Short course of grass allergen peptides immunotherapy over 3 weeks reduces seasonal symptoms in allergic rhinoconjunctivitis with/without asthma: A randomized, multicenter, double-blind, placebo-controlled trial.

BACKGROUND: Immunotherapy with peptide hydrolysates from Lolium perenne (LPP) is an alternative treatment for seasonal allergic rhinitis with or without asthma. The aim of this study was to assess the clinical efficacy and safety of a cumulative dose of 170 µg LPP administered subcutaneously over 3 weeks. METHODS: In a randomized, double-blind, placebo-controlled trial, 554 adults with grass pollen rhinoconjunctivitis were randomized (1:2 ratio) to receive 8 subcutaneous injections of placebo or 170 µg LPP administered in increasing doses in 4 visits over 3 weeks. The primary outcome was the combined symptom and medication score (CSMS) measured over the peak pollen season. Reactivity to conjunctival provocation test (CPT) and quality of life (QOL) was assessed as secondary endpoints. RESULTS: The mean reduction in CSMS in the LPP vs placebo group was -15.5% (P = .041) during the peak period and -17.9% (P = .029) over the entire pollen season. LPP-treated group had a reduced reactivity to CPT (P < .001) and, during the pollen season, a lower rhinoconjunctivitis QOL global score (P = .005) compared with placebo group. Mostly mild and WAO grade 1 early systemic reaction (ESR) were observed ≤30 minutes in 10.5% of LPP-treated patients, whereas 3 patients with a medical history of asthma (<1%) experienced a serious ESR that resolved with rescue medication. CONCLUSION: Lolium perenne pollen peptides administered over 3 weeks before the grass pollen season significantly reduced seasonal symptoms and was generally safe and well-tolerated.

A randomized, double-blind, placebo-controlled, dose-finding trial with Lolium perenne peptide immunotherapy.

BACKGROUND: A novel subcutaneous allergen immunotherapy formulation (gpASIT+™) containing Lolium perenne peptides (LPP) and having a short up-dosing phase has been developed to treat grass pollen-induced seasonal allergic rhinoconjunctivitis. We investigated peptide immunotherapy containing the hydrolysate from perennial ryegrass allergens for the optimum dose in terms of clinical efficacy, immunogenicity and safety. METHODS: This prospective, double-blind, placebo-controlled, phase IIb, parallel, four-arm, dose-finding study randomized 198 grass pollen-allergic adults to receive placebo or cumulative doses of 70, 170 or 370 µg LPP. All patients received weekly subcutaneous injections, with the active treatment groups reaching assigned doses within 2, 3 and 4 weeks, respectively. Efficacy was assessed by comparing conjunctival provocation test (CPT) reactions at baseline, after 4 weeks and after completion. Grass pollen-specific immunoglobulins were analysed before and after treatment. RESULTS: Conjunctival provocation test (CPT) response thresholds improved from baseline to V7 by at least one concentration step in 51.2% (170 µg; P = .023), 46.3% (370 µg), and 38.6% (70 µg) of patients receiving LPP vs 25.6% of patients receiving placebo (modified per-protocol set). Also, 39% of patients in the 170-µg group became nonreactive to CPT vs 18% in the placebo group. Facilitated allergen-binding assays revealed a highly significant (P < .001) dose-dependent reduction in IgE allergen binding across all treatment groups (70 µg: 17.1%; 170 µg: 18.8%; 370 µg: 26.4%). Specific IgG4 levels increased to 1.6-fold (70 µg), 3.1-fold (170 µg) and 3.9-fold (370 µg) (mPP). CONCLUSION: Three-week immunotherapy with 170 µg LPP reduced CPT reactivity significantly and increased protective specific antibodies.

Lolium perenne peptide immunotherapy is well tolerated and elicits a protective B-cell response in seasonal allergic rhinitis patients.

