BRET between G-protein subunits. Thus we used Gα–Rluc (Renilla reniformis luciferase) fusion proteins, with Rluc fused to the G-protein α subunit Gαo in the loop between helices αA and αB in the h-α domain , and the Venus-tagged Gγ2 subunit generously provided by Dr C. Galès (Inserm U858, Toulouse, France).
HEK-293 cells were grown in complete medium [DMEM supplemented with 10% (v/v) FBS, 4.5 g/l glucose, 100 units/ml penicillin, 0.1 mg/ml streptomycin and 1 mM glutamine (Invitrogen)]. Transient transfections were performed with the indicated plasmids using Lipofectamine™ 2000 (Invitrogen) following the manufacturer's protocol. All constructs were verified by sequencing, and their expression in HEK-293 cells was confirmed by BRET signals.
For BRET experiments, 48 h after transfection, HEK-293 cells were washed with PBS, and coelenterazine h substrate (Promega) was added at a final concentration of 5 μM in the total volume of 30 μl/well at 37 °C. For kinetic analysis, BRET readings were then performed immediately after coelenterazine addition. After BRET signal stabilization (≈100 s), the different ligands were injected by the Mithras LB 940 plate reader (Berthold Biotechnologies). BRET signals were expressed in milliBRET units of BRET ratio as described previously .
To determine cell survival and proliferation, 2-DIV-cultured astrocytes were grown in the complete culture medium supplemented with 10% (v/v) FBS. Cells were either incubated in the absence or presence of URP, UII, ET-1, PACAP or IL-1β for 24 h in the same complete medium [WST-1 (water-soluble tetrazolium salt 1) assay, 2 h] or rinsed in FBS-free medium and incubated after 24 h in the absence or presence of URP or UII for 48 h (cell counting). Astrocytes were rinsed three times in PBS (37 °C), and cell proliferation was quantified by fluorescence measurement with a FL600 microplate reader (450/750 nm) or by counting the cell number with an electronic cell counter (Z2, Beckman Coulter).
To calculate the Hill coefficient (h), competition curves were first fitted to the Hill equation via a non-linear regression using the computerized curve-fitting package GraphPad Prism version 4 for Windows. The data were then re-fitted using equation for displacement of radioligand by competitors to one or two binding sites. A two-site curve-fitting model was retained when this model fitted the data significantly better than a one-site model, as determined by a F-test at a significance level of P<0.05. IC50 values derived from these latter fits were converted into apparent competition-receptor dissociation equilibrium constants K1 for the very-high-affinity and K2 for the high-affinity binding sites. For biological activities, values presented in the Figures are means±S.E.M. Statistical significance of differences was determined by using the one-way ANOVA followed by a Dunnett's post-hoc test or the non-parametric Friedman's test.
Binding characteristics of URP on cultured rat astrocytes
Competition experiments performed with [125I]UII (0.2 nM) as a tracer and unlabelled UII (1 pM–1 μM) generated a biphasic curve with a Hill coefficient of 0.81±0.06 (Figure 2A). In contrast, graded concentrations of unlabelled URP (1 pM–1 μM) generated a monophasic displacement curve with a Hill coefficient close to 1. Statistical analysis of the competition experiments using the F-test showed that data best fitted with a two-site model (P<0.05) and the calculated dissociation constants for the very-high- (K1) and high- (K2) affinity sites were 0.31±0.12 and 32.9±18.7 nM for UII and with a one-site model (P<0.05) yielding a dissociation constant for a high-affinity (K) site of 1.62±0.74 nM for URP (Figure 2A, lower panel). We thus tested the ability of URP and UII, which exhibit sequence similarities with somatostatin in their cyclic portion, to displace the binding of the somatostatin analogue [125I-Tyr0,D-Trp8]somatostatin-14 (0.4 nM). Graded concentrations of unlabelled URP (10 pM–10 μM) showed a biphasic competition curve with a Hill coefficient of 0.32±0.09 (Figure 2B, lower panel) and dissociation constants for high- (K1) and low- (K2) affinity sites of 9.16±5.32 nM and 4.42±1.72 μM respectively. In contrast, UII (3 nM–30 μM) failed to displace [125I-Tyr0,D-Trp8]somatostatin-14 below micromolar concentrations (Figure 2B).
