Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Recently developed nonpeptide angiotensin II (Ang II) AT₁ receptor antagonists comprise a new generation of antihypertensive agents that reduce increased peripheral vascular resistance without eliciting reflex changes in cardiac output and heart rate or interfering with kinin metabolism (Townsend and Ford, 1996). Losartan, the first of this novel class of orally active AT₁ receptor antagonists, blocks most known Ang II-mediated responses and is clinically effective in the management of hypertension and congestive heart failure (Timmermans et al., 1993; Goa and Wagstaff, 1996). Losartan reduces high blood pressure in hypertensive patients, as well as in animal models of hypertension, such as renal hypertensive rats, spontaneously hypertensive rats (SHR) and renin transgenic hypertensive rats (Wong et al., 1990; Moriguchi et al., 1994; Townsend and Ford, 1996). It has been noted that acute administration of losartan is more potent than the peptide AT receptor antagonist saralasin and angiotensin converting enzyme (ACE) inhibitors in lowering blood pressure in SHR (Ohlstein et al., 1992). Losartan reduces constrictor responses to thromboxane A₂ in blood vessels of SHR by stimulating production of nitric oxide (Maeso et al., 1997). Grove and Speth (1993) showed that binding of [³H]losartan has a greater density than the labeled Ang II in different tissues. These studies indicate that although the antihypertensive action of losartan is attributed mainly to its ability to antagonize the AT₁ receptor, additional mechanisms may be involved in the therapeutic effects of losartan.

Recent findings from our and others laboratories demonstrated that the AT₁ receptor antagonist losartan and its active metabolite EXP3174 interact with the thromboxane A₂/prostaglandin endoperoxide H₂ (TxA₂/PGH₂) receptor (Liu et al., 1992; Bertolino et al., 1994; Li et al., 1997, 1998; Gueraa-Cuesta et al., 1999). TxA₂ is a potent endogenous vasoconstrictor and mediator of platelet aggregation. Abnormal production of TxA₂ is linked to the pathophysiology of renal and Ang II-dependent hypertension, coronary artery spasm, and arterial thrombosis (Tada and Kuzuya, 1985; Dai et al., 1992; Lin et al., 1994). We reported that losartan and EXP3174 inhibit TxA₂ analog U46619-induced vasoconstriction and platelet aggregation in SHR (Li et al., 1998). These studies showed that the inhibiting effect of losartan on the TxA₂/PGH₂ receptor is not shared by the AT₁ receptor antagonist candesartan, indicating that this separate and specific cardiovascular action of losartan may be related to its chemical characteristics rather than specific for the class of AT₁ receptor antagonists.

Irbesartan (BMS186295, SR 47436) is one of the newly developed orally active nonpeptide AT₁ receptor antagonists (Cazaubon et al., 1993). Studies in animal models and human subjects of hypertension showed that irbesartan effectively reduced high blood pressure with a potency similar to losartan at equal doses (Pool et al., 1998). This novel compound has a similar structure to losartan (Fig. 1), but no study has been conducted to evaluate the interaction of irbesarten with the TxA₂/PGH₂ receptor. In the present study, we investigated whether irbesartan interacts with TxA₂/PGH₂ in isolated canine coronary arteries and human platelets. In addition, we also evaluated whether irbesartan competes for the specific binding of [³H]SQ29,548, a specific TxA₂/PGH₂ antagonist, in canine coronary arteries.

