Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Irbesartan (SR47436, BMS-186295), chemically designated as 2-butyl-3-[[2'-(1H-tetrazole-5-yl)[1,1'-biphenyl]-4-yl]methyl]-1,3diazaspiro[4,4]non-1-en-4-one, is a potent, long-acting angiotensin II (AII)¹ receptor antagonist, with high selectivity for the AT₁ subtype (Cazaubon et al., 1993). The chemical structure of irbesartan is shown in fig. 1. Irbesartan and other AII receptor antagonists have the potential to offer advantages in safety and tolerability over earlier classes of drugs for the treatment of hypertension, diabetic nephropathy, and heart failure (Eberhardt et al., 1993; Gottlieb et al., 1993; Lacourcière et al., 1994). A clinical study in hypertensive subjects has demonstrated that irbesartan effectively lowers blood pressure with once-daily administration (van den Meiracker et al., 1995). The pharmacokinetics of irbesartan in healthy subjects has been determined (Necciari et al., 1994; Vachharajani et al., 1995). Irbesartan was found to undergo N-glucuronidation on the tetrazole moiety during in vitro incubations with hepatic microsomes from rats, monkeys, and humans (Perrier et al., 1994). In addition to the glucuronide conjugate of irbesartan, a number of oxidized metabolites were identified by LC/MS as metabolites of irbesartan in Macaca fascicularis monkeys (Tronquet et al., 1996). This report describes the isolation and identification of the metabolites of irbesartan from human urine and the quantification of some of these metabolites in plasma, urine, and feces after oral and iv administration of [¹⁴C]irbesartan to humans.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Structure of irbesartan. The site labeled with 14C is indicated with an asterisk. Designations of protons used for the interpretation of the NMR spectra of irbesartan and its metabolites are also shown and are that of Perrier et al. (1994)

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. [¹⁴C]Irbesartan, with a specific radioactivity of 19.8 µCi/mg, was provided by Sanofi Recherche (Montpellier, France). The radiochemical purity of the [¹⁴C]irbesartan was 98.8%. Nonradiolabeled irbesartan, SR 49498, chemically designated as 1-[(1-oxopentyl)amino]-N-[[2'-(1H-tetrazol-5-yl)[1,1'-biphenyl]-4-yl]methyl]cyclopentanecarboxamide, and the tetrazole N²-glucuronide of irbesartan were also obtained from Sanofi Recherche. Metabolites M1 through M8 were isolated from human urine and were used as reference standards. Ecolite liquid scintillation cocktail was purchased from ICN Biomedicals, Inc. (Costa Mesa, CA). For solid phase extractions, Bond Elut C₁₈ 6-cc cartridges were obtained from Analytichem International (Harbor City, CA). HPLC-grade solvents were used throughout. Acetonitrile was obtained from Burdick and Jackson (Muskegon, MI). Diethylamine, DMSO-d₆ (100.0 atom %), and methanol-d₄ (99.95 atom %) were obtained from Aldrich. Combined -glucuronidase/sulfatase from Helix pomatia (type H-1, G 0751), phenolphthalein glucuronic acid, sodium salt (P 0376), and D-saccharic acid-1,4-lactone (S 0375) were obtained from Sigma. All other chemicals were reagent grade or better.

Dosage and Sample Collection. Plasma, urine, and/or fecal samples were obtained from three clinical studies as follows.

Clinical study 1. In an open label, balanced, randomized, 3-way crossover study, each of 12 healthy male volunteers received a single 50-mg dose of irbesartan administered as follows: [¹⁴C]irbesartan iv solution infused over 30 min, [¹⁴C]irbesartan solution orally or nonradiolabeled irbesartan in a capsule. Blood, urine, and fecal samples were collected over 168 hr after the radiolabeled doses (75 µCi per subject). HPLC biotransformation profiles were determined for pooled plasma (1 and 6 hr) and urine samples (0-48 hr) from subjects that received [¹⁴C]irbesartan in this study. There were numerous missing fecal samples; thus, biotransformation profiles were not determined for fecal samples from this study. This study was later repeated (clinical study 2).

