Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Allergic and inflammatory diseases are complex, multifactorial processes that involve formation and/or release of many different mediators, including histamine and PAF. Histamine is released from mast cells and basophils by antigenic stimulation causing smooth muscle contraction, increased vascular permeability and mucus formation (White et al., 1987). Like histamine, PAF is known to provoke bronchoconstriction and increased vasopermeability, and it may also be responsible for bronchial hyperreactivity, a common feature of asthma (Pretolani and Vargaftig, 1993). PAF is also a very potent chemotactic stimulus for inflammatory cells, such as eosinophils (Wardlaw et al., 1986) and polymorphonuclear neutrophils (Pretolani et al., 1989) and is believed to be involved in inducing shock states (Sánchez-Crespo, 1993) and mediating inflammatory diseases (Koltai et al., 1994). Moreover, PAF and histamine are known to complement each other in vivo; histamine is a mediator of early response, being released from preformed reservoirs in mast cells, whereas PAF is mainly synthesized de novo (Sciberras et al., 1988; Snyder, 1994). Furthermore, each mediator is able to promote the release of the other in some tissues and cells (Presscott et al., 1987). Thus it seems reasonable to infer that the blockade of both PAF and histamine receptors may be of greater clinical effectiveness than the blockade of only one of them, and this is the premise behind the search for new chemical entities that possess dual activity. With the aim of obtaining a compound that offers such features, we have studied the pharmacological profile of a series of benzocycloheptapyridine derivatives (Carceller et al., 1994). From this work we identified rupatadine (fig. 1) as a novel, orally active and potent antagonist of histamine and PAF-induced responses in several animal species. The histamine and PAF antagonist activity of rupatadine in vitro and in vivo is here compared with that of several reference compounds: WEB-2086, a potent, orally active and selective PAF antagonist (Casals-Stenzel et al., 1987); loratadine, a second-generation antihistamine, and SCH-37370, which is the drug prototype with both histamine and PAF antagonist properties (Billah et al., 1990).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Structure of rupatadine, fumarate.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. The following drugs and chemicals were used in this study: [³H]-pyrilamine (23 Ci/mmol) and [³H]WEB-2086 (13.5 Ci/mmol) (New England Nuclear, Boston, MA), C₁₈-PAF, rupatadine (UR-12592:8-chloro-6,11-dihydro-11-[1-[(5-methyl-3-pyridinyl)methyl]-4-piperidinylidene]-5H-benzo[5,6]cyclohepta[1,2-b]pyridine), SCH-37370 (1-acetyl-4-(8-chloro-5,6-dihydro-11H-benzo[5,6]cyclohepta[1,2-b]pyridin-11-ylidene)piperidine), WEB-2086 (3-[-4-(2-chlorophenyl)-9-methyl-6H-thieno[3,2-f][1,2,4]triazolo[4,3-a][1,4]diazepin-2-yl]-1-(4-morpholinyl)-1-propanone) and loratadine (Chemistry Laboratories of J. Uriach & Cía, Barcelona, Spain), terfenadine, ketotifen, clemastine, diphenhydramine, pyrilamine, clorpheniramine, promethazine, histamine, ACh, serotonin, LTD₄, propanolol, BSA and lipopolysaccharide from E. coli 0127:B8 (Sigma Chemical Co., St. Louis, MO), arachidonic acid (Fluka, Buchs, Switzerland), ADP (Stago, Asnières, France), lipopolysaccharide from E. coli 0111:B4 (Difco, Detroit, MI). BN-52021 was a generous gift from laboratories Henri Beaufour (Le Plessis Robinson, France). All other chemicals used were of reagent grade or of the purest commercially available grade. Most of the studies were carried out using the fumarate salt of rupatadine. Throughout this manuscript, however, the test compound is referred to as rupatadine.