BACKGROUND: Systemic allergic reactions are a risk for allergen immunotherapy that utilizes intact allergen preparations. We evaluated the safety, efficacy and immune mechanisms of short-course treatment with adjuvant-free Lolium perenne peptides (LPP) following a 6-week dose-escalation protocol. METHODS: In a prospective, dose-escalation study, 61 grass pollen-allergic patients received 2 subcutaneous injections of LPP once weekly for 6 weeks. Safety was assessed evaluating local reactions, systemic reactions and adverse events. The clinical effect of LPP was determined by reactivity to the conjunctival provocation test (CPT). Specific IgE, IgG4 and blocking antibodies were measured at baseline (V1), during (V6) and after treatment (V8). RESULTS: No fatality, serious adverse event or epinephrine use was reported. Mean wheal diameters after injections were <0.6 cm and mean redness diameters <2.5 cm, independent of dose. Transient and mostly mild adverse events were reported in 33 patients. Two patients experienced a grade I and 4 patients a grade II reaction (AWMF classification). At V8, 69.8% of patients became nonreactive to CPT. sIgG4 levels were higher at V6 (8.1-fold, P < .001) and V8 (12.2-fold, P < .001) than at V1. The sIgE:sIgG4 ratio decreased at V6 (-54.6%, P < .001) and V8 (-71.6%, P < .001) compared to V1. The absolute decrease in IgE-facilitated allergen binding was 18% (P < .001) at V6 and 25% (P < .001) at V8. CONCLUSION: Increasing doses of subcutaneous LPP appeared safe, substantially diminished reactivity to CPT and induced blocking antibodies as early as 4 weeks after treatment initiation. The benefit/risk balance of LPP immunotherapy remains to be further evaluated in large studies.

Influence of DNA condensation state on transfection efficiency in DNA/polymer complexes: an AFM and DLS comparative study.

Atomic force microscopy (AFM) is used to describe the formation process of polymer/DNA complexes. Two main objectives of this research are presented. The first one is to apply AFM as an effective tool to analyse DNA molecules and different polycation/DNA complexes in order to evaluate their degree of condensation (size and shape). The other one is to search for a relationship between the condensation state of DNA and its transfection efficiency. In this study, linear methacrylate based polymers and globular SuperFect polymers are used in order to induce DNA condensation. Ternary complexes, composed of methacrylate based polymers and polyethylene glycol (PEG)-based copolymers, are also investigated. AFM allows us to confirm good condensation conditions and relate them (or not) to transfection efficiencies. These AFM results (obtained after drying in air) are compared with measurements deduced from Dynamic Light Scattering (DLS) experiments performed in water. This comparison allowed us to identify the structural modifications resulting from deposition on the mica surface.

Involvement of multiple receptors in the actions of extracellular ATP: the example of vascular endothelial cells.

The role of ATP and ADP as intercellular mediators is now well established. The presence of the nucleotides in extracellular fluids can result from several mechanisms: cell lysis, selective permeabilization of the plasma membrane and exocytosis of secretory vesicles, such as platelet dense bodies. Extracellular adenine nucleotides are rapidly degraded by ectonucleotidases expressed inter alia on the surface of endothelial cells. They act on cells via the family of P2 receptors which encompasses more than 5 subtypes, some of which have been cloned recently. The P2T, P2U and P2Y receptors belong to the superfamily of receptors coupled to G proteins, whereas the P2X receptor is a cation channel and the P2Z receptor a non-selective pore. ATP and ADP stimulate the endothelial production of prostacyclin (PGI2) and nitric oxide (NO), two vasodilators and inhibitors of platelet aggregation, via an increase in cytosolic Ca2+. This action of adenine nucleotides is believed to limit the extent of intravascular platelet aggregation and to help localize thrombus formation to areas of endothelial damage. The endothelial response to nucleotides is mediated by at least two distinct subtypes of P2 receptors, P2Y and P2U, both coupled to phospholipase C.

Evidence that a form of ATP uncomplexed with divalent cations is the ligand of P2y and nucleotide/P2u receptors on aortic endothelial cells.

1. The response of bovine aortic endothelial cells to adenosine 5'-triphosphate (ATP) is mediated by both P2y and nucleotide/P2u receptors. In order to determine which form of the nucleotide is the true ligand of these receptors, we have investigated the effects of divalent cations on ATP-, uridine 5'-triphosphate (UTP)- and 2 methylthioadenosine 5'-triphosphate (2MeSATP)-induced inositol phosphate accumulation in these cells. 2. Omisson of Mg2+ from a calcium-free incubation buffer caused a shift to the left of the ATP concentration-action curve. 3. In the presence of EDTA (1 mM), the basal level of inositol trisphosphate (InsP3) was markedly increased and the absolute maximal response to ATP was decreased; however, the response to low concentrations of ATP was enhanced. 4. When the results were plotted in terms of calculated ATP4- concentrations, the concentration-response curves obtained in the presence of 1.25 mM Mg2+ lay closer to the respective curves obtained when Mg2+ was omitted from the medium or when Mg2+ was omitted and EDTA (1 mM) was added. The curves became almost superimposable when the baseline value was subtracted. 5. A similar shift to the left of the concentrations-action curves was also observed with both UTP and 2MeSATP. 6. Our data provide evidence that a form of ATP uncomplexed with divalent cation is the preferential agonist of both the nucleotide/P2u and the P2y receptors expressed on bovine aortic endothelial cells.