Clinical study 2. In an open label, balanced, randomized, 2-way crossover study, each of 6 healthy male volunteers received a single 150-mg oral dose (75 µCi) and a single 50-mg iv dose (75 µCi, 30-min infusion) of [¹⁴C]irbesartan on two separate occasions. Blood, urine, and fecal samples were collected over a period of 168 hr. HPLC biotransformation profiles were determined for representative pooled plasma (1 and 6 hr), urine (0-48 hr), and fecal samples (0-120 hr) from this study. These time intervals were chosen for pooling of the urine and fecal samples because most of the radioactivity (90% or greater) recovered in urine was recovered during the first 48 hr after dosing and most of the radioactivity in feces was recovered during the first 120 hr after dosing.

Clinical study 3. This study was a placebo-controlled, double blind within dose group, single and multiple dose study. Four groups of 9 volunteers each received a 150, 300, 600, or 900-mg dose of nonradiolabeled irbesartan on day 1 followed by placebo for days 2, 3, and 4. The study subjects at each dose level then received a daily dose of irbesartan for 1 week (days 5-11).

Biotransformation Profiles. HPLC was performed on a Hewlett Packard model 1090 liquid chromatograph equipped with a photodiode array ultraviolet detector. Chromatography was performed on a Waters µBondapak C₁₈ analytical column (3.9 × 300 mm, Part No. 27324, Millipore Corp., Milford, MA). A gradient mobile phase system was utilized for determination of biotransformation profiles. The mobile phase consisted initially of 20% acetonitrile and 80% water containing 0.1% diethylamine, adjusted to pH 5.2 with glacial acetic acid. The acetonitrile content of the mobile phase was increased in a linear gradient over a period of 15 min to a final composition of 40% acetonitrile. This composition was then maintained until the end of the run at 25 min. The mobile phase flow rate was 1.0 ml/min. All injections were performed using an automatic injector, which delivered 225 µl of sample. Nonradiolabeled reference irbesartan and irbesartan glucuronide (the glucuronide was not added to all samples, as it was in very limited supply) were added to urine and extracts of plasma and feces before HPLC analysis to allow unambiguous identity of the radioactivity corresponding to these metabolites. HPLC fractions were collected at 0.5-min intervals using a Gilson model 202 fraction collector (Gilson Medical Electronics, Middleton, WI). Each fraction of column eluate was mixed with 15 ml of Ecolite and counted for radioactivity using a Packard Tri-Carb 1900CA liquid scintillation analyzer. The radioactive fractions corresponding to each radioactive peak were summed to determine the relative percentage distribution of radioactivity for each metabolite (or metabolites, as some metabolites were not completely resolved by the HPLC system utilized). The assignment of the isolated metabolites (the isolation and identification of the metabolites of irbesartan from urine is described below) to the proper radioactive peaks in the biotransformation profile was further accomplished by separate HPLC runs using the same conditions as described above, where the radioactive metabolites isolated and identified from human urine (M1-M8) were co-injected with the urine or with the extract of the fecal samples (this procedure could not be utilized for plasma because of the low amount of radioactivity in the plasma samples). For HPLC runs with co-injected radioactive metabolites, an increase in a radioactive peak indicated the added metabolite was contained within that peak. A number of co-injection runs had to be performed for each sample to allow unambiguous assignment of the identity of each radioactive peak. Some radioactive peaks contained more than one metabolite, as some metabolites were not completely resolved by the HPLC system utilized.

Human Urine. A 360-µl aliquot of human urine was spiked with up to 20 µl of a methanol solution containing irbesartan as a reference standard. Following high speed centrifugation (10,000g) for 1 min, a 225-µl aliquot was injected into the HPLC.

Human Plasma. Plasma samples (1.0 ml) were deproteinized by adding 2.0 ml of acetonitrile while the sample was mixed on a Vortex mixer. After centrifugation of the acetonitrile/water mixture for 10 min at 700g, the supernatant fraction was removed and saved, and the protein precipitate was washed twice with 1.0 ml of acetonitrile/water (2:1,v/v). The combined supernatant fractions were evaporated to dryness under nitrogen and reconstituted in 430 µl of the HPLC mobile phase (initial conditions). The extraction recovery from plasma was 90% or greater in all cases. The reconstituted sample was spiked with approximately 20 µl of a methanol solution containing irbesartan and irbesartan glucuronide as a reference standard, and following high speed centrifugation for 1 min, a 225-µl aliquot was injected into the HPLC.