[³H]-WEB-2086 binding to PAF receptors in rabbit platelet membranes. Whole blood of male New Zealand rabbits (9 volumes) was drawn by cardiac puncture into 1 volume of 3.8% (w/v) sodium citrate and centrifuged at 150 × g for 15 min at 4°C. The PRP was carefully removed and centrifuged for 15 min at 1000 × g at 4°C. The platelet pellet was then washed three times in buffer A (10 mM Tris-HCl pH 7.4, 150 mM NaCl, 5 mM MgCl₂ · 6H₂O, 2 mM EDTA). Then the pellet was resuspended in buffer A, quickly frozen in liquid nitrogen and slowly thawed at room temperature. This freeze/thaw cycle was repeated three times. The lysed platelets were further centrifuged at 100,000 × g for 30 min at 4°C. The platelet membrane homogenate was resuspended in buffer B [10 mM HEPES pH 7.4, 145 mM NaCl, 5 mM KCl, 1 mM MgCl₂ · 6H₂O, 0.6 mM NaH₂PO₄, 0.4 mM K₂HPO₄, 6 mM glucose, 0.1% (w/v) BSA], stored at 80°C and used within 2 weeks. Protein content in membrane suspensions was determined by the Bradford method (1976) using a commercial protein assay reagent (Bio-Rad, Hercules, CA) and BSA as the standard.

11

5

[³H]-Pyrilamine binding to histamine (H₁) receptors in guinea pig cerebellum membranes. Cerebellum was isolated from adult male Dunkin-Hartley guinea pig and homogenized in 50 mM phosphate buffer, pH 7.5 (1:20, w/v). The homogenate was centrifuged at 270 × g for 2 min at 4°C to remove cell nuclei and debris. The supernatant was centrifuged at 60,000 × g for 10 min at 4°C, and the pellet was washed again. The resulting pellet was resuspended in phosphate buffer and stored at 80°C. Protein content in membrane suspensions was determined by the Bradford method, using Bio-Rad protein assay reagent and BSA as the standard.

Platelet aggregation assay. Aggregation studies were performed by previously described methods (Born, 1962). Blood was collected in 3.8% sodium citrate (1 volume per 9 volumes of blood) by cardiac puncture from male New Zealand rabbits (2-2.5 kg b.wt.) or from healthy volunteers. PRP was prepared by centrifuging the blood at 250 × g for 10 min at 4°C. In the human study, the HPRP was diluted with platelet-poor plasma obtained by further centrifuging at 3000 × g for 10 min. The platelet number was adjusted to 3.5 × 10⁵ cells/mm³. Rabbit PRP was further processed to obtain washed platelets, as described by Lalau Keraly et al. (1984). Briefly, PRP was mixed with three volumes of modified calcium-free Tyrode's solution containing 137 mM NaCl, 2.7 mM KCl, 11.9 mM NaHCO₃, 1 mM MgCl₂, 5.5 mM glucose, 0.23 mM EGTA and 2.5 g/l of gelatin, pH 6.5. A platelet pellet was obtained by centrifuging at 900 × g for 10 min at 4°C. The pellet was washed twice with calcium-free Tyrode's, pH 6.5. The last pellet was resuspended in a Tyrode's solution similar to the former but without EGTA, containing 53.3 mM CaCl₂, pH 7.4 (3 × 10⁵ platelets/mm³, final concentration). Platelet aggregation was induced by C₁₈-PAF (4 × 10⁶ M, HPRP; 5 × 10¹⁰ M, WRP) and measured by using a dual-channel aggregometer Chrono-log 560 (Havertown, PA). Platelet aggregation in the absence and in the presence (5-min incubation) of the test compounds was recorded. Activity of the inhibitors was expressed as the IC₅₀ values. To assess selectivity, rupatadine was tested against other aggregating agents, including arachidonic acid (1 mM) and ADP (5 µM), in WRP. Dose-response curves for PAF-induced aggregation in WRP were obtained in the absence of rupatadine and in its presence at various concentrations (3 × 10⁷-3 × 10⁵ M). Data were analyzed using the method of Arunlakshana and Schild (1959).

Histamine-induced contraction in guinea pig ileum. We used male Dunkin-Hartley guinea pigs (b.wt. 300-350 g), fasted overnight. Animals were stunned, the abdomen was opened and ileum sections 4 cm long were cut off. The sections were placed in a Petri dish containing Tyrode's solution at 37°C and continuously bubbled with carbogen. The ileum fragments were washed with Tyrode's solution and then were transferred to an organ bath. Ileum contraction was measured with an isometric transducer. The initial load was 1 g. After a stabilization period of 20 min in which the tissue was immersed in Tyrode's solution at 37°C continuously bubbled with carbogen, noncumulative stimuli with submaximal doses of histamine (5 × 10⁷ M) were given. The contraction in the absence and in the presence (5-min incubation) of the test compounds was recorded. The activities of the antagonists were expressed as IC₅₀ values. In another set of experiments, dose-response curves for histamine were obtained in the absence and in the presence of rupatadine at various concentrations (10⁹-3 × 10⁸ M). Data was analyzed using the method of Arunlakshana and Schild (1959). In selectivity studies, guinea pig ileum was also contracted using ACh (10⁷ M), serotonin (3 × 10⁶ M) and LTD₄ (10⁸ M).