View larger version (9K): [in this window] [in a new window]   Fig. 1.   Comparison of the structures of irbesartan (left) and losartan (right).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Preparation of Coronary Artery Rings

Following approval by the Institutional Animal Care and Use Committee, 18 male adult dogs (15-25 kg b.wt.) were anesthetized with ketamine (20 mg/kg i.m.) and 2% halothane inhalation, and then the dogs were euthanized with sodium pentobarbital (50 mg/kg i.v.). The heart was harvested immediately and immersed on ice-cold modified Krebs buffer. The left anterior descending coronary artery was carefully dissected free of fat and adhering connective tissues and was cut into 3-mm-long rings. Vascular segments were suspended by two stainless steel wire triangles in organ chambers as previously described (Li et al., 1997). Krebs' solution (118.3 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25 mM NaHCO₃, 0.026 mM CaNa-EDTA, and 11 mM glucose) was aerated with 95% O₂ and 5% CO₂ at 37°C (pH 7.4). The rings were allowed to equilibrate for 60 min at 1 g initial resting tension, and then basic tension was increased individually in a step-by-step fashion until the optimal length-tension relationship was obtained by repeated exposure to 40 mM KCl. The optimal basic tension was 4 to 6 g in our preparations. In some rings, the endothelial cells of vascular rings were denuded by gentle mechanical rubbing with a stainless steel wire. Isometric tension of vascular rings was measured continuously with force-displacement transducer (FTO3; Grass Instrument Co., Quincy, MA) connected to a Grass polygraph. The functional integrity of endothelium of vascular rings was confirmed by the presence of acetylcholine (Ach)-induced relaxation in preconstricted rings with 10⁸ M U46619 (9,11-dideoxy-11, 9-epoxymethano-prostaglangdin F₂) (>90% relaxation at 10⁷ M Ach) and the absence of Ach-induced relaxation in vessels following mechanical denudation of the vascular endothelium.

Experimental Protocol

Vascular Reactivity. Control cumulative concentration-response curves for the TxA₂ analog U46619 (10¹⁰-3 × 10⁶ M) were generated after 1 h equilibration in intact quiescent rings. Irbesartan (10⁷-10⁵ M) was used to pretreat the coronary artery rings for 30 min and the concentration-response curves for U46619 were then repeated. To determine whether irbesartan interacts with other vasoconstrictors, concentration-response curves for prostaglandin F₂ (PGF₂) (10¹⁰-10⁵ M) and KCl (10-80 M) also were constructed in the absence and presence of irbesartan (10⁶ M) in isolated coronary vascular rings. In addition, to compare the potency and selectivity of irbesartan on the TxA₂ receptor in coronary arteries, the potent, selective TxA₂/PGH₂ receptor antagonist SQ29,548 (Ogletree et al., 1985) was used to pretreat the tissues for 30 min and then concentration-response curves for U46619 were determined.

6

5

6

w

4

9

4

In Vitro Autoradiography. Canine coronary artery was obtained from four dogs. Vessels were frozen and sectioned at 14 µm on a cryostat (Bright Instrument Company Ltd., Huntingdon, Cambs, England). Sections were mounted onto Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA) and stored at 80°C until use. Incubations were carried out in 50 mM Tris buffer, pH 7.2, containing 5 mM ethylene glycol bis(-aminoethyl)-N,N,N',N'-tetraacetic acid and 0.1% BSA (Shin et al., 1993). The autoradiography procedure was modified from Li et al. (1996). Sections were preincubated in buffer for 30 min at room temperature (22°C), and then incubated for 45 min at room temperature with 9 nM [³H]SQ29,548 (51 Ci/mmol). Specific binding in the presence of 1 or 10 µM unlabeled SQ29,548 added to the incubation mixture was ~60%. U46619 (10⁸-10⁵ M) or irbesartan (10⁷-10⁴ M) were added to the incubation buffer for competition studies. After incubation, slides were rinsed in ice-cold double distilled water and were dipped into Tris buffer at 4°C for 1 min with a final brief rinse in ice-cold double distilled water. All sections were dried rapidly under a stream of cool dry air and stored with desiccant at 4°C overnight. Slides were apposed to Biomax MR (Kodak, Rocherster, NY) with ¹⁴C-labeled standards in cassettes at room temperature for 7 months. The relative density of binding sites was estimated by densitometric analysis with a computerized imaging device (Imaging Research, Ontario, Canada). Measurements were taken from at least two and as many as three different tissue sections from each animal in the absence or presence of each competitor. The percentage of inhibition by competing ligands is presented as means ± S.E. and values were compared with a constant (0) to determine whether significant competition occurred. K_i estimates were calculated with PRISM (GraphPad Software, San Diego, CA).