Human Feces. Fecal homogenates (1.0 g) were extracted by addition of 2.0 ml of acetonitrile while the sample was mixed on a Vortex mixer. The mixtures were sonicated for 5 min and then shaken mechanically for 20 min. After centrifugation of the acetonitrile/water mixture for 10 min at 700g, the supernatant fraction was removed and saved, and the precipitate was washed twice with 2.0 ml of acetonitrile/water (2:1,v/v). The extraction recovery from feces was approximately 100%. A 1.5-ml aliquot of the combined supernatant fraction was evaporated to dryness under nitrogen and reconstituted in 430 µl of the HPLC mobile phase (initial conditions). The reconstituted sample was spiked with 20 µl of a methanol solution containing irbesartan as a reference standard, and following high speed centrifugation for 1 min, a 225-µl aliquot was injected into the HPLC.

Combined -Glucuronidase/Sulfatase Enzyme Hydrolysis. Pooled urine (0-48 hr) was also incubated with combined -glucuronidase/sulfatase from H. pomatia. Incubations were conducted at 37^oC for 24 hr in 0.09 M sodium acetate (pH 4.8) at a final concentration of 909 -glucuronidase units/ml. Incubations were performed using phenolphthalein glucuronide (0.45 mM) as a positive control and with the -glucuronidase inhibitor, D-saccharic acid 1,4-lactone (24 mM), added to selected incubations to differentiate sulfatase activity from -glucuronidase activity.

Isolation of Metabolites from Human Urine. HPLC was performed on a Hewlett Packard model 1090 liquid chromatograph equipped with a photodiode array ultraviolet detector. Chromatography was performed on a Whatman Partisil 10 ODS-3 Magnum 9 semi-preparative reversed phase column (C₁₈, 9.4 mm × 50 cm). The mobile phase flow rate was 4.0 ml/min. Fractions of HPLC eluent were collected at 0.5- or 1-min intervals. HPLC runs were profiled for radioactivity by mixing eluent fractions with either 6 or 15 ml of Ecolite and counting using a Packard Tri-Carb 1900CA liquid scintillation analyzer.

System I. The mobile phase consisted initially of 15% acetonitrile and 85% water containing 0.1% diethylamine adjusted to pH 5.2 with glacial acetic acid. The acetonitrile content of the mobile phase was increased in a linear gradient over a period of 60 min to a final composition of 40% acetonitrile. This composition was then maintained till the end of the run at 65 min.

System II. Isocratic elution with 20% acetonitrile and 80% 0.05 M ammonium acetate was adjusted to pH 5.2 with glacial acetic acid.

System III. Isocratic elution with 21% acetonitrile and 79% 0.05 M ammonium acetate was adjusted to pH 5.2 with glacial acetic acid.

System IV. Isocratic elution with 24% acetonitrile and 76% 0.05 M ammonium acetate was adjusted to pH 4.5 with glacial acetic acid.

System V. Isocratic elution with 24% acetonitrile and 76% 0.05 M ammonium acetate was adjusted to pH 5.2 with glacial acetic acid.

System VI. Isocratic elution with 28% acetonitrile and 72% 0.05 M ammonium acetate was adjusted to pH 4.5 with glacial acetic acid.

System VII. Isocratic elution with 25.5% acetonitrile and 74.5% 0.05 M ammonium acetate was adjusted to pH 5.2 with glacial acetic acid.

System VIII. Isocratic elution with 30% acetonitrile and 70% 0.05 M ammonium acetate was adjusted to pH 5.2 with glacial acetic acid.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Scheme for the isolation and purification of irbesartan metabolites from human urine.

Purification of Fractions 1, 2, 3, 4, 5b, 6, and 7. Each fraction was further purified to yield a single metabolite by use of preparative HPLC utilizing the mobile phase systems indicated in fig. 2. This was accomplished by evaporating each fraction to dryness, redissolving the residue in mobile phase, and making multiple injections of 1.8 ml each until the entire fraction was processed. Eluate fractions from all runs that corresponded to a single peak were combined and evaporated to dryness in vacuo. The metabolites were each redissolved in approximately 3 ml of an acetonitrile/water mixture. To remove salt remaining from the mobile phase, each metabolite solution was submitted to solid phase extraction. The resulting solutions were each evaporated to dryness under nitrogen, and the residues were each redissolved in 25 ml of water. These aqueous samples were each loaded onto two 6-cc C₁₈ solid phase cartridges, which had been mounted one on top of the other. The cartridges were then washed with 25 ml of water, 25 ml of acetonitrile, and 25 ml of acetonitrile/water (50:50). The final two washes, which contained the metabolites, were combined and evaporated to dryness in vacuo and redissolved in 1 or 2 ml of methanol. Following HPLC analysis, which indicated each metabolite was pure, each of the metabolites (M1, M2, M3, M4, M7, M8, and M9) were submitted for NMR and mass spectral analysis.