PAF- and histamine-induced hypotension in normotensive rats. Male Sprague-Dawley rats weighing 180 to 220 g were anesthetized with sodium pentobarbital (50 mg/kg i.p.). Blood pressure was recorded from the left carotid artery using a Statham pressure transducer coupled to a Beckman R611 recorder. Right and left femoral veins were catheterized to inject the test compound and PAF (0.5 µg/kg) or histamine (25 µg/kg). Test compounds were administered by i.v. injection (1 ml/kg, dissolved in saline) 3 min before PAF or histamine injection. Blood pressure was monitored and percentage inhibition of PAF- or histamine-induced hypotension with respect to controls was calculated. The results were expressed as ID₅₀ values.

PAF- and histamine-induced bronchospasm in guinea pigs. Male Dunkin-Hartley guinea pigs weighing 300 to 350 g were anesthetized with urethane (1.5 g/kg i.p.). After suppression of spontaneous breathing with gallamine (5 mg/kg i.v.), the animal was connected to a Harvard ventilator (56 strokes/min). Bronchoconstriction was recorded using the method of Konzett and Rössler (1940). Histamine-induced bronchoconstriction was achieved at doses of 10 to 20 µg/kg i.v. For studies involving PAF, 3 mg/kg i.p. propanolol was given 20 min before the administration of 100 ng/kg i.v. C₁₈-PAF. When repetitive PAF or histamine responses were achieved, the test compound was administered i.v. 5 min before a new PAF or histamine injection. Percentage inhibition with respect to the initial response was calculated, and the results were expressed as ID₅₀ values.

PAF-induced mortality in mice. Groups of 10 fasted male Swiss mice weighing 22 to 26 g were used. A dose of 100 µg/kg of C₁₈-PAF plus 1 mg/kg of propranolol was administered through a lateral tail vein 5 min after i.v. administration of the test compounds (10 ml/kg) or saline (control group). Mortality was recorded 2 h after PAF injection. Following this protocol, we obtained a consistent mortality of 70% to 100% in the control group. Percentage inhibition of mortality attributable to treatment in comparison with control group mortality was calculated. The results were expressed as ID₅₀ values.

Endotoxin-induced mortality in mice and rats. Groups of 10 fasted male Swiss mice weighing 22 to 26 g or male Sprague-Dawley rats weighing 125 to 150 g were used. We administered 20 mg/kg of endotoxin from E. coli 0111:B4 (mice) or 5 mg/kg of endotoxin from E. coli 0127:B8 (rats) through a lateral tail vein 5 min or 30 min after i.v. or p.o. administration, respectively, of the test compound. Vehicle (saline or 1% Tween 80 in distilled water) was administered to control groups. Mortality was recorded 7 days after endotoxin injection. Following this protocol, we obtained a consistent mortality of 80% to 100% in the control group. Percentage inhibition of mortality attributable to treatment in comparison with control group mortality was calculated. The results were expressed as ID₅₀ values.

PAF- and histamine-induced cutaneous wheal reaction in dogs. Male beagle dogs weighing 10 to 14 kg were used. Animals were fasted 12 h before the experiment. A 2% w/v Evans blue dye solution in saline was administered i.v. (0.4 ml/kg) 30 min before intradermal injection of 10-µl solutions of PAF (0.15 mg/ml in 0.25% w/v BSA in saline) and histamine (0.25 mg/ml in saline) in previously shaved lateral thoracic skin (base line). Test compounds were administered p.o. (time 0). PAF and histamine were subsequently injected at different times (1, 2, 4, 6, 8, 12, 34 and 48 h). Wheal areas were measured 10 or 25 min after histamine and PAF injections, respectively, by using a computerized image analyzer system (MICROM). Wheal area for each time was compared to the base line, and percentage inhibition was calculated.

