Introduction

Abstract Introduction Results Discussion References

Leukocyte elastase is a serine protease capable of proteolytic degradation of a variety of substrates, including elastin and collagen, which are components of connective tissue. Specific inhibitors of leukocyte elastase are being explored as potential therapeutic agents for the treatment of inflammatory diseases, such as cystic fibrosis and rheumatoid arthritis where high amounts of extracellular elastase, either free or bound to its natural inhibitors, ₁-proteinase inhibitor and secretory leukocyte proteinase inhibitor, have been detected extracellularly (1-3).

Several classes of inhibitors of elastase have been synthesized and evaluated to date (4-19). L-694,458 (N-[1(R)-(1,3-benzodioxol-5-yl)butyl]-3,3-diethyl-2(S)-[4-[(4-methyl-1-piperazinyl)carbonyl]phenoxy]-4-oxo-1-azetidinecarboxamide,^1,2 fig. 1) is a member of the monocyclic -lactam series of elastase inhibitors, most of which are active after oral administration (8-14). It differs from the previously studied -lactam inhibitors, L-680,833 (11) and L-683,845 (13, 14), in that it contains a weakly basic methyl piperazine amide moiety at the 4-position of the phenol at the azetidinone C-4, instead of an acetic acid group. Unlike the acidic inhibitors, L-694,458 can inhibit elastase in circulating leukocytes as well as elastase released extracellularly (13). Similar to L-680,833 and L-683,845, L-694,458 is highly selective for the human enzyme, being much less active against elastase and elastase-like enzymes from other species, including rhesus monkey. The present report describes its disposition in rats and rhesus monkeys, the two species used in the toxicological evaluation of this drug candidate. Studies on its stability in blood and plasma and its effect on cytochrome P450 also are reported.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Chemical structure of L-694,458.

Materials and Methods

Chemicals. L-694,458 (fig. 1) and all other related chemicals, such as the monodemethylated metabolite of L-694,458, M5 and the monosubstituted urea, D1, were synthesized at Merck Research Laboratories (Rahway, NJ) according to published procedures ((19) and U.S. patents 183564 and 5591737). L-694,458 was supplied as the amorphous malate salt and the crystalline free base. All data presented in this report are expressed in free base equivalents. The radiolabeled compound, [butyl-3,4-³H]L-694,458 (radiochemical purity ~98%) was prepared with ³H at the positions shown in fig. 1. Pentadeuterated L-694,458, used as the internal standard in the LC-MS/MS assay, was synthesized with deuterium atoms at the positions indicated in fig. 1. Testosterone and its metabolites, including 6-hydroxytestosterone and 4-androstene-3, 17-dione were obtained from Sigma Chemical Co. (St. Louis, MO).

Microsomes. Fresh rat liver and frozen human and rhesus monkey liver tissue were used for the preparation of microsomes. Microsomes containing recombinant human cytochrome P450 isozymes were obtained from Gentest Corp. (Woburn, MA). The approximate specific activities of the CYP3A4- and CYP2C9-containing microsomes were 500 and 130 pmol/mg/min, as determined by testosterone 6-hydroxylation and diclofenac 4-hydroxylation, respectively.

Animal Studies. All animal procedures were performed in accordance with the highest standards for the humane handling, care, and treatment of research animals and were pre-approved by the Merck Institutional Animal Care and Use Committee. The care and use of research animals at Merck Research Laboratories meets or exceeds all applicable local, national, and international laws and regulations.

Rats. Male Sprague-Dawley rats were used in all experiments. For the pharmacokinetic studies, rats weighing 240-270 g (N = 3-5/dose) were implanted with cannulas in the jugular and femoral veins for dosing and blood collection, respectively. They were dosed iv with the malate salt at 1 and 5 mg/kg and orally with the malate salt and the free base at 10 mg/kg and with the free base at 40 mg/kg. For tissue distribution, mass balance and biliary excretion studies, rats weighing 300-400 g (N = 3-5/dose) were dosed with [³H]L-694,458 at 9 or 10 mg/kg po and 5 mg/kg iv. Intravenous dosing for the mass balance and tissue distribution studies was via the tail vein and for the biliary excretion studies via a previously implanted cannula in the femoral vein. Oral dosing was by gavage.