Purification of Fraction 5a. Fraction 5a contained two metabolites (M5 and M6), which were purified by preparative HPLC system VI. The column eluate corresponding to M5 was combined and evaporated to dryness under nitrogen and submitted to solid phase extraction as previously described. M5 was redissolved in 2 ml of methanol. HPLC analysis indicated that M5 was pure. The column eluate corresponding to M6 was combined and evaporated to dryness under nitrogen and submitted to solid phase extraction as previously described. HPLC analysis indicated that M6 still contained some M5. The sample was chromatographed in HPLC system VI. Column eluate that corresponded to M6 was combined and evaporated to dryness in vacuo. M6 was submitted to solid phase extraction as previously described and was redissolved in 2 ml of methanol. HPLC analysis indicated that M6 was pure.

Spectroscopy. Proton NMR spectroscopy was performed on a Varian Unity Plus 400. All compounds were dissolved in DMSO-d₆. All chemical shifts (_H) are reported in ppm relative to tetramethylsilane as an internal standard. One-dimensional spectra were collected using both the S2PUL and PRESAT pulse sequences, whereas GCOSY and CYCLENOE pulse sequences were used for acquiring two-dimensional correlation spectra and one-dimensional NOE data, respectively.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Biotransformation profiles of pooled plasma (6 hr), urine (0-48 hr), and fecal samples (0-120 hr) from six healthy male volunteers given a single 150-mg oral dose of [¹⁴C]irbesartan are shown in fig. 3. Reference standards or previously isolated radioactive metabolites of irbesartan (M1-M8, isolation and identification of these metabolites from urine are discussed below) were co-injected with the pooled urine and fecal samples for determination of their retention times in this HPLC system. Co-injection of samples with isolated radioactive metabolites was accomplished in separate HPLC runs. The retention times of these metabolites are indicated in fig. 3, and the relative distribution of each metabolite in pooled plasma (1 and 6 hr), urine (0-48 hr), and feces (0-120 hr) is summarized in table 1. Biotransformation profiles of pooled plasma (1 and 6 hr) and urine (0-48 hr) from a separate clinical study where 12 healthy male volunteers received single 50-mg oral and iv doses of [¹⁴C]irbesartan (clinical study 1) gave similar results.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Biotransformation profiles of pooled plasma, urine, and fecal samples obtained from healthy subjects given a 150-mg oral dose of [14C]irbesartan. Biotransformation profiles were determined by HPLC. Retention times of irbesartan and its metabolites were determined by co-injection (solid arrows, co-injection of samples with isolated radioactive metabolites M1-M7 was accomplished in separate HPLC runs); radioactivity in plasma was too low to allow co-injection of isolated radioactive metabolites, so retention times of M1-M7 (dotted arrows) are those from the urine sample.

Biotransformation profiles of the plasma samples also were similar after either oral or iv administration of [¹⁴C]irbesartan. At 1 hr after drug administration, the amount of unchanged irbesartan ranged from about 81 to 88% of the plasma radioactivity, and the amount of glucuronide (M8) was about 7% of the plasma radioactivity; none of the other metabolites constituted greater than 4% of the plasma radioactivity. At 6 hr, the amount of unchanged irbesartan ranged from about 76 to 83% of the plasma radioactivity, and the amount of the glucuronide conjugate (M8) was about 5% of the plasma radioactivity. Other metabolites were present to a greater extent at 6 hr than at 1 hr but still constituted a minor portion of the plasma radioactivity, with the carboxylic acid metabolite (M3) accounting for about 9% and other metabolites less than 5% of the plasma radioactivity. These minor metabolites were estimated solely based upon retention times, as the peaks were barely above background noise and were too small to be properly co-chromatographed with isolated metabolites.