Palmitoyl-L-carnitine increases the release of prostacyclin from vascular endothelial cells.

Prostacyclin biosynthesis is dramatically increased in patients with acute myocardial infarction. As palmitoylcarnitine accumulates during myocardial ischemia, the action of this metabolite on the endothelial production of prostacyclin was studied. Palmitoyl-L-carnitine (10-100 microM) enhanced the release of prostacyclin and free arachidonic acid from bovine aortic endothelial cells. This action was mimicked by lysophosphatidylcholine, but by none of the following compounds: acetylcarnitine, carnitine, palmitic acid, sphingosine, dihydrosphingosine and N-stearoyl-dihydrosphingosine. In addition to mobilizing free arachidonate, palmitoylcarnitine induced the release of free choline and phosphorylcholine presumably via the activation of phospholipases C and D. Palmitoyl-L-carnitine had also a cytotoxic effect on the endothelial cells. These data suggest that the increased biosynthesis of prostacyclin in myocardial infarction might be partially explained by the accumulation and release of palmitoyl-L-carnitine.

Evidence that ATP, ADP and AMP are not ligands of the striatal adenosine A2A receptors.

It has been claimed recently that, in several cell types, ATP can induce a stimulation of cAMP production which is sensitive to methylxanthine inhibition and is not mediated by the ATP degradation product, adenosine. One explanation for these results would be direct activation of adenosine A2 receptors by ATP itself. We have therefore investigated whether adenine nucleotides are ligands of adenosine A2A receptors from bovine striatum. We show here that ATP, ADP, AMP and their phosphorothioates analogues (ATP gamma S, ADP beta S and AMP alpha S), at a 100 microM concentration, produced a 83-91% inhibition of the binding of [3H]CGS21680, an adenosine A2A receptor agonist, to striatum membranes. However, this action was inhibited by adenosine deaminase or by adenosine 5'-O-(alpha, beta-methylene)diphosphate (APCP), an inhibitor of 5'-nucleotidase-mediated AMP degradation. The effects of adenosine deaminase and APCP were dependent on their concentration. These results indicate that ATP, ADP and even AMP can exert an effect on the adenosine A2A receptors only through their breakdown into adenosine by ectonucleotidases.

Pharmacological characterization of the human P2Y4 receptor.

The P2Y4 receptor is a new member of the P2Y family which functionally behaves as a pyrimidinergic receptor. The pharmacological properties of the human P2Y4 receptor have been characterized following its stable expression in 1321N1 astrocytoma cells. UTP induced a biphasic accumulation of inositol trisphosphates, with an early peak at 30 s followed by a smaller but more sustained accumulation. ATP was a pure antagonist at early times and later behaved as a partial agonist. At 20 min, the rank order of potency of various nucleotides was the following: UTP > UDP = deoxy UTP > 5-bromo-UTP > ITP > ATP. Diadenosine polyphosphates also stimulated the production of inositol trisphosphates (after 20 min), more potently than ATP, but their maximal effect represented only 20-25% of that of UTP. Pyridoxal-phosphate-6-azophenyl-2',4'-disulphonic acid inhibited strongly the UTP response, whereas suramin was inactive and reactive blue 2 had an intermediate effect. Pertussis toxin inhibited the response to UTP at early times (62 +/- 5% inhibition at 30 s), but its effect was no longer observed at 5 or 20 min. It is speculated that the P2Y4 receptor can exist in two distinct activation states differing in terms of time-course, specificity for uridine nucleotides and G-protein coupling.

Metabolism of inositol phosphates in ATP-stimulated vascular endothelial cells.