Rhesus monkeys. The studies were conducted in three periods using six male rhesus monkeys (4.5-7 kg). The animals were chaired without anesthesia and dosed iv via the saphenous vein and orally using a pediatric nasogastric tube. In the first period, three monkeys were dosed at 5 mg/kg iv, and another three monkeys were dosed orally at 10 mg/kg. In the second period, the first group of monkeys were dosed at 10 mg/kg po and the second group at 40 mg/kg po. The malate salt was used for both the iv and oral doses in the first and second periods. In the third period, the second group of monkeys were dosed orally with a suspension of the free base either at 30 mg/kg (N = 1) or 40 mg/kg (N = 2).

Dosing solutions. The malate salt of L-694,458 was dissolved in ethanol:saline (10:90) for iv dosing and in ethanol:0.5% methylcellulose (2:98) for oral dosing. The free base was prepared as a suspension in 0.5% methylcellulose. For iv studies with radiolabeled compound, [³H]L-694,458 was diluted with the malate salt to a specific activity of 135 µCi/mg (tissue distribution) or 15 µCi/mg (mass balance and biliary excretion). For oral studies, [³H]L-694,458 was diluted with the malate salt to a specific activity of 20 µCi/mg (tissue distribution) or with the free base to a specific activity of 15 µCi/mg (mass balance and biliary excretion).

Collection and handling of biological samples. In the pharmacokinetic studies, blood was collected from the femoral vein using a previously implanted cannula, and plasma, obtained by centrifugation at 4°C, was aliquoted and stored at 20°C until analyzed, within a week for the rat samples and within a month for the monkey samples. In the tissue distribution studies, blood was collected by heart puncture from three rats each at 30 min, 2, 6, and 24 hr after iv dosing, and at 2 and 6 hr after oral dosing. Tissues were removed and combusted either whole (cut into small pieces) or after homogenization. In the biliary excretion studies, bile was collected every hour for the first 8 hr and at 8-24, 24-48, and 48-72 hr from a previously implanted cannula in the bile duct. Urine and feces in the biliary excretion and mass balance studies were collected at 24-hr intervals for 3 days. Radioactivity in the dose, plasma, bile, urine, and HPLC fractions was determined by direct counting of aliquots. Radioactivity in the tissues and feces was determined by combustion of homogenates and counting of the resulting ³H₂O. Acetonitrile extracts of plasma, liver, lung, bile, urine, and feces were analyzed by HPLC using method 1, as described in the section on HPLC analysis.

In Vitro Stability in Blood and Plasma. The in vitro stability of [³H]L-694,458 (10 and 25 µM) in rat, monkey, and human blood and plasma was determined after incubation at 37°C for 1, 4, and 24 hr. Acetonitrile supernatants were analyzed by HPLC using method 2 and by LC-MS/MS. For the isolation of the degradation product D2, ethyl acetate extracts of incubations with plasma were chromatographed on a phenyl column eluted isocratically with acetonitrile:water (55:45). The isolated product was analyzed by NMR spectroscopy. Possible degradation of L-694,458 during storage and sample processing was investigated by storing samples of rat and monkey plasma containing known amounts of [³H]L-694,458 at 20°C for periods up to 2 months and processing them by solid-phase extraction, as described for the pharmacokinetic studies, except that the reconstituted extracts were analyzed by HPLC methods 1 and 2.

In Vitro Metabolism of L-694,458. L-694,458 was incubated at 37°C with liver microsomal preparations from male rats, rhesus monkeys, and humans in the presence of an NADPH-regenerating system. Incubations to determine the in vitro metabolite profiles were carried out at 10, 25, and 100 µM concentrations of L-694,458 using 0.5-2 mg (rat and monkey) or 1 mg (human) of liver microsomal protein/ml for 10 min to 1 hr. To determine the V_max and K_m, [³H]L-694,458 was incubated with 0.2 mg of rhesus monkey liver microsomal protein/ml for 4 min. The rates of disappearance of L-694,458 and appearance of the three major metabolites, M1, M5, and M6, were determined at six substrate concentrations (0.25 to 15 µM), and the data were fitted to the Michaelis-Menten equation using nonlinear least squares. Incubations also were carried out with suspensions of freshly isolated rat hepatocytes (1 × 10⁶ cells/ml), and with small segments of monkey jejunum (0.1 g of tissue cut into ~1 cm squares/ml of phosphate buffer). Acetonitrile extracts of incubation mixtures were analyzed using HPLC method 1, as described in the section on HPLC analysis.