Biotransformation profiles of urine samples were also similar after oral or iv administration of [¹⁴C]irbesartan. Recovery of radioactivity in urine accounted for about 20% of the radiolabeled dose. In pooled human urine (0-48 hr), irbesartan and its glucuronide conjugate (M8) each accounted for about 10% or less of the urinary radioactivity. The predominant metabolite excreted in urine was M4, the -1 monohydroxylated metabolite, which constituted approximately 25% of the urinary radioactivity. The carboxylic acid metabolite (M3) constituted approximately 15% of the urinary radioactivity. Irbesartan and the eight metabolites identified accounted for about 90% of the urinary radioactivity (table 1).

Pooled urine (0-48 hr) from clinical study 1 was incubated with -glucuronidase/sulfatase, and only the radioactive peak corresponding to M8 was susceptible to enzymic hydrolysis. The -glucuronidase inhibitor, D-saccharic acid 1,4-lactone, successfully inhibited the enzymic hydrolysis of this metabolite. After enzymic hydrolysis, the radioactive irbesartan peak increased by an amount corresponding to the radioactive M8 peak before hydrolysis, indicating that M8 was a glucuronide conjugate of irbesartan.

About 30% of the fecal radioactivity corresponded to irbesartan; the remainder corresponded primarily to the metabolites identified in urine (table 1 and fig. 3). About 3.6% of the fecal radioactivity corresponded to SR 49498 after either the oral or iv dose of [¹⁴C]irbesartan. SR49498 is the dihydroimidazole ring-opened product of irbesartan (structure shown in fig. 4) and was present as a minor impurity in the radiolabeled dose.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Biotransformation of irbesartan in humans. Arrows indicate potential metabolic pathways. The structure of SR 49498, the dihydroimidazole ring-opened product of irbesartan, is also shown.

For metabolite isolation, an aliquot of the urine collected from subjects given an oral 50-mg dose of radiolabeled irbesartan in clinical study 1 was added as a tracer to urine collected on day 11 from subjects that received the 900-mg nonradiolabeled doses of irbesartan in clinical study 3. Urinary metabolites were concentrated on an XAD-2 column, extracted with ethyl acetate, and purified by extensive preparative HPLC to give metabolites M1 through M9.

Structural elucidation of the isolated metabolites was based upon mass spectrometry, proton NMR, COSY analysis, and chromatographic retention time using HPLC. Mass spectral analyses were performed using two different ionization techniques. Ion spray MS/MS produced ions resulting from cleavage of the methylene bridge between the butyl diazospirononenone and the biphenyl tetrazole moieties of irbesartan and its metabolites. This fragmentation of the molecule into two readily interpretable product ions allowed metabolic changes to be assigned to one or the other of these two ring systems. Fast atom bombardment (FAB) on the JEOL HX-110 mass spectrometer gave mass spectra of high resolution, allowing the determination of the exact mass of the (M+H⁺) ion of the isolated metabolites. A fragmentation scheme and a summary of both the ion spray MS/MS and FAB exact mass determinations are shown in table 2.

NMR spectral data are summarized in table 3. Proton NMR and interpretable COSY spectra were obtained on each metabolite. Initial attempts to obtain interpretable proton NMR spectra of the metabolites in MeOH-d₄ failed due to the gradual exchange of the 1b protons of the butyl side chain with deuterium. Thus, the proton NMR of all metabolites was determined in DMSO-d₆. The aromatic protons of the two phenyl rings present in irbesartan are clearly visible downfield in the NMR spectra between about  6.8 and  7.7. The methyl protons of the butyl side chain demonstrate a distinct triplet at about  0.8, and the methylene protons of this same butyl side chain are differentiated from each other at chemical shift values between  1.2 and  2.4. The benzylic methylene protons produced a sharp singlet at about  4.66. 

A scheme for the biotransformation of irbesartan in humans is presented in fig. 4. The evidence and the rationale for the structures assigned to these nine metabolites isolated from human urine follow.

Identification of M1. FAB mass spectrometry of M1 produced a protonated molecular ion (M+H)⁺ of m/z 461. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 461.2301 (m/z 461.2301 calculated for C₂₅H₂₉O₃N₆⁺), which was consistent with dihydroxylation of irbesartan. MS/MS of the m/z 461 parent ion using ion spray resulted in product ions at m/z 235, 207, 192, and 180 indicative of an unchanged biphenyl tetrazole ring system. A product ion at m/z 227 indicated two additional oxygen atoms were added to the butyl diazospirononenone ring of irbesartan.