The accumulation of InsP1, InsP2, InsP3 and InsP4 isomers was investigated in bovine aortic endothelial cells labelled with [3H]inositol and stimulated with ATP. The separation of these isomers was performed by ion-pairing reverse-phase h.p.l.c. on a mu Bondapack C18 column for the InsP3 and InsP4 isomers and by ion-exchange h.p.l.c. on a Partisil SAX column for the InsP1 and InsP2 isomers. In unstimulated endothelial cells, a large amount of material was co-eluted with InsP5 and InsP6, whereas amounts of InsP3 and InsP4 were small. The addition of ATP (100 microM) induced a striking (35-fold stimulation) and transient increase of Ins(1,4,5)P3 that was maximal around 15 s. This peak was followed by a more sustained accumulation of Ins(1,3,4,5)P4 and Ins(1,3,4)P3, but the amounts of these two metabolites accumulated in response to ATP were much smaller than that of Ins(1,4,5)P3. The increase in InsP2 isomers in response to ATP had similar characteristics: a rapid and transient accumulation of Ins(1,4)P2, followed by an increase of Ins(3,4)P2 and Ins(1,3)P2, which was more sustained but had a smaller magnitude. ATP also induced the accumulation of both Ins1P and Ins4P, but with different time courses: the level of Ins4P was maximal at 1 min (60 times the control value) and returned to baseline after 5 min, whereas the increase in Ins1P was undetectable at 1 min and reached a maximum after 5 min, which represented 240% of the basal level. These data indicate that Ins(1,4,5)P3, which is rapidly formed in aortic endothelial cells as a result of activation of P2Y receptors, is preferentially metabolized at early times (less than 1 min) by a 5-phosphatase, with the sequential formation of Ins(1,4)P2 and Ins4P. Afterwards, a small but sustained increase in the content of Ins(1,3,4)P3, Ins(1,3)P2, Ins(3,4)P2 and Ins1P was observed, reflecting the activation of the Ins(1,4,5)P3 3-kinase.

Dual role of GTP-binding proteins in the control of endothelial prostacyclin.

Pretreatment of bovine aortic endothelial cells with pertussis toxin inhibited partially the accumulation of inositol phosphates in response to ATP, whereas cholera toxin had no effect. Both pertussis and cholera toxins enhanced the stimulatory effect of ATP on prostacyclin release from the same cells. This action of cholera toxin was mimicked neither by an increase of cyclic AMP nor by the dissociated subunits of the toxin. Cholera and pertussis toxins, as well as aluminum fluoride, also potentiated the release of prostacyclin induced by ionophore A23187. These results suggest that a pertussis toxin-sensitive GTP-binding protein is involved in the coupling between P2-purinergic receptors and phospholipase C. In addition, another GTP-binding protein would play a crucial role at a further step in the control of PGI2 biosynthesis.

Heterogeneity of ATP receptors in aortic endothelial cells. Involvement of P2y and P2u receptors in inositol phosphate response.

Extracellular ATP plays an important role in the regulation of prostacyclin and nitric oxide release from vascular endothelial cells. These cellular responses to ATP are generally attributed to the stimulation of the P2y subtype of P2 purinergic receptors. However, it has recently been suggested that two types of ATP receptors might coexist on endothelial cells. To evaluate this hypothesis, we examined the effects of P2y receptor agonists 2-methylthioadenosine 5'-triphosphate (2MeSATP) and 2'- and 3'-O-(4-benzoylbenzoyl)adenosine 5'-triphosphate (BzATP) and of UTP on the accumulation of inositol phosphates in bovine aortic endothelial cells. BzATP, 2MeSATP, and UTP produced a smaller maximal effect than ATP. The effects of 2MeSATP and UTP were additive, whereas the effects of ATP and either UTP or 2MeSATP were not. Prior exposure to UTP reduced the subsequent response to UTP to 12% of the control response, whereas the response to 2MeSATP was decreased to 61%. Reciprocally, preincubation with 2MeSATP reduced the subsequent response to 2MeSATP to 23% of the control response, whereas the response to UTP was reduced to 73%. Pertussis toxin pretreatment decreased the response to both ATP and UTP (65% and 70% inhibition, respectively), whereas the response to 2MeSATP was not modified. Our data support the hypothesis that two classes of receptors recognizing ATP are expressed on bovine aortic endothelial cells.

Cloning and functional expression of a human uridine nucleotide receptor.