Spontaneous locomotor activity in mice. Male Swiss mice weighing 22 to 26 g were used in groups of three. Thirty minutes after a p.o. dose of test compound (or vehicle in control groups), the animals were placed in plastic cages (22 × 22 × 14.5 cm) located in a soundproof room. Five minutes later, movement was measured for 25 min by means of a Panlab (Barcelona, Spain) actimeter. Not less than four groups of three animals for each compound were used. Activity counts for treated groups were averaged and expressed in percentages of mean control counts.

Barbiturate-induced narcosis in mice. Male Swiss mice weighing 22 to 26 g were used. Sixty minutes after a p.o. dose of test compound (or vehicle in control groups), each animal was given an s.c. injection of sodium pentobarbital (35 mg/kg). The time from the loss of the righting reflex until its recovery was monitored for each animal. Results were given as the mean sleeping time and its S.E.M. for each treatment group.

Statistical analysis. Analysis of the pharmacological data (i.e., ID₅₀, IC₅₀ and pA₂ values and their 95% confidence limits) were performed using a standard pharmacology program (Tallarida and Murray, 1981). The IC₅₀ and ID₅₀ values were obtained by linear regression from an experimental curve with no fewer than four data points, each point being the mean percentage inhibition at a given concentration/dose obtained from two or more independent experiments. Statistical comparisons were made by ordinary or repeated-measure analysis of variance followed by Bonferroni's test using the INSTAT program. Binding data were analyzed using a weighted, nonlinear fitting program (LIGAND).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Inhibition of [³H]-WEB-2086 binding to PAF receptors in rabbit platelet membranes. Rupatadine (fig. 1) displaced [³H]WEB-2086 from its binding sites in rabbit platelet membranes, with a K_i value of 550 nM. Under the same conditions, WEB-2086, SCH-37370 and BN-52021 showed K_i values of 13, 22 and 221 nM, respectively (table 1).

Inhibition of [³H]-pyrilamine binding to histamine (H₁) receptors in guinea pig cerebellum membranes. Rupatadine displaced [³H]-pyrilamine from its binding site in guinea pig cerebellum membranes, with a K_i value of 102 nM, which was similar to the results obtained with terfenadine, loratadine and SCH-37370 and was at least two orders of magnitude higher than that obtained with pyrilamine (table 2).

Inhibition of PAF-induced aggregation of WRP and in HPRP. Rupatadine showed an in vitro PAF antagonist activity in the submicromolar range (IC₅₀ = 0.20 and 0.68 µM in WRP and in HPRP, respectively) (table 3). Selective PAF antagonists such as BN-52021 and WEB-2086 were more potent, by about one order of magnitude, whereas the antihistamine compounds loratadine or pyrilamine showed marginal or no activity in these tests. Schild analysis gave a pA₂ value of 6.68 ± 0.08 in WRP (fig. 2). The slope was 0.59 ± 0.03. A slope less than unity (absolute value) is commonly found in Schild plots for PAF antagonists (Floch and Cavero, 1990) and can be attributed to the internalization of PAF in washed platelets (Homma et al., 1987). Rupatadine is not a nonspecific platelet aggregation inhibitor; it did not inhibit platelet aggregation induced by ADP or arachidonic acid at concentrations up to 100 µM.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Dose-response curves of PAF-induced aggregation in washed rabbit platelets were obtained in the absence () or presence of 3 × 107 (), 106 (), 3 × 106 (), 105 () and 3 × 105 () M rupatadine. Each point is the mean percentage of response (n = 3) at a given concentration of PAF with respect to the maximum aggregation obtained in the control curve (). The Schild plot of the data is given in the insert. pA2 = 6.68 ± 0.08; slope = 0.59 ± 0.03.