Human Platelet Aggregation

Blood for platelet aggregation studies was collected by venous puncture from healthy human volunteers (n = 6) who were free of aspirin-like agents for at least 2 weeks. Platelet-rich plasma (PRP) was obtained as described by Hink et al. (1989). Whole venous blood was mixed with 3.8% trisodium citrate buffer (9:1 v/v) and then centrifuged at 200g for 15 min at room temperature to obtained PRP. Platelet-poor plasma (PPP) was prepared by centrifugation of blood at 4000 rpm for 5 min and was used for adjusting the platelet concentration of PRP to 250,000-300,000 cells/µl and calibrating the aggregometer at maximal light transmittance. Platelet aggregation of PRP was monitored at 37°C with an aggregator (Bio-data, Hatsboro, PA) connected to a computer for data recording. U46619 (0.1-10 µM) was used to elicit dose-dependent platelet aggregation as a control response. Irbesartan (1 and 10 µM) was used to pretreat platelets for 10 min and U46619-induced platelet aggregation was repeated. Fresh blood was used and aggregation studies were finished within 2 to 3 h.

Drugs and Chemicals

Irbesartan was a generous gift from Bristol-Myers Squibb pharmaceutical research institute (New Brunswick, NJ). Losartan was obtained from DuPont Merck Company (Wilmington, DE) and valsartan from Novartis Corp. (Summit, NJ). PD 123319 was generously supplied by Parke-Davis (Ann Arbor, MI) and CV-11974 by Takeda Chemical Industries, Ltd. (Osaka, Japan). N-nitro-L-arginine methyl ester and SQ29,548 were purchased from Research Biochemicals International (Natick, MA). [³H]SQ29,548 was purchased from DuPont NEN (Boston, MA). Other chemicals were obtained from Sigma Chemical Co. (St. Louis, MO). Indomethacin, CV-11974, irbesartan, and valsartan were dissolved in 0.2 N Na₂CO₃ solution and diluted with Krebs' buffer. U46619 was prepared as stock in ethanol and diluted with Krebs' buffer. The concentrations of drugs reported in the text are at final concentration in organ chambers.

Data and Statistical Analysis

The concentration of U46619 causing 50% of the maximal contraction (EC₅₀) and the concentration of irbesartan causing 50% of the maximal relaxation (IC₅₀) were calculated with a nonlinear regression sigmoid curve fitting program of PRISM. The apparent dissociation constant (K_b) was calculated with the equation K_b = [B] · ([A']/[A]  1)¹, where [B] is the concentration of the antagonist and [A] and [A'] are the EC₅₀ values obtained in each artery before and after the addition of each antagonist (Corriu et al., 1995). Data are expressed as means ± S.E. One-way ANOVA followed by Newman-Keuls multiple comparisons and Student's t test for paired observations was used for statistical evaluation. P < .05 was considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Irbesartan on U46619-Induced Contraction in Coronary Arteries. Thromboxane A₂ analog U46619 produced concentration-dependent vasoconstriction in canine coronary artery rings (Fig. 2A). Preincubation with irbesartan inhibited the dose-response curve of U46619 and shifted the concentration-response curve of U46619 to the right in a dose-dependent manner without a change in the maximum constriction. The EC₅₀ of U46619 (11.6 ± 1.5 nM; control) was shifted 6- and 35-fold by pretreatment with 1 and 10 µM irbesartan, respectively (P < .01 compared with control) (Table 1). The apparent dissociation constant K_b averaged 214.5 ± 46.7 and 235.6 ± 45.8 nM in the presence of 1 and 10 µM irbesartan, respectively. The potent, selective TxA₂/PGH₂ receptor antagonist SQ29,548 markedly shifted the concentration-response curves of U46619 to the right without a change of maximal contraction [Fig. 2B, EC₅₀: 11.6 ± 1.5 (control) versus 179.6 ± 24.6 and 620.3 ± 33.8 nM at 0.01 and 0.1 µM of SQ29,548, respectively]. Pretreatment with 100 µM irbesartan or 1 µM SQ29,548 for 30 min nearly abolished the contractile responses of U46619 at the concentrations tested (data not shown). The potency of antagonism of SQ29,548 on the U46619-induced vasoconstriction was 100-fold greater than irbesartan (each at 0.1 µM) [K_b: 2.2 ± 0.4 versus 222 ± 16 nM, SQ29,548 versus irbesartan, P < .01].