Identification of the Major Cytochrome P450 Isozyme that Metabolizes L-694,458 in Human Liver Microsomes. The effect of specific P450 inhibitors and/or substrates on the metabolism of L-694,458 was determined by incubating [³H]L-694,458 (10 µM) with human liver microsomes (1 mg protein/ml) at 37°C for 30 min in the presence of an NADPH-regenerating system. The reversible inhibitors, ketoconazole (inhibitor of CYP3A), sulfaphenazole (CYP2C), quinidine (CYP2D6), enoxacin (CYP1A2), and 4-methylpyrazole (CYP2E1) were added at the same time as L-694,458. The CYP3A suicide inhibitors, troleandomycin (TAO) and gestodene were pre-incubated with microsomal preparations in the presence of an NADPH-regenerating system for 30 min before [³H]L-694,458 was added. Ketoconazole and sulfaphenazole were added at 1, 10, and 100 µM, and all other inhibitors at 100 µM. Incubations with microsomal preparations containing recombinant human P450 isozymes were conducted at 37°C for 1-4 hr using 2-4 mg microsomal protein/ml. Cytochrome c reductase was added to increase metabolism, except in incubations with CYP2C9-containing microsomes, which contained co-expressed cytochrome P450 reductase. Acetonitrile supernatants were analyzed by HPLC method 1. The concentrations of L-694,458 and its metabolites were determined from the percentage of radioactivity eluting in each peak.

Effect of L-694,458 on the In Vitro Metabolism of Testosterone. Human and monkey liver microsomal preparations were pre-incubated at 37°C for 30 min with L-694,458 (10 and 100 µM) or solvent (control) in the presence of an NADPH-regenerating system. Testosterone (100 µM) was added and aliquots were removed at 10, 20, and 30 min and analyzed by HPLC on a Zorbax Rx-C8 column eluted with water:acetonitrile:TFA (60:40:0.1). The identity of the testosterone metabolites was determined by co-elution with authentic materials.

In Vitro Interaction of L-694,458 with Rhesus Monkey Liver Microsomal Cytochrome P450. Rhesus monkey liver microsomes were incubated with L-694,458 in the presence or absence of an NADPH-regenerating system at 37°C for 0.5, 1, and 4 hr and aliquots were scanned from 500 to 400 nm on a Shimadzu UV160 UV-visible spectrophotometer against a reference cuvette containing microsomes and the NADPH-regenerating system.

Analytical Procedures. Quantitation of L-694,458 in plasma by LC-MS/MS. Concentrations of L-694,458 in plasma were determined by LC-MS/MS after solid-phase extraction. Aliquots of plasma (rat, 0.2 ml; monkey, 0.5 ml) were mixed with the internal standard (pentadeuterated L-694,458), diluted with water to 1 ml, and loaded onto CN cartridges. After a wash with water, the cartridges were eluted with methanol. The reconstituted extracts were analyzed by LC-MS/MS on a Spherisorb phenyl column using a mobile phase consisting of acetonitrile:50 mM ammonium acetate:formic acid (70:30:0.1). The column effluent was analyzed directly on a SCIEX API III mass spectrometer using the heated nebulizer interface. The precursor/product ion combinations of m/z 565/389 and 570/392 were used for selected reaction monitoring of L-694,458 and its internal standard, respectively. Quantitation was achieved using a standard curve generated from the mean of five replicates which were made by adding increasing amounts of L-694,458 (0.5-500 ng) and 20 ng of internal standard to control plasma. The lower limit of quantitation in most studies was 1 ng/ml for 0.2 ml rat plasma and 1 ng/ml for 0.5 ml monkey plasma.

Pharmacokinetic calculations. Pharmacokinetic parameters were determined using standard noncompartmental methods. Plasma AUC and AUMC were calculated using UNICUE with log-linear trapezoidal method for interpolation (20). Extrapolation to infinity was done using the terminal rate constant determined by linear regression from the slowest phase of the log plasma concentration-time curve in each animal and at each dose, except as indicated.

Radioactivity measurements. Direct liquid scintillation counting was carried out using a Beckman LS5000TD counter. Combustion was done in a Packard TriCarb Sample Oxidizer, Model B301, followed by counting of the resulting ³H₂O. Quench correction was carried out by the external ratio method.