Inhibition of histamine-induced contraction in guinea pig ileum. Rupatadine showed higher activity (IC₅₀ = 3.8 nM) than the nonsedative antihistamines terfenadine, loratadine and cetirizine (IC₅₀ values were 362, 286 and 90 nM, respectively) (table 4). Rupatadine's activity was comparable to that of the most potent antihistamines, pyrilamine being the only test compound with potency that was slightly higher than rupatadine's. Schild plot analysis showed a pA₂ value of 9.29 ± 0.06 (fig. 3). The slope was 1.15 ± 0.06, which indicates that rupatadine behaves as a competitive antagonist, at least in the range of concentrations tested. On the other hand, rupatadine showed practically no activity against ACh-, serotonin- and LTD₄-induced guinea pig ileum contraction, with IC₅₀ values >100, >100 and 41 µM, respectively.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Dose-response curves for histamine-induced contraction of guinea pig ileum were obtained in the absence of () or presence of 109 (), 3 × 108 (), 108 () and 3 × 108 () M rupatadine. Each point is the mean percentage of response (n = 4-8) at a given concentration of histamine with respect to the maximum contraction obtained in the control curve (). The Schild plot of the data is given in the insert. pA2 = 9.29 ± 0.06; slope = 1.15 ± 0.06.

Inhibition of PAF- and histamine-induced hypotension in normotensive rats. Rupatadine showed inhibitory effects in both PAF- and histamine-induced hypotension tests (ID₅₀ values were 0.44 and 1.4 mg/kg i.v., respectively; see table 5). This behavior was shared by the dual inhibitor SCH-37370, although this compound was somewhat less active than rupatadine as a PAF antagonist (ID₅₀ = 2.0 and 1.9 mg/kg i.v. against PAF- and histamine-induced hypotension, respectively). Other tested compounds showed preferential inhibition of one of the mediators: PAF antagonist WEB-2086 blocked only responses elicited by PAF (ID₅₀ = 0.17 mg/kg i.v.), whereas loratadine showed a moderate antihistamine activity (ID₅₀ = 4.7 mg/kg i.v.) but no PAF antagonism.

Inhibition of PAF- and histamine-induced bronchospasm in guinea pigs. Rupatadine inhibited bronchospasm induced by PAF and histamine (ID₅₀ = 0.0096 and 0.11 mg/kg i.v., respectively). SCH-37370 also showed dual activity, although its PAF antagonist activity is about one order of magnitude lower than that of rupatadine (table 5). WEB-2086 was active only against PAF (ID₅₀ = 0.0042 mg/kg i.v.), whereas loratadine showed no PAF antagonist effects (ID₅₀ >0.3 mg/kg i.v.) and an antihistamine activity (ID₅₀ = 1.6 mg/kg i.v.) lower than that of either rupatadine or SCH-37370.

Inhibition of PAF-induced mortality in mice. Rupatadine protected against mortality induced by PAF (ID₅₀ = 0.31 and 3.0 mg/kg i.v. and p.o., respectively), with a potency that was about one-third that of WEB-2086 (ID₅₀ = 0.084 and 0.97 mg/kg i.v. and p.o., respectively), whereas SCH-37370 was less active (ID₅₀ = 1.1 and 31 mg/kg i.v. and p.o.) than either rupatadine or WEB-2086 (table 6). Loratadine displayed no activity in this test (ID₅₀ >3 and >30 mg/kg i.v. and p.o., respectively).

Inhibition of endotoxin-induced mortality in mice and rats. Rupatadine and WEB-2086 behaved similarly in both mice and rats (ID₅₀ values for rupatadine and WEB-2086 were 1.6 and 1.9 mg/kg i.v., mice; 0.66 and 0.47 mg/kg i.v., rats) (table 6). Loratadine and SCH-37370 were not active (ID₅₀ >5 mg/kg i.v., rats; ID₅₀ >10 mg/kg i.v., mice).

Inhibition of PAF- and histamine-induced cutaneous wheal reaction in dogs. Rupatadine dose-dependently (0.3-10 mg/kg p.o.) inhibited both histamine- and PAF-induced wheal response (fig. 4, A and B), although a higher potency against histamine-induced reactions was obtained. Peak activity was observed 4 h after product administration, and the effect lasted for 12 to 48 h, depending on the dose. When rupatadine was compared with reference substances (all at 1 mg/kg p.o.), its H₁ antihistamine potency (E_max about 75% inhibition at 4 h after administration) and duration of activity (42% and 35% inhibition at 24 h, respectively) were similar to those of loratadine (fig. 5A). Rupatadine compared favorably with WEB-2086, showing similar PAF antagonist potency (E_max about 55% inhibition at 4 h) but longer duration of activity (fig. 5B). Loratadine displayed no PAF antagonist activity, and WEB-2086 showed no H₁ antihistamine activity at the dose tested.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Mean percentage of inhibition (± S.E.) of wheal areas induced by 2.5 µg histamine (A) or 1.5 µg PAF (B) at different times after the p.o. administration (time 0) of 0.3 (), 1 (), and 10 () mg/kg of rupatadine. Wheal areas were calculated as described in "Materials and Methods."