View larger version (25K): [in this window] [in a new window]   Fig. 2.   A, average dose-response contraction curve to U46619 during control conditions (n = 16 dogs) and after pretreatment with increasing concentrations of irbesartan (107-105 M) (n = 5-6 dogs). B, average dose-response curve to U46619 during control conditions (n = 16 dogs) and after pretreatment with increasing concentrations of SQ29,548 (108-107 M) (n = 4 - 6 dogs). Values are means ± S.E.

Selectivity of Nonpeptide Ang II Receptor Antagonists and ACE Inhibitor on TxA₂/PGH₂ Receptor. As we previously reported (Li et al., 1997), losartan at 1 µM significantly shifted the concentration-response curves of U46619 to the right (EC₅₀: 11.6 ± 1.5 versus 36.5 ± 2.8 nM, control versus losartan, P < .01). However, losartan was 2-fold less potent than irbesartan in the inhibition of U46619-induced vasoconstriction at equal molar concentration (K_b 480 ± 10.5 versus 214.5 ± 46.7 nM, losartan versus irbesartan, P < .01). In contrast, pretreatment with the AT₁ receptor antagonists CV11974 and valsartan did not affect the concentration-response curve of U46619 at 1 µM, [EC₅₀: 11.6 ± 1.5 (control) versus 12.7 ± 3.3 (CV11974) and 13.3 ± 2.5 (valsartan) nmol/l, P > .05 compared with control] (Fig. 3A and Table 1). Both CV11974 and valsartan at concentration of 10 µM had no effect on U46619-induced vasoconstriction (data not shown). In addition, incubation with the AT₂ antagonist PD123319 or the ACE inhibitor lisinopril each at 1 µM for 30 min had no effect on the concentration-response curves of U46619 (Fig. 3B and Table 1).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   A, average dose-response contraction curve to U46619 during control (; n = 16 dogs) and after preincubation with the Ang II AT1 receptor antagonist CV11974 (), losartan (), valsartan (), and irbesartan () (n = 5-6 dogs in treated groups; all antagonists at concentration of 1 µM). B, average dose-response curve to U46619 during control conditions (; n = 16 dogs) and after pretreatment with the AT2 antagonist PD123319 () and the ACE inhibitor lisinopril () at 106 M (n = 5-6 animals). Values are means ± S.E.

Specificity of Irbesartan for Vasoconstrictor-Induced Contraction in Coronary Arteries. Pretreatment with irbesartan (1 µM) did not affect the coronary vasoconstrictor response to the smooth muscle cell depolarizing agent KCl (Fig. 4A). However, irbesartan (1 µM) did block the TxA₂/PGH₂ receptor agonist PGF₂-induced dose-dependent vasoconstriction without a change of maximal contraction (Fig. 4B, EC₅₀: 0.8 ± 0.2 versus 5.1 ± 0.3 µM; maximal contraction: 4.1 ± 0.6 versus 3.9 ± 0.5 g, control versus irbesartan, n = 5). Phenylephrine and arginine vasopressin caused minimal contractile responses in canine coronary arteries, and irbesartan did not change these contractile effects (data not shown).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   (A) Pretreatment with irbesartan (1 µM) did not affect the coronary constrictor response to KCl (10-80 mM). [control () versus irbesartan treated ()]. Values are mean ± S.E., (n = 5 dogs). (B) PGF2 elicited concentration-dependent contraction of coronary arteries before (, control) and after irbesartan pretreatment (). Irbesartan at 1 µM inhibited the contraction of PGF2. Each group contained 4 to 6 animals.