Displacement curves and binding parameter tables comparing competition of [125I]UII and [125I-Tyr0,
D-Trp8]somatostatin-14 binding by UII, URP and somatostatin-14 on cultured rat astrocytes
Similar effects of URP and UII on resting membrane potential and endozepine release from cultured rat astrocytes
To investigate the action of URP and UII on the electrical activity of cultured astrocytes, amphotericin-perforated patch-clamp recordings were performed. Under current-clamp conditions (holding potential at −52.12±3.08 mV; n=32), focal application of URP (Figure 3A, upper panel; 100 nM; 2 min) or UII (Figure 3A, lower panel; 100 nM; 2 min) elicited a membrane depolarization in 69% of astrocytes (20/29) accompanied by a decrease in the input resistance. The majority of responding astrocytes remained in a depolarized state during the washout period (Figure 3A). The amplitudes of membrane depolarization induced by URP at concentrations of 10 nM (9.6±2.6 mV, n=5) and 100 nM (16.6±4.3 mV, n=8) were very similar to those evoked by UII (10 nM, 9±4 mV, n=3; 100 nM, 15.5±2.8 mV, n=12, Figure 3B).
Effects of URP and UII on the resting membrane potential and on the release of ODN-LI from cultured rat astrocytes
Graded concentrations of URP (1 pM–10 μM, 1 h) induced a dose-related increase in the release of ODN-LI with an EC50 value of 3.95±1.52 nM and a maximum effect at a concentration of 1 μM (results not shown). Incubation of cells with 0.1 nM, 10 nM and 1 μM URP or UII provoked a concentration-dependent increase in ODN-LI release, non-significantly different between URP and UII (Figure 3C).
Similar mechanism of URP and UII on [Ca2+]c mobilization in cultured rat astrocytes
URP and UII (0.1 μM each) induced a rapid [Ca2+]c increase (peak) followed by a sustained response (plateau) . Pre-incubation of astrocytes with the Gi/o-protein inhibitor PTX (0.2 μg/ml, 18 h) did not affect the Ca2+ responses induced by both peptides (Figure 4A). In contrast, the URP- and UII-induced Ca2+ mobilizations were completely blocked by the PLC inhibitor U73122 (10 μM, 20 min) and were strongly reduced by the InsP3-induced Ca2+ release and store-operated Ca2+ channel blocker 2-APB (2-aminoethoxydiphenylborane) (50 μM, 20 min), by 83 and 80% respectively (Figures 4B and 4C). In order to gain insight into the Ca2+ pools recruited by UII and URP, [Ca2+]c was measured in the absence or presence of the extracellular Ca2+ chelator EGTA (1 mM, 20 min). Peak responses were partially attenuated by 53% (URP) and 58% (UII) and plateau phases were totally blocked in the presence of both peptides (Figure 4D). The [Ca2+]c increase evoked by UII and URP was also suppressed in the presence of the voltage-dependent T-type Ca2+ channel blocker mibefradil (1 μM, Figure 4E, 20 min), whereas the specific L- and N-type Ca2+ channel blockers nifedipine and ω-conotoxin GVIA (1 μM each, 20 min) respectively did not significantly affect the peptide-evoked elevation of [Ca2+]c in astrocytes (Figures 4F and 4G).
Mechanisms of URP- or UII-induced increase in [Ca2+]c in cultured rat astrocytes
Divergent actions of URP and UII on [Ca2+]c mobilization and PIP turnover in cultured rat astrocytes
As shown in Figure 5(A), exposure of astrocytes to increasing concentrations of URP and UII (1 pM–10 μM each) provoked a robust and sustained increase in [Ca2+]c, followed by a plateau phase. The dose–response study of the effects of URP and UII on [Ca2+]c mobilization revealed that URP and UII induced a mono- and bi-phasic dose-dependent [Ca2+]c increase (Figure 5A). The maximum stimulations (URP, 157%; UII, 151%) were obtained at 1 μM for both peptides. The peptide-induced Ca2+ responses were also investigated at a single astrocyte level using a digital imaging system. Application of 0.1 μM URP resulted in a rapid and transient increase in [Ca2+]c levels (136.9±3.5%) in a subset of 59% astrocytes (16/24; Figure 5B). Surprisingly, the UII-evoked [Ca2+]c response was characterized by a rapid increase followed by a sustained or developing increase in the [Ca2+]c level, reaching 137.1±5.2% stimulation after 400 s of recording in 70% of cells (21/26; Figure 5C).