Identification of M2. FAB mass spectrometry of M2 produced a protonated molecular ion (M+H)⁺ of m/z 459. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 459.2133 (m/z 459.2145 calculated for C₂₅H₂₇O₃N₆⁺), which was consistent with dihydroxylation and further oxidation of irbesartan. MS/MS of the m/z 459 parent ion using ion spray resulted in product ions at m/z 235, 207, and 192 indicative of an unchanged biphenyl tetrazole ring system. A product ion at m/z 225 indicated two additional oxygen atoms were present on the butyl diazospirononenone ring and that, in addition to dihydroxylation, further oxidation had occurred on this part of the molecule.

Identification of M3. FAB mass spectrometry of M3 produced a protonated molecular ion (M+H)⁺ of m/z 459. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 459.2148 (m/z 459.2144 calculated for C₂₅H₂₇O₃N₆⁺), which was consistent with dihydroxylation and further oxidation of irbesartan. MS/MS of the m/z 459 parent ion using ion spray resulted in product ions at m/z 235, 207, 192, and 180 indicative of an unchanged biphenyl tetrazole ring system. A product ion at m/z 225 indicated two additional oxygen atoms were present on the butyl diazospirononenone ring and that, in addition to dihydroxylation, further oxidation also occurred on this part of the molecule.

Identification of M4. FAB mass spectrometry of M4 produced a protonated molecular ion (M+H)⁺ of m/z 445. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 445.2372 (m/z 445.2352 calculated for C₂₅H₂₉O₂N₆⁺), which was consistent with monohydroxylation of irbesartan. MS/MS of the m/z 445 parent ion using ion spray resulted in product ions at m/z 235, 207, 192, and 180 indicative of an unchanged biphenyl tetrazole ring system. A product ion at m/z 211 indicated one additional oxygen atom was present on the butyl diazospirononenone ring.

Identification of M5. FAB mass spectrometry of M5 produced a protonated molecular ion (M+H)⁺ of m/z 445. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 445.2350 (m/z 445.2352 calculated for C₂₅H₂₉O₂N₆⁺), which was consistent with hydroxylation of irbesartan. MS/MS of the m/z 445 parent ion using ion spray resulted in product ions at m/z 235, 207, 192, and 180 indicative of an unchanged biphenyl tetrazole ring system. A product ion at m/z 211 indicated one additional oxygen atom was present on the butyl diazospirononenone ring.

Identification of M6. FAB mass spectrometry of M6 produced a protonated molecular ion (M+H)⁺ of m/z 443. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 443.2183 (m/z 443.2195 calculated for C₂₅H₂₇O₂N₆⁺), which was consistent with hydroxylation and further oxidation of irbesartan. MS/MS of the m/z 443 parent ion using ion spray resulted in product ions at m/z 235, 207, 192, and 180 indicative of an unchanged biphenyl tetrazole ring system. A product ion at m/z 209 indicated one additional oxygen atom was present on the butyl diazospirononenone ring and that, in addition to hydroxylation, further oxidation also occurred on this part of the molecule.

Identification of M7. FAB mass spectrometry of M7 produced a protonated molecular ion (M+H)⁺ of m/z 445. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 445.2341 (m/z 445.2352 calculated for C₂₅H₂₉O₂N₆⁺), which was consistent with monohydroxylation of irbesartan. MS/MS of the m/z 445 parent ion using ion spray resulted in product ions at m/z 235, 207, 192, and 180 indicative of an unchanged biphenyl tetrazole ring system. A product ion at m/z 211 indicated one additional oxygen atom was present on the butyl diazospirononenone ring.

Identification of M8. The isolated M8 co-chromatographed with authentic tetrazole N²-glucuronide of irbesartan obtained from Sanofi. FAB mass spectrometry of M8 produced a protonated molecular ion (M+H)⁺ of m/z 605. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 605.2744 (m/z 605.2724 calculated for C₃₁H₃₇O₇N₆⁺), which was consistent with conjugation of irbesartan with glucuronic acid. MS/MS of the m/z 605 parent ion using ion spray resulted in a product ion at m/z 429, indicating the loss of glucuronic acid to give irbesartan. The product ion at m/z 383 (207 + 176), resulting from cleavage of the bond between the biphenyl tetrazole ring system and the methylene bridge carbon, confirmed the position of glucuronidation as the tetrazole ring.