Effects of Production of Vasoactive Prostaglandins and NO on Irbesartan's Attenuation of U46619-Induced Vasoconstriction. Copreincubation of 1 µM irbesartan and 10 µM of the cyclooxygenase inhibitor indomethacin with rings for 30 min shifted the concentration-response curve of U46619 significantly to the right compared with pretreatment with 10 µM indomethacin alone (Fig. 5A, EC₅₀: 11.5 ± 2.5 versus 57.6 ± 3.6 nM, control versus irbesartan). This occurred without a change of maximal contraction [maximal contraction (6.7 ± 0.7 versus 7.6± 1.4 g, control versus irbesartan). There was no difference in the attenuated response in irbesartan-treated rings with and without indomethacin copretreatment. Similarly, copretreatment of 1 µM irbesartan and the NO synthase inhibitor L-NAME (100 µM) with rings did not change irbesartan's attenuation of U46619-induced vasoconstriction (Fig. 5B, EC₅₀: 13.5 ± 2.5 versus 71.6 ± 4.3 nM; maximal contraction: 6.2 ± 1.3 versus 6.3 ± 0.6 g, control versus irbesartan). Comparison of the irbesarten-treated rings with and without L-NAME revealed that there was no effect of blockade of NO release on irbesartan's attenuated U46619-elicited response (EC₅₀: 71.6 ± 4.3 versus 72.5 ± 3.4 nM, P > .05). However, irbesartan dilated 10 nM U46619-preconstricted coronary rings in a dose-dependent manner. Removal of the endothelium did not affect the relaxation induced by irbesartan. There were no differences in the IC₅₀ of irbesartan-induced vasodilation (IC₅₀: 0.7 ± 0.2 versus 0.9 ± 0.4, µM (intact rings versus endothelium-denuded rings; P > .05). In addition, in intact coronary rings preconstricted with 40 mM KCl, irbesartan (10⁹-10⁴ M) had no vasodilatory effects (data not shown; n = 4).

View larger version (21K): [in this window] [in a new window]   Fig. 5.   A, effect of copretreatment with irbesartan and indomethacin on U46619 dose-response curve. Irbesartan combined with indomethacin () significantly attenuated U46619-induced contraction-response curve compared with indomethacin (105 M)-treated rings alone (). Blockade of cyclooxygenase did not affect the antagonistic action of irbesartan. B, effect of copretreatment with irbesartan () and NO synthase inhibitor L-NAME (104 M) on the attenuation by irbesartan in U46619-induced vasoconstriction compared with L-NAME-treated rings alone (). L-NAME did not affect the inhibitory action of irebsartan. Studies were conducted in five or six dogs. Values are means ± S.E.

Competition by Irbesartan at TxA₂/PGH₂ Receptors. The effects of irbesartan to compete for the binding site of [³H]SQ29,548 to canine coronary vessel sections are illustrated in Fig. 6. U46619 competed for specific [³H]SQ29,548 binding at concentrations of 100 nM to 10 µM (percentage of competition: 98 ± 1 at 100 nM, 100 ± 0 at 1 µM, and 96 ± 4 at 10 µM, P < .05 compared with 0), but not at 10 nM (24 ± 12%). In contrast, irbesartan competed significantly for [³H]SQ29,548 binding only at the highest concentration of 100 µM (100 ± 0; P < .05 compared with 0). Lower concentrations were without significant effect on binding percentage of competition: 14 ± 9 at 100 nM, 35 ± 16 at 1 µM, and 25 ± 10 at 10 µM. From these data, the estimated K_i was 7 nM for U46619 and 10 µM for irbesartan.