Divergent effects of URP and UII on [Ca2+]c in cultured rat astrocytes
As observed on Ca2+ levels, exposure of astrocytes to graded concentrations of URP (1 pM–1 μM) induced a dose-dependent increase in [3H]inositol incorporation into PIP with an EC50 value of 2.19±1.42 nM (Figure 6A). In contrast, UII (1 pM–1 μM) evoked a biphasic dose-dependent stimulation of PIP turnover, yielding an EC501 of 9.1±4.6 pM and an EC502 of 11.0±4.6 nM (Figure 6A). The amplitude of the stimulatory effect of URP (10 nM; 141.5±8.6% of control) on PIP metabolism was not significantly different from that obtained with UII (10 nM; 154.4±12.8% of control). Co-application of URP (10 nM) with a saturating concentration of UII (10 nM) induced an increase in [3H]inositol incorporation into PIP (160±12.8%) that did not significantly exceed the effect of each peptide administered alone (Figure 6B). In the same set of experiments, PTX (0.2 μg/ml; 18 h) significantly decreased the incorporation of [3H]inositol into PIP induced by UII (100 nM, −45%), but failed to significantly affect the stimulatory effect of URP (100 nM, −26%) (Figure 6C).
Distinct activities of URP and UII on PIP metabolism in cultured rat astrocytes
Specific activation of Gαo-protein by UII and URP in cells expressing recombinant hUT
To control whether UT might be coupled to Gi/o-proteins in the presence of UII, but also URP, we used the BRET approach to directly monitor Gαo-protein activation in living cells [30,31]. This assay is based on the G-protein complex conformational change and the dissociation of the Gαo fused with the energy donor Rluc, from the Gγ2 fused with the energy acceptor YFP (yellow fluorescent protein) upon receptor activation. These fused G-protein subunits were shown to be correctly expressed and to allow an efficient coupling of receptors to their effectors (results not shown). In HEK-293 cells co-expressing Gαo–Rluc, Gγ2–YFP and either recombinant hUT (HEK-293–hUT) or PAR-1 (HEK-293–PAR-1), a significant BRET signal was detected compared with cells expressing only Gαo. When HEK-293–hUT cells were treated with PBS or thrombin (50 units/ml), the BRET signal was not modified (Figure 7A). In contrast, kinetic analysis showed that UII and URP (10 μM each) injection provoked a BRET signal decrease between Gαo and Gγ2 subunits, which occurred rapidly and persisted for more than 1 min (Figure 7B). In order to control the specificity of the UT–Gαo coupling, BRET signal was measured on HEK-293–PAR-1 in the presence of thrombin, UII and URP. In these cells, thrombin induced a BRET decrease, whereas URP and UII failed to activate G-protein dissociation (Figure 7C).
BRET kinetic analysis of the effect of UII and URP on Gαo-protein activation
Effect of URP and UII on cell viability/proliferation in cultured rat astrocytes
Rat astrocytes (2 DIV) (non-synchronized) were cultured in complete medium with FBS added and then incubated in an FBS-free medium containing URP or UII (10 nM each). Cell viability was measured 24 h later by using the WST-1 assay, allowing detection of mitochondrial activity related to the number of viable cells. Figure 8(A) shows that URP failed to affect the cell viability, whereas UII stimulated astrocyte growth by 132.7%. PACAP (10 nM), IL-1β (10 ng/ml) and ET-1 (10 nM) were also tested as positive controls and gave rise respectively to 126.3, 169.3 and 171.1% increases (Figure 8A). The dose–response effects of URP and UII on cell growth were also investigated in non-synchronized (WST-1, Figure 8B) and synchronized (18 h of harvesting, counting, Figure 8C) astrocytes. A 24 h (Figure 8B) or 48 h (Figure 8C) treatment with graded concentrations of URP failed to significantly modify the rate of cell growth. In contrast, UII induced a dose-dependent cell growth, yielding an EC50 value of 2.37±0.02 nM and a maximum stimulatory effect of 132.9±7.0% at 1 μM (Figure 8B) in non-synchronized cultures, and evoked a significant cell number increase (120.1±2.9%) at 0.1 μM under synchronized conditions (Figure 8C).