Identification of M9. The isolated M9 co-chromatographed with irbesartan. FAB mass spectrometry of M9 produced a protonated molecular ion (M+H)⁺ of m/z 429. Exact mass determination of the (M+H)⁺ ion resulted in an experimental value of m/z 429.2399 (m/z 429.2403 calculated for C₂₅H₂₉ON₆⁺), which was consistent with M9 being irbesartan. MS/MS of the m/z 429 parent ion using ion spray gave the same product ion spectrum as authentic irbesartan.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The primary route of metabolism of irbesartan in humans was oxidative leading to mono- and dihydroxylated metabolites or metabolites resulting from further oxidation of these hydroxylated metabolites (i.e. keto, hydroxy-keto, and carboxyl metabolites). Hydroxylation occurs primarily at the -1 carbon of the butyl side chain or on the spirocyclopentyl ring. Hydroxylation of the spirocyclopentyl ring occurs on the carbon atom  to the spiro carbon atom. As two stereochemical orientations are possible at each of these two carbon atoms, there are four possible isomers. Two of the three monohydroxylated metabolites identified were isomers resulting from hydroxylation of the spirocyclopentyl ring  to the spiro carbon atom. Attempts to distinguish between these isomers by use of NOE techniques were unsuccessful. Glucuronidation of the N² nitrogen of the tetrazole ring was also observed in humans. This tetrazole-N²--glucuronide was previously identified from in vitro incubations of irbesartan with microsomes from rat, monkey, and humans (Perrier et al., 1994). A tetrazole-N²--glucuronide conjugate has been identified as a metabolite of losartan (Stearns et al., 1991, 1992) and other biphenyl tetrazole AII receptor antagonists (Huskey et al., 1993; Kondo et al., 1996).

Irbesartan was the largest drug-related component in feces (about 30% of recovered radioactivity). Any irbesartan glucuronide conjugate excreted in the bile would be hydrolyzed by intestinal microflora to irbesartan. Thus, the irbesartan found in feces may represent excretion of both intact irbesartan and its glucuronide conjugate. The polar M1 seems to be present to a lesser extent in the feces; however, the other urinary metabolites are also present in feces. A minor amount of SR 49498 was observed in the biotransformation profiles, but the amount was only slightly more than the amount of SR 49498 present as an impurity in the radiolabeled dose of irbesartan. In fact, fecal homogenates spiked with [¹⁴C]irbesartan showed additional minor amounts (1-2%) of SR 49498 apparently formed during storage at -20°C. Incubation of fecal homogenate containing [¹⁴C]irbesartan for 18 hr at 37°C indicated little, if any, metabolism of irbesartan by the microflora present in the fecal samples. This suggests the SR 49498 observed in the biotransformation profiles resulted from the minor amount of impurity in the irbesartan dose with perhaps a trace amount of degradation of irbesartan that occurred during storage. However, the production of minor amounts of SR 49498 as a result of systemic metabolism cannot be excluded.

N-Dealkylation of irbesartan does not seem to be a major metabolic pathway in humans. N-Dealkylation of losartan was observed in incubations with liver slice preparations from rats and humans (Stearns et al., 1992). This cleavage was postulated to occur via oxidation of the benzylic carbon atom of losartan, followed by decomposition of the resulting hemiaminal metabolite. An analogous N-dealkylation is possible with irbesartan. If this cleavage were to occur with the [¹⁴C]irbesartan used for the clinical studies, the label would be retained in the resulting dihydroimidazole moiety. This N-dealkylation of irbesartan has been observed in monkey (Tronquet et al., 1996). Three monohydroxylated metabolites of the butyl spirocyclopentyl dihydroimidazole moiety were identified in the urine of monkeys given a 10 mg/kg oral dose of irbesartan. These metabolites would be expected to elute at or near the void volume in the HPLC system used for this study (fig. 3). The unidentified radioactivity in urine that eluted as polar metabolites at or near the void volume (fig. 3) could be metabolites resulting from N-dealkylation of irbesartan. However, this radioactivity corresponds to less than 10% of the urinary radioactivity and thus must be a minor metabolic pathway.

None of the identified human metabolites of irbesartan (M1-M8) are of greater potency than irbesartan (J. Gougat, unpublished results). M3, the major circulating metabolite in plasma (about 9% of the plasma radioactivity at 6 hr), was found to be more than 1000-fold less potent than irbesartan as an inhibitor of the AII receptor. These results suggest that irbesartan, and not its metabolites, is responsible for inhibition of the AII-mediated pressor response in humans.