View larger version (59K): [in this window] [in a new window]   Fig. 6.   Photomicrographs of TxA2 binding in 14-µm sections from the canine coronary artery. The first panel shows binding (total) detected with the ligand [3H]SQ29,548. Excess unlabeled 1 µM SQ29,548 (SQ), 1 µM U46619 (U46), and 100 µM irbesartan (Irb) significantly competed for binding.

Inhibition of U46619-Induced Human Platelet Aggregation by Irbesartan. Freshly isolated platelets from healthy volunteers were used for platelet aggregation studies. The TxA₂ agonist U46619 (0.1-10 µM) elicited dose-dependent human platelet aggregation in PRP (Fig. 7). At 1 µM, U46619 elicited near maximal aggregation. The degree of aggregation was not significantly increased at higher concentrations of U46619 (10 µM). Preincubation with 1 µM irbesartan for 10 min significantly inhibited platelet aggregation induced at the lower concentration of U46619 (0.1 and 0.5 µM) (P < .01), whereas at higher concentrations of U46619 (1 and 10 µM) irbesartan had no effect on platelet aggregation. Irbesartan at 10 µM abolished lower concentration of U46619-induced platelet aggregation and significantly prevented platelet aggregation of PRP at 1 µM U46619 (Fig. 8) (P < .01). In addition, pretreatment with 10 µM irbesartan for 10 min did not change 5 µM ADP-induced human platelet aggregation of PRP (data not shown; n = 2).

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Typical traces of concentration-dependent human platelet aggregation induced with U46619 during control conditions (A-D) and after pretreatment with 10 µM irbesartan (E-H).

View larger version (29K): [in this window] [in a new window]   Fig. 8.   U46619-stimulated human platelet aggregation of PRP in a concentration-dependent manner. Irbesartan pretreatment inhibited the platelet aggregation in dose-dependent fashion (106-105 M). Aggregation studies were conducted in six healthy human volunteers.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we demonstrated for the first time that irbesartan, the newly developed orally active, nonpeptide Ang II AT₁ antagonist, inhibits the TxA₂ receptor agonist U46619-induced vasoconstriction of canine coronary arteries and human platelet aggregation without changing the maximal response. Irbesartan also displaced TxA₂ receptor binding sites of coronary arteries. These findings are consistent with irbesartan acting as a competitive antagonist of the TxA₂/PGH₂ receptor in canine coronary arteries and human platelets. We reported previously that the nonpeptide AT₁ receptor antagonists losartan and its active metabolite EXP3174 competitively blocked TxA₂/PGH₂ receptors in canine coronary arteries and inhibited U46619-induced vasoconstriction (Li et al., 1997). In the present study, we found that irbesartan was more potent than losartan and equivalent in potency with EXP3174 in blocking U46619-mediated vasoconstriction (Li et al., 1997). The inhibiting effects of irbesartan in U46619-induced vasoconstriction were independent of the endothelium and not mediated by the release of either vasoactive prostaglandins or NO from the coronary arteries because the responses were not altered by preincubation with inhibitors of cyclooxygenase, NO synthase, and removal of vascular endothelium. Irbesartan specifically reversed U46619-induced vasoconstriction because KCl-induced preconstricted vascular rings were not influenced by irbesartan pretreatment. In contrast, the AT₁ receptor antagonists candesartan and valsartan, the AT₂ antagonist PD123319, and the ACE inhibitor lisinopril did not interact with the TxA₂/PGH₂ receptor in coronary arteries. This study demonstrates that the antagonistic effect of irbesartan on the TxA₂/PGH₂ receptor in coronary arteries and human platelets is specific for irbesartan, and these effects are not shared by the AT₁ antagonists candesartan and valsartan, the AT₂ antagonist PD 123319, and the ACE inhibitor lisinopril, at least at the doses studied. These studies suggest that irbesartan acts as a dual receptor antagonist, i.e., a "dipharmocophore," (Wexler et al., 1996) at both the AT₁ and the TxA₂/PGH₂ receptors in blood vessels and platelets. This action of the drug may play an important role in the prolonged blood pressure-lowering effects, renal protection and antiproatherogenic actions, that are linked to vasoconstriction and platelet aggregation (Schafer, 1996; Ruilope, 1997).