Effects of UII and URP on cultured rat astrocyte viability
Although the cardiovascular effects of the urotensinergic system are now well characterized, little is known concerning the roles of UII and URP in the CNS. Immunohistochemical studies have shown the presence, in the rat brain, of UT-like immunoreactivity in GFAP-positive cells within the brainstem, hypothalamus and thalamus . In agreement with these observations, we have demonstrated previously that rat cortical astrocytes express UT mRNA, contain the UT protein and exhibit very-high- and high-affinity UII-binding sites . In these cells, UII specifically activates a PLC/PIP/Ca2+ transduction pathway, via both PTX-sensitive and -insensitive G-proteins . The identification of a UII paralogue, URP, which apparently exhibits the same pharmacological characteristics as UII , raises the question as to whether URP functions as a UII backup in the brain. In the present study, we have looked for the existence of URP-binding sites in cultured rat astrocytes and we have compared the mechanisms of action of URP and UII on astroglial cell activity.
Competition experiments revealed that URP could displace [125I]UII at nanomolar concentrations (K=1.62 nM) with a Hill coefficient close to 1, suggesting the existence of a homogeneous population of URP-binding sites in astrocytes. However, the occurrence of two affinity binding sites for UII has been previously described in CHO cells expressing recombinant hUT  and in rat cortical astrocytes . In our model, URP is 5-fold less potent in competing for the very-high-affinity UII-binding site and 20-fold more potent in displacing UII from the high-affinity site, indicating that UII and URP exhibit different binding characteristics in astrocytes. Interestingly, URP and the C-terminal octapeptide of human UII, UII4–11, were 10-fold more potent than human UII in competing with [125I]UII at recombinant UT [3,32], suggesting that the N-terminal segment of UII decreases the affinity of the peptide for its receptor. It has been shown previously that cultured rat astrocytes express UT mRNA and that UT-like immunoreactivity is located at the plasma membrane of astroglial cells [22,23]. Thus the present study suggests that the high-affinity site bound by URP actually corresponds to UT. The cyclic hexapeptide core of UII and URP exhibits substantial similarities with the biologically active region of somatostatin-14 [33,34] (Figure 1) and UT shares sequence identity with somatostatin receptors (sst) . In spite of this relatively high degree of sequence homologies, UII exhibits very low affinity for sst (sst1, sst2 and sst4) either transfected in cell lines  or naturally expressed in astrocytes [20,22,37]. The present study shows that URP could displace [125I-Tyr0,D-Trp8]somatostatin-14 binding to both high- and low-affinity sites. This observation indicates that, unlike rat UII that selectively binds to UT, URP can interact with both UT and sst expressed in rat cortical astrocytes. In agreement with these data, a comparative genomic study has established the existence of a close evolutionary relationship between urotensinergic and somatostatinergic gene families, particularly between URP and somatostatin-14 peptides . Moreover, it has recently been demonstrated that high concentration of URP stimulates [Ca2+]c levels and proliferation of recombinant sst2-expressing CHO cells . Taken together, it is suggested that shorter urotensinergic peptides in their N-terminal sequence (Figure 1; human UII and URP) were more potent than longer UII (rat UII) in competing for the high-affinity UII-binding site UT, and also acquired the ability to bind sst.