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Mean percentage of inhibition (± S.E.) of wheal areas induced by 2.5 µg histamine (A) or 1.5 µg PAF (B) at different times after the p.o. administration (time 0) of 1 mg/kg of rupatadine (), loratadine (), WEB-2086 () and SCH-37370 (). Wheal areas were calculated as described in "Materials and Methods."

Effects of rupatadine on spontaneous locomotor activity in mice. At a dose of 100 mg/kg p.o., rupatadine and the antihistamines loratadine and terfenadine did not affect spontaneous motor activity in mice (table 7). Under the same conditions, diazepam showed 46% reduction of activity at a dose of 10 mg/kg.

Effects of rupatadine on barbiturate-induced narcosis in mice. Neither rupatadine nor terfenadine at a dose of 100 mg/kg prolonged sleeping time induced by sodium pentobarbital (table 7). At the same dose, loratadine and SCH-37370 produced significant potentiation of barbiturate effect, with a nearly 3-fold increase in sleeping time (320 vs. 113 min in the loratadine-treated and control groups, respectively). The potentiation of sodium pentobarbital-induced narcosis caused by loratadine and SCH-37370 disappeared at doses below 30 mg/kg (data not shown).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present work demonstrates that the novel compound rupatadine behaves as a potent, orally active and long-lasting histamine (H₁) and PAF receptor antagonist. This dual activity has been demonstrated in vitro and in a wide array of pharmacological models in vivo in several species, including mice, rats, guinea pigs, rabbits, dogs and humans.

Rupatadine displaces both [³H]WEB-2086 and [³H]-pyrilamine from their binding sites in rabbit platelet membranes and guinea pig cerebellum membranes, respectively. Thus the inhibition of PAF- and histamine-mediated effects produced by rupatadine in vitro and in vivo should be ascribed to its interaction with specific receptors and not to physiological antagonism. Several antihistamines (e.g., loratadine and cetirizine) have revealed some PAF antagonist properties, such as inhibition of PAF-induced eosinophil chemotaxis, but these effects cannot be attributed to specific interaction with PAF receptors (Townley and Okada, 1992; Eda et al., 1993). H₁ receptor binding constants of second-generation antihistamines do not predict their potency in other in vitro tests (Zhang et al., 1991), and this was also the case for rupatadine in our study. Thus rupatadine inhibits histamine-induced contractions of guinea pig ileum at concentrations in the nanomolar range. The discrepancies may be a result of very slow dissociation from receptors (Laduron et al., 1982; Ter Laak et al., 1993), time-dependent binding (Zhang et al., 1991) or high nonspecific binding (Laduron et al., 1982), all of which can bring about underestimation of binding constants. Another possibility is that antihistamines may interfere with signal transduction mechanisms beyond interaction with specific receptors. Loratadine, for example, is able to mobilize calcium from intracellular stores and also increases calcium influx (Berthon et al., 1994). Studies are to verify the validity of these hypotheses for rupatadine under way. Competitive antagonism is shown by Schild analysis of guinea pig ileum data, in which a pA₂ value of 9.29 similar to that obtained with the most potent first-generation antihistamines, such as pyrilamine, was obtained (reported pA₂ in guinea pig ileum is 9.4; Bowman and Rand, 1980). The activity shown by rupatadine in guinea pig ileum is much greater than that of the nonsedative second-generation antihistamines such as terfenadine and loratadine (the latter has a pA₂ value of 7.22, data obtained in our laboratory). Nevertheless, it must be kept in mind that both terfenadine and loratadine possess active metabolites, some of which are even more active than the parent compound (Simons and Simons, 1991). The presence of active metabolites and their slow kinetics of dissociation from their receptor account for the fact that the potency in vivo of several antihistamines is higher than could be expected on the basis of in vitro results.

Studies performed in guinea pig ileum suggest that rupatadine's inhibition of histamine-induced responses is selective, because no anticholinergic, antiserotoninergic or LTD₄ receptor antagonist activity was found. Rupatadine also showed selectivity for H₁ receptor-mediated histamine activity, being about two orders of magnitude less potent than cimetidine in inhibiting the increase of heart rate induced by histamine in isolated guinea pig atria through H₂ receptor mediation (data not shown).