Our findings that the AT₁ receptor antagonists irbesartan, losartan, and EXP3174, but not candesartan and valsartan, interact with the TxA₂ receptor provide evidence that not all nonpeptide AT₁ receptor antagonists have similar pharmacological actions. The interaction of selective AT₁ receptor antagonists with the TxA₂ receptor indicates that these three AT₁ antagonists share cardiovascular actions that are dependent on a similar chemical structure rather than specific for the class of drugs. Irbesartan has a similar structure to losartan in that they both have an imadizole ring differing in side chain substitutions (Wexler et al., 1996). Irbesartan has carbonyl and ketone groups that replace the chloride and the hydroxymethyl groups of losartan, respectively. EXP3174, the active metabolite of losartan in vivo, is structurally very similar to losartan having a carboxylic acid in place of 5-hydroxymethyl group in the imidazole moiety of losartan. Like losartan, EXP3174 antagonizes both AT₁ and TxA₂/PGH₂ receptors; however, it has greater potency than losartan at both receptors (Wong et al., 1990; Li et al., 1997). The difference between losartan and EXP3174 at the AT₁ and TxA₂ receptors may be due to the change in electrical charge of the imidazole moiety of the drug (Wexler et al., 1996). The similarity of these three AT₁ antagonists at the TxA₂ receptor contrasts with the lack of effect in blocking U46619-induced vasoconstrictor responses of two other nonpeptide AT₁ receptor antagonists, candesartan and valsartan, and the nonselective, peptide Ang II receptor antagonist Sar¹Thr⁸-Ang II (Li et al., 1997). Structurally, candesartan has the imidazole ring fused with another heterocyclic ring with a carboxylic acid, and valsartan is a nonheterocyclic antagonist in which the imidazole ring of losartan is replaced with an acylated amino acid (Wexler et al., 1996). The peptide antagonist Sar¹Thr⁸Ang II nonselectively antagonizes all subtypes of angiotensin receptors, including the AT₁ receptor, but structurally does not overlap with any of the nonpeptide antagonists capable of acting at the TxA₂ receptor. Thus, our studies indicate that imidazole moiety of biphenyl tetrazole is the key structure of AT₁ antagonists required for blocking the TxA₂ receptor in blood vessels. In further support of this concept, we reported direct interactions of losartan with imidazole receptors in a previous study (Li et al., 1996).

Losartan is the prototype of orally active, nonpeptide AT₁ receptor antagonists without intrinsic agonist effects. It blocks Ang II-induced vasoconstrictor and dipsogenic responses, aldosterone secretion and catecholamine release (Timmermans et al., 1993) and reduces high blood pressure in most hypertension models studied (Moriguchi et al., 1994; Goa and Wagstaff, 1996). Losartan has an in vivo active hepatic metabolite EXP3174 after oral administration. EXP3174 has a longer plasma half-life and is ~10- to 15-fold more potent than losartan in antagonizing AT₁ receptors (Wong et al., 1990; Sachinidis et al., 1993). These findings may in part account for the long-lasting antihypertensive effects of losartan. Irbesartan is one of the newly developed orally active nonpeptide Ang II AT₁ receptor antagonists that selectively acts at AT₁ receptors with high affinity in different tissues. Irbesartan displaces labeled Ang II-binding sites and antagonizes the pressor response to Ang II in vivo (Cazaubon et al., 1993). In addition, irbesartan reduces high blood pressure with comparable potency to losartan in hypertensive animals and patients (Reeves et al., 1998). Irbesartan does not require biotransformation to become pharmacologically active in vivo.