In order to isolate new subtypes of P2 purinoceptors, sets of degenerate oligonucleotide primers were synthesized on the basis of the best conserved segments in the published sequences of the chick brain P2Y/P2Y1 and murine neuroblastoma P2U/P2Y2 receptors. Their use in polymerase chain reaction (PCR) experiments on human genomic DNA amplified, among other things, a 712-base pair sequence, that was used as a probe to screen a human genomic DNA library. Several clones corresponding to a single locus were isolated, and the sequence analysis revealed an intronless 1095-base pair open reading frame. The deduced amino acid sequence is consistent with a G protein-coupled receptor and exhibits 51% identity with the human P2Y2 receptor and 35% with the chick P2Y1 receptor. A close comparison with the human P2Y2 sequence reveals the conservation of histidine 262, arginine 265, lysine 289, and arginine 292, which were reported to be involved in nucleotide binding (Erb, L., Garrad, R., Wang, Y., Quinn, T., Turner, J. T., and Weisman, G. A. (1995) J. Biol. Chem. 270, 4185-4188). Northern blot analysis detected a 1.8-kilobase messenger RNA in human placenta. The coding sequence was inserted in the pcDNA3 vector in order to transfect 1321N1 human astrocytoma cells. In cells stably expressing the receptor, UTP and UDP stimulated the formation of inositol phosphates with equivalent potency and maximal effect, ATP behaved as a partial agonist, and ADP was almost inactive. We have thus cloned a new member of the G protein-coupled P2 purinergic receptor family, which functionally behaves as a pyrimidinergic receptor.

Coexpression of P2Y and P2U receptors on aortic endothelial cells. Comparison of cell localization and signaling pathways.

Depending on the vascular bed considered, the actions of ATP on the endothelium are mediated by either P2Y or P2U receptors. The two types of receptors seem to coexist on bovine aortic endothelial cells, where they are both coupled to phospholipase C. In this study, we have investigated whether they are truly coexpressed on the same cells and whether their signaling pathways diverge beyond phospholipase C activation. Measurements of [Ca2+]i in single cells showed that almost all bovine aortic endothelial cells are responsive to both 2-methylthio-ATP (2MeSATP), an agonist of P2Y receptors, and UTP, an agonist of P2U receptors. UTP stimulated the release of prostacyclin from freshly isolated bovine aortic endothelial cells, even when they were exposed to cycloheximide at the time of their collection: this indicates that P2U receptors must already be expressed on endothelial cells in situ and do not appear during cell culture. The time course of inositol phosphate (InsP) accumulation and the relative proportion of Ins(1,4,5)P3, Ins(1,3,4,5)P4, and Ins(1,3,4)P3 were similar in cells stimulated by 2MeSATP or UTP. UTP and 2MeSATP both stimulated the hydrolysis of phosphatidylcholine by phospholipase D, as reflected by the release of [3H]choline from prelabeled cells. The responses to both agents were blocked after downregulation of protein kinase C, resulting from a prolonged exposure to phorbol 12-myristate 13-acetate: this blockade occurred at a step distal to phospholipase C activation. A single difference between the two pathways has been identified: the effect of 2MeSATP on InsP3 was significantly more inhibited after a short exposure to phorbol 12-myristate 13-acetate than that of UTP.(ABSTRACT TRUNCATED AT 250 WORDS)

Adenine nucleotides modulate phosphatidylcholine metabolism in aortic endothelial cells.

ATP and ADP, in concentrations ranging from 1-100 microM, increased the release of [3H]choline and [3H]phosphorylcholine (P-choline) from bovine aortic endothelial cells (BAEC) prelabelled with [3H]choline. This action was detectable within 5 minutes and was maintained for at least 40 minutes. ATP and ADP were equiactive, and their action was mimicked by their phosphorothioate analogs (ATP gamma S and ADP beta S) and adenosine 5'-(beta, gamma imido) triphosphate (APPNP), but not by AMP, adenosine, and adenosine 5'-(alpha, beta methylene)triphosphate (APCPP): these results are consistent with the involvement of P2Y receptors. ATP also induced an intracellular accumulation of [3H]choline: the intracellular level of [3H]choline was increased 30 seconds after ATP addition and remained elevated for a least 20 minutes. The action of ATP on the release of choline metabolites was reproduced by bradykinin (1 microM), the tumor promoter phorbol 12-myristate 13-acetate (PMA, 50 nM), and the calcium ionophore A23187 (0.5 microM). Down-regulation of protein kinase C, following a 24-hour exposure of endothelial cells to PMA, abolished the effects of PMA and ATP on the release of choline and P-choline, whereas the response to A23187 was maintained. These results suggest that in aortic endothelial cells, ATP produces a sustained activation of a phospholipase D hydrolyzing phosphatidylcholine. The resulting accumulation of phosphatidic acid might have an important role in the modulation of endothelial cell function by adenine nucleotides. Stimulation of phospholipase D appears to involve protein kinase C, activated following the release of diacylglycerol from phosphatidylinositol bisphosphate by a phospholipase C coupled to the P2Y receptors (Pirotton et al., 1987a).