Astrocytes have long been considered as a non-excitable support of neuronal cells in the brain. It is now well established that astroglial cells express functional channels and receptors enable to sense synaptic transmission . In addition, astrocytes participate in neuronal functioning not only via feeding and structural support, but also through release of gliotransmitters such as glutamate, D-serine, ATP, cytokines or endozepines . Application of URP or UII in the vicinity of astrocytes in culture induced a slow onset of membrane depolarization accompanied by a decrease in the input resistance. It has also been demonstrated that UII depolarizes nervous cells on brainstem cholinergic neurons and evokes a membrane conductance decrease [40,41]. Astrocytes in culture or acutely dissociated from different brain regions may express various ion channels including both T- and L-type Ca2+ channels, different K+ channels and also store-operated Ca2+ channels . Thus URP as well as UII may inhibit K+ channel activity and/or stimulate voltage-activated Ca2+ channel opening either directly or upon depolarizing stimulus. Astrocytes are able to synthesize, store and release endozepines via Ca2+-dependent and/or cAMP/PKA (protein kinase A)-dependent ways . It is likely that depolarization of glial cells evoked by both peptides subserves neurotransmitter secretion, transcription or proliferation.
It is now clearly established that activation of UT expressed in native cells or in transfected cell lines is associated with an increase in PIP turnover causing mobilization of Ca2+ from intracellular stores [43,44]. Exposure of astrocytes to URP and UII was associated with [Ca2+] mobilization, i.e. a rapid peak response followed by a plateau phase. Peptide-induced [Ca2+]c increase was PLC-dependent (U73122-sensitive) and resulted from activation of InsP3 receptor pools (2-APB-sensitive). These data are in good agreement with previous observations made in sartorius muscle, vascular tissues or rabdomyosarcoma cell lines [44–46]. Since PLC-inducing Ca2+ mobilization is typically mediated via Gq- and/or Gi/o-proteins, and UT sites might be linked to Gi/o-protein types [3,22,46], the ability of URP compared with UII was tested in PTX-pre-treated astrocytes. PTX, which catalyses ADP-ribosylation of Gi/o-type G-proteins and consequently inactivates their transduction mechanisms , did not affect the Ca2+ response evoked by both peptides, indicating that URP and UII tested at the micromolar range activate a receptor coupled to a PLC/InsP3/Ca2+ pathway in a Gi/o-independent manner. The involvement of the extracellular Ca2+ pool was also tested by means of experiments in the presence of EGTA and voltage-gated Ca2+ channel blockers. It is thus demonstrated that extracellular chelation of Ca2+ attenuated the peak and totally blocked the plateau responses, indicating that the transient Ca2+ peak is partially attributable to Ca2+ release from the sarcoplasmic reticulum, whereas the sustained plateau phase is due to Ca2+ influx. Surprisingly, [Ca2+]c increase induced by both peptides was completely abolished by the specific inhibition of the T-type Ca2+ channel (mibefradil) and appeared insensitive to both blockage of L- and N-type channels by nifedipine and ω-conotoxin GVIA respectively. These observations suggest that the high level of UT activation leads to the recruitment of both extra- and intra-cellular Ca2+ pools and to a cross-talk between InsP3 receptors and at least voltage-dependent T-type channels. T-type channels are represented by three different Ca2+ channels, i.e. α1G or Cav3.1, α1H or Cav3.2 and α1I or Cav3.3 , and mRNA transcripts encoding the functional α1G T-type subunit channels have been found in cultured astrocytes . An interesting property of T-type Ca2+ channels is their ability to sustain a continuous Ca2+ influx in glia at rest by a window current mechanism . Interestingly, changes in membrane potential and intracellular Ca2+ may affect Gq-protein-coupled InsP3 production itself [51,52]. The components of the basic Gq-protein-coupled receptor/InsP3 production pathway would be located in or associated with the plasma membrane, making them potentially susceptible to changes in membrane potential lipid in the inner leaflet of the plasma membrane . These observations may unveil a similar UT–Gq-associated mechanism in astrocytes, where blockage of T-type Ca2+ channels may hyperpolarize astrocytes and inhibit InsP3-induced Ca2+ release.