A number of studies with losartan provide the basis for the suggestion that its pharmacological actions may not be mediated by Ang II receptor blockade exclusively. Ohlstein et al. (1992) reported that 48 h after administration of a single dose of losartan blood pressure was still reduced in the presence of normal responses to Ang I and Ang II in SHR. Several non-Ang II-related actions of losartan may explain its cardiovascular actions, which include stimulating production of vasodilator prostaglandins and NO (Jaiswal et al., 1991; Catalioto et al., 1994) and blocking ₁-receptors (Maeso et al., 1995), tachykinin (Picard et al., 1995), and imidazoline/guanidinium receptors in the central nervous system (Li et al., 1996). In addition to the studies described above from our and other laboratories (Liu et al., 1992; Bertolino et al., 1994; Corriu et al., 1995) showing that losartan and EXP3174 are competitive TxA₂/PGH₂ receptor antagonists in rat blood vessels and platelets (Li et al., 1998), we found that losartan enhanced acetylcholine-induced NO release and blocked the contractile effects of endothelium-derived contracting factor from blood vessels of aged SHR. The actions of losartan in these studies are consistent with our previous findings because the actions of endothelium-derived contracting factor have been ascribed to TxA₂/PGH₂ (Auch-Schwelk et al., 1990). In the perfused rat hindlimb preparation, irbesartan blocked the increase in U46619-induced perfusion pressure (K.B.B., unpublished data). These findings may provide further explanation for the diverse pharmacological actions of losartan and irbesartan due to their interaction with the TxA₂ receptor.

In our studies, we found that irbesartan blocked the TxA₂/PGH₂ receptor of human platelets. Platelets play an important role in arterial thrombosis and the onset of acute myocardial infarction after atherosclerotic plaque rupture (Verstraete, 1995). Inhibition of platelet aggregation has become a critical step in preventing thrombotic events that are associated with stroke, heart attack, and peripheral arterial thrombosis. TxA₂ synthase inhibitors and/or TxA₂ receptor antagonists have been considered to have advantage over the aspirin-like drugs in preventing platelet adhesion and aggregation (Schafer, 1996). Demonstration of an additional property of irbesartan as an agent that inhibits platelet aggregation is potentially of therapeutic value.

In our study, irbesartan blocked the TxA₂/PGH₂ receptor of canine coronary arteries as a competitive antagonist with K_b values ~215 to 236 nM. The estimated K_i from the competition data was 10 µM. This indicates that the apparent affinity of irbesartan for the TxA₂ receptor was at least 110-fold lower compared with that for the AT₁ receptor (Cazaubon et al., 1993). In the rat circulation, the concentration of irbesartan was estimated to reach ~390 µM after a dose of 10 mg/kg i.v. (Lacour et al., 1994), well within the estimated concentration of irbesartan effective at the TxA₂ receptors. In humans, the concentration of irbesartan in plasma was 1.5 to 2.8 µg/ml (~4-7 µM) after an oral therapeutic dose of irbesartan (150-300 mg) (Brunner, 1997). In our studies, we demonstrated that irbesartan at 1 µM significantly inhibited the TxA₂ analog U46619-induced vasoconstriction and human platelet aggregation in vitro. These findings suggest that the antagonistic effect of irbesartan on the TxA₂/PGH₂ receptor in blood vessels and platelets may contribute to blood pressure reduction and prevention of thrombosis during long-term treatment. Thus, irbesartan with a separate and specific action on the TxA₂/PGH₂ receptor may have additional therapeutic benefits over more selective angiotensin AT₁ receptor antagonists in preventing the vasoconstriction and platelet aggregation of TxA₂.