P2-purinergic receptors in vascular endothelial cells: from concept to reality.

ATP exerts at least 2 actions on arterial endothelial cells: it stimulates the release of endothelium-derived relaxing factor, a still unidentified vasodilator, and of prostacyclin, a potent inhibitor of platelet aggregation. A study of agonist specificity indicates that these responses are mediated by P2-purinergic receptors. We have now demonstrated that in these cells, the P2-receptors are coupled to a phospholipase C hydrolysing phosphatidylinositol-bisphosphate and that this coupling involves a pertussis toxin-sensitive GTP-binding regulatory protein.

Involvement of inositol 1,4,5-trisphosphate and calcium in the action of adenine nucleotides on aortic endothelial cells.

ADP and ATP, in the 1-100 microM range of concentrations, increased the formation of inositol phosphates in bovine aortic endothelial cells. The accumulation of inositol trisphosphate in response to adenine nucleotides was rapid (maximum at 15 s) and transient. This material was identified as the biologically active isomer inositol 1,4,5-trisphosphate on the basis of its retention time by high-performance liquid chromatography on an anion-exchange resin. AMP and adenosine have no effect on inositol phosphates. The action of ATP and ADP was mimicked with an equal potency and activity by their phosphorothioate analogs, ATP gamma S and ADP beta S, and with a lower potency by adenosine 5'-(beta,gamma-imido)triphosphate, whereas adenosine 5'-(alpha,beta-methylene)triphosphate, was inactive. In the same range of concentrations, ADP and ATP induced an efflux of 45Ca2+ from prelabeled bovine aortic endothelial cells and increased the fluorescence emission by cells loaded with quin-2. Here, too, AMP and adenosine were completely inactive. The outflow of 45Ca2+ induced by ADP was partially maintained in a calcium-free medium. These data suggest that in aortic endothelial cells, P2-purinergic receptors, of the P2Y subtype, are coupled to the hydrolysis of phosphatidylinositol bisphosphate by a phospholipase C. It is likely that the release of prostacyclin and endothelium-derived relaxing factor in response to ADP and ATP is a consequence of this initial event.

Ecto-phosphorylation on aortic endothelial cells. Exquisite sensitivity to staurosporine.

One- and two-dimensional gel electrophoresis of proteins from bovine aortic endothelial cells (BAEC) incubated with [gamma-32P]ATP revealed the preferential labelling of a cell-associated 21 kDa substrate. The labelling of this band was detectable within 30 s, increased up to 30 min and was stable for at least 3 h following the wash-out of the ATP. This protein was also labelled after incubation of the cells with [gamma-35S]ATP. Incorporation of radioactivity into the 21 kDa band did not occur if the endothelial cells were treated with low concentrations of trypsin (0.01%) before or after the labelling period. The pattern of BAEC protein phosphorylation by [gamma-32P]ATP was completely different from that of the fetal calf serum used for the cell culture. The presence of serum during the incubation of BAEC with [gamma-32P]ATP did not modify qualitatively the labelling pattern and, in particular, did not enhance the phosphorylation of the 21 kDa substrate; this suggests that neither the kinase nor the 21 kDa substrate are adsorbed serum proteins. Staurosporine, a protein kinase inhibitor with low specificity, decreased the labelling of the 21 kDa protein with an IC50 of 2 nM. In contrast, at 100 nM, staurosporine did not decrease the accumulation of inositol phosphates induced by ATP via the activation of P2y receptors. These data indicate the presence of aortic endothelial cells of an ecto-kinase which uses extracellular ATP to produce the selective and long-lived phosphorylation of a 21 kDa endothelial substrate. Ecto-phosphorylation of this protein might play a role in the modulation of endothelial cell functions by ATP, in addition to the P2y receptors [Boeynaems & Pearson (1990) Trends Pharmacol. Sci. 11, 34-37]. The exquisite sensitivity of ecto-phosphorylation to inhibition by staurosporine and its specific inhibition by some isoquinolinesulphonamide compounds provide potential pharmacological tools to investigate this hypothesis.