The dose–response curves of URP and UII on [Ca2+]c show mono- and bi-phasic profiles respectively. The agonist potency of URP (EC50 ~0.7 nM) is similar to that one of the two sites of UII (EC50 ~0.7 nM), and in close agreement with the binding characteristic at the common high-affinity sites of URP and UII, positively coupled to a PLC–Gq-protein pathway. It thus appears that the very-high-affinity binding site of UII is not activated by URP and is functionally coupled to [Ca2+]c. Single-cell digital imaging techniques demonstrated that a majority of cells responded to URP and UII, even if the cell–cell variations in shape and amplitude can be observed. More surprisingly, URP (0.1 μM) evoked a transient and reversible increase in [Ca2+]c, whereas a single UII (0.1 μM) infusion induced a Ca2+ response that remained stable during more than 400 s of recording. This last effect may be attributable to the low dissociation rate of UII, as already described for rat and human UII on UT-transfected cells, skeletal muscle myoblasts and astrocytes [22,27,46,54]. Thus the slow dissociation might be specific to UII and would probably account for the sustained and washout-resistant contractile responses induced by UII on primate arteries and [Ca2+]c increase in rat cortical astrocytes.
As observed for Ca2+ responses, [3H]inositol incorporation in PIP shows mono- and bi-phasic profiles for URP and UII respectively. Thus URP and UII produced a concentration-dependent increase in [Ca2+]c consistent with a UT–Gq coupling to PLC and InsP3 formation-inducing Ca2+ release from intracellular stores. The percentage increase in PIP metabolism produced by a concomitant administration of URP and UII at their maximum effective concentrations was not significantly different from the effects of both peptides tested individually, indicating that this pathway might be common to the both peptides.
In the present study, we observed that the binding and the stimulatory effect of URP on PIP metabolism were not significantly affected by pre-treatment with PTX, whereas half of the UII-induced PIP formation was blocked by PTX. This observation indicates that URP would not be promiscuously linked to the Gi/o PTX-sensitive binding site in astrocytes. In order to clarify a possible URP- and UII-induced UT coupling to Gi/o-proteins, kinetic BRET signal changes were measured in HEK-293 cells transiently co-transfected with Gαo–Rluc and Gγ2–YFP, and with either hUT or the control receptor PAR-1. Under these conditions, UII and also URP provoked a fast decrease in BRET exclusively in cells expressing hUT, resulting from UT–Gαo activation and Gαo–Gγ2 dissociation. These data indicate that, in a recombinant system, both UII and URP are able to activate a UT–Gi/o PTX-sensitive coupling system.
In astrocytes, PTX partially inhibited the effect of UII and, to a much lesser extent, of URP on PIP, whereas it failed to block the UII/URP-evoked Ca2+ response, suggesting the existence of a specific PTX-insensitive UT–Gq–PLC–InsP3–Ca2+ preferential coupling, and a PTX-sensitive UT–Gi/o–PIP pathway more specific to UII. These two coincident cascades probably converge to give rise to an increase in [3H]inositol incorporation into the PIP. Indeed, the stimulatory effect of UII and URP on PIP turnover can be ascribed to (i) the activation of PtdInsP2 hydrolysis into diacylglycerol and InsP3 by PLC, culminating in the release of Ca2+ from internal stores, and (ii) the production of PtdInsP, PtdInsP2, and PtdInsP3 upon the activation of phosphatidylinositol kinases in a Ca2+-independent manner. In particular, PIPs synthesized by PI3K (phosphoinositide 3-kinase) are known to play a critical role in cell survival by membrane recruitment and activation of Akt/PKB (protein kinase B) . Since PI3K has been shown recently to be involved in the UII-induced natriuretic peptide release from rat atria  and the MAPK (mitogen-activated protein kinase) pathway clearly implicated in several UII processes, including vascular and airway smooth muscle cell proliferation and migration, fibroblast proliferation and cardiomyocyte hypertrophy , we hypothesize that a UT–Gi/o–PI3K–PTX-sensitive pathway mainly activated by UII, and, to a lesser extent, by URP might thus be involved in Ca2+-independent astrocyte activities.