PAF antagonist potency is commonly assayed by taking advantage of the ability of this mediator to aggregate platelets from several species (Cargill et al., 1983). Rupatadine shows a PAF antagonist activity in the submicromolar range (IC₅₀ = 0.20 and 0.68 µM in WRP and in HPRP, respectively). Rupatadine's in vitro activity lies between that of the pure antihistamines, which lack PAF antagonist properties in these tests, and that of the selective PAF receptor blockers, which are about one order of magnitude more potent than rupatadine. In the in vivo tests, however, the differences are clearly reduced, and rupatadine's effect is very much like that of WEB-2086. Moreover, rupatadine shows even greater inhibitory potency when hypotension (rats) or bronchoconstriction (guinea pigs) is induced by PAF (ID₅₀ = 0.44 and 0.0096 mg/kg i.v., respectively) than when these effects are elicited by histamine (ID₅₀ = 1.4 and 0.11 mg/kg i.v., respectively). Rupatadine's activity in vivo also clearly surpasses that of the standard dual antagonist SCH-37370. The reasons why rupatadine shows better PAF antagonist activity in vivo than expected from in vitro results remain unexplained by our data, although rupatadine's good pharmacokinetic profile could account for the results observed.

The i.v. injection of PAF may cause death in mice. This is probably due to a combination of effects, including increased vascular permeability, depression of cardiac and pulmonary functions, activation of neutrophils and severe hypotension (Terashita et al., 1987). Pretreatment with rupatadine confers protection against mortality (ID₅₀ = 0.31 and 3.0 mg/kg i.v. and p.o., respectively), with a potency near that of the selective PAF antagonist WEB-2086. Pure antihistamines, on the other hand, are not active, and the dual-activity compound SCH-37370 shows a slight effect. The PAF antagonist activity of rupatadine was further corroborated in a PAF-dependent model: endotoxin-induced shock in rodents. The i.v. administration of bacterial lipopolysaccharides produces a shock state characterized by systemic hypotension, bronchoconstriction and multiorgan failure, events that together lead to death (Parrillo, 1993). It has been demonstrated that PAF may be a key mediator in the development of shock (Lefer, 1989), and PAF antagonists are known to prevent some of the symptoms present in this model, including mortality (Herbert et al., 1991; DeJoy et al., 1993). Thus, given the remarkable effect displayed by rupatadine against PAF-induced mortality, the protection it affords against endotoxin is hardly surprising. This activity should be attributed to rupatadine's PAF antagonist properties, because pure antihistamines are inactive. Even though high levels of histamine have been detected in plasma in a late phase of endotoxic shock, this mediator does not seem to play a substantial role in promoting shock (Brackett et al., 1990).

Rupatadine dose-dependently inhibits the wheal response induced by histamine in dogs. Given p.o. at a dose of 1 mg/kg, rupatadine exhibits antihistamine activity that mirrors that of loratadine in both potency and duration, while surpassing the PAF antagonist activity of WEB-2086 by offering a similar maximum inhibition (about 55%) but a longer duration of effect. PAF's effects on plasma extravasation can be only partially blocked by rupatadine, however. Although at low doses (0.3-1 mg/kg), a dose-response relationship can be observed, the administration of higher doses (10 mg/kg) does not provide an increase in maximum inhibition. Limited distribution of the product into the peripheral compartment (in this case into the skin) could explain this lack of linearity. Another interesting observation is the similarity with which the histamine and PAF antagonist activities decrease over time. SCH-37370, on the other hand, is known to lose its PAF antagonist properties faster than its antihistamine activity both in animal models (Billah et al., 1990) and in humans (Billah et al., 1992).

In conclusion, rupatadine administered p.o. possesses a potent, long-lasting and balanced in vivo dual PAF and histamine antagonist activity. Although compounds with either more potent antihistaminic properties or higher PAF antagonist activity are available, rupatadine uniquely combines both activities at a high level of potency. Rupatadine's good pharmacological properties, as well as its absence of sedative effects in preliminary tests, suggest that this compound could be an efficacious alternative to the drugs currently used in the treatment of allergy and asthma. Further experiments using animal models of such pathologies followed by clinical trials are needed to test this hypothesis.