Pathological processes in the brain are usually accompanied by a significant astrocyte proliferation referred as to astrogliosis. In the present study, we have demonstrated that UII, but not URP, significantly stimulated rat cortical astrocyte proliferation. Numerous studies have demonstrated that vasoactive peptides or interleukins play important roles in astrogliosis . Thus we have demonstrated that, in the same set of experiments, PACAP and ET-1 tested at the same concentration as URP and UII, as well as IL-1β, also markedly increased cell growth. Moreover, graded concentrations of URP and UII were tested on either non-synchronized or synchronized cells, and the results demonstrate the existence of a UII-specific dose-dependent increase in cell viability. A similar activity for UII has already been observed on rat vascular smooth muscle cells , on human endothelial cells  and on UT-transfected CHO cells . Considering that URP has been shown to activate porcine sst2 , we proposed that the absence of a mitogenic role for URP would be linked to its ability to bind sst expressed by astrocytes, as sst2 [20,37]. It is conceivable that concomitant UT–Gq and sst2–Gi recruitment by URP may (i) also lead to [Ca2+]c increase, and/or (ii) antagonize some specific very-high-affinity UII/UT-mediating transduction pathways involved in cell proliferation.
In conclusion, the present study has demonstrated for the first time that URP and UII activate functional receptors in native rat cortical astrocytes, exhibiting similar and distinct modes of action. URP and UII bind a high-affinity site that exhibits the pharmacological characteristics of UT, mediating cell depolarization, endozepine release and PLC-dependent Ca2+ mobilization through intra- (InsP3 receptor) and extra- (T-type Ca2+ channels) cellular pools (Figure 9). Ca2+ entry though T-type channels appears to be essential for InsP3-induced Ca2+ increase, whereas Ca2+ release from intracellular stores induces Ca2+ release. In addition UII, and, to a lesser extent, URP are able to activate a higher-affinity site coupled to a PTX-sensitive G-protein that would mediate a hypothetical PI3K activation, relaying an important physiological astrocyte behaviour, i.e. cell proliferation (Figure 9). Finally, URP, but not UII, may bind sst expressed in rat astrocytes whose activation would mimic the UT–Ca2+ coupling and/or counteract the mitogenic role of UT (Figure 9). These data provide the first understanding of the involvement of the two vasoactive peptides UII and URP in nerve cell activity, stressing the specific role of UII in the (patho)physiology of astrocytes.
Proposed model depicting distinct and common transduction pathways activated by URP and UII on cultured rat astrocytes
bioluminescence resonance energy transfer
cytosolic Ca2+ concentration
central nervous system
days in vitro
Dulbecco's modified Eagle's medium
fetal bovine serum
glial fibrillary acidic protein
Hanks balanced salt solution
human embryonic kidney
pituitary adenylate cyclase-activating polypeptide
protease-activated receptor 1
Renilla reniformis luciferase
water-soluble tetrazolium salt 1
yellow fluorescent protein
Marie Jarry and Mickaël Diallo performed most of the experiments (binding, [Ca2+]c, PIP metabolism, cell viability and cell growth), Céline Lecointre and Laurent Prézeau carried out BRET experiments, Laurence Desrues generated receptor constructions for BRET and prepared radiolabelled peptides, Tursonjan Tokay measured endozepine release, David Chatenet and Jérôme Leprince synthesized UII and URP, Oriana Rossi initiated the binding studies, Hubert Vaudry and Marie-Christine Tonon contributed to the supervision, Hélène Castel performed electrophysiological experiments, co-supervised the project and contributed to the writing of the paper, Pierrick Gandolfo supervised the project and contributed to the writing of the paper.
We gratefully acknowledge Mr Sébastien Arthaud, Mrs Huguette Lemonnier and Mr Gérard Cauchois for skillful technical assistance. We thank Laetitia Comps-Agrar, Claire Vol, and Mohammed Akli Ayoub for their technical and scientific assistance, and the Pharmacology Screening-Interactome Platform facilities of the Institut Fédératif de Recherche 3 of Montpellier (France). We thank Professor Alain Fournier (INRS, Pointe-Claire, Canada), who generously provided PACAP.
This work was supported by Inserm (U982), the European Institute for Peptide Research (IFRMP 23), the Lille–Amiens–Rouen–Caen Neuroscience Network and the Conseil Régional de Haute-Normandie. M.D. was a recipient of a fellowship from the Lille–Amiens–Rouen–Caen Neuroscience Network and the Conseil Régional de Haute-Normandie. H.V. was recipient of the De Bétancourt-Perronet award for scientific co-operation between France and Spain.
These authors contributed equally to this work.