HPLC analysis. Two HPLC methods were used for the analysis of samples from the studies with L-694,458. In method 1, samples were analyzed on a Zorbax SB-CN column, eluted isocratically with a mixture of water:acetonitrile (60:40, v/v) with 10 mM ammonium acetate and 0.1% TFA in both solvents. In method 2, samples were analyzed on a Zorbax phenyl column eluted isocratically with acetonitrile:water:TFA (54:46:0.2,v/v). The UV absorbance of the eluate was monitored at 220 nm, and radioactivity was measured with an on-line radiometric detector. Parent and metabolite identification was accomplished by co-elution with authentic material and by LC-MS/MS analysis of extracts, as reported in the accompanying paper (21).

NMR analysis. Proton and HMQC (22) NMR spectra of D2, the plasma degradant of L-694,458, were recorded on a Varian Unity 500 MHz spectrometer in DMSO-d₆ and were referenced to the residual solvent signal at 2.49 ppm for ¹H and 49.0 ppm for ¹³C.

	Results

Abstract Introduction Results Discussion References

Pharmacokinetics. Rats. After iv dosing at 1 and 5 mg/kg, concentrations of L-694,458 in plasma declined rapidly within the first hour, and then more slowly with a terminal t_1/2 of ~1.5 hr (fig. 2 and table 1). The mean values of plasma clearance, mean residence time (MRT), and Vd_ss of L-694,458 were similar at the two doses studied, ~26 ml/min/kg, 2.5 hr, and 4 liters/kg, respectively (table 1).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Mean plasma concentrations of L-694,458 in rats. - 5 mg/kg iv; -- 10 mg/kg po (free base); -  - 40 mg/kg po (free base).

Rhesus monkeys. In rhesus monkeys dosed iv at 5 mg/kg, concentrations of parent compound in plasma (fig. 3) were similar to those measured in rats at the same dose. Plasma clearance, terminal t_1/2, and Vd_ss also were similar (table 1). In contrast, after oral dosing, L-694,458 concentrations were lower and more variable than in rats. Peak concentrations of L-694,458 in six monkeys dosed with the malate salt at 10 mg/kg ranged from 0.02 to 0.35 µg/ml (fig. 3 and table 2) and were reached at a later time than in rats (between 2 and 6 hr). The bioavailability of the 10 mg/kg dose was 39%. Drug concentrations also were variable when a solution of the salt was dosed at 40 mg/kg. At this dose, concentrations of L-694,458 in the plasma of two of the three animals were still increasing between 8 and 24 hr. In a third animal, peak concentrations were reached at ~8 hr. Similar results were obtained when the same monkeys were dosed with a suspension of the free base. In two monkeys dosed at 40 mg/kg, peak concentrations of L-694,458 in plasma were 0.53 and 0.66 µg/ml at 24 hr, declining with a much longer t_1/2 (6 hr) than after iv dosing, to 2 and 4 ng/ml at 72 hr. Delayed t_max also was observed in a third monkey dosed at 30 mg/kg, with peak concentrations of 0.39 µg/ml at 24 hr. As in rats, plasma AUC increased more than proportionally with the increase in the oral dose, and bioavailability of the higher doses could not be determined accurately.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Mean plasma concentrations of L-694,458 in rhesus monkeys. (), 5 mg/kg iv; (), 10 mg/kg po (salt); () 40 mg/kg po (salt);() 40 mg/kg po (free base).

Biliary Excretion and Mass Balance in Rats. [³H]L-694,458 was well absorbed after oral dosing at 10 mg/kg, as indicated by the high recovery of radioactivity in bile and urine of bile duct-cannulated rats. About 83% of the dose was excreted in bile and another 6% in urine over a 72-hr period. Only 11% of the dose was recovered in the feces. When rats were dosed iv at 5 mg/kg, ~77% of the dose was excreted in bile, 9% in urine, and 3% in feces. In noncannulated rats, 92% of an iv dose and 75% of an oral dose were excreted in feces, with another 13% and 14%, respectively, recovered in urine (data not shown).

Tissue Distribution in Rats. The distribution of radioactivity to selected rat tissues was determined at 2 and 6 hr after a 9 mg/kg oral dose of [³H]L-694,458. In all seven tissues examined (adrenals, heart, kidneys, liver, lung, pancreas, and spleen), radioactivity concentrations were much higher than in plasma with tissue to plasma ratios of 4-16 at 2 hr and 2-14 at 6 hr. Concentrations were highest in the liver (22.6 and 13.8 µg drug eq/g tissue at 2 and 6 hr, respectively), and lung (20.1 and 11.3 µg drug eq/g at 2 and 6 hr, respectively). A similar tissue distribution pattern was observed after iv dosing of [³H]L-694,458 at 5 mg/kg. At 24 hr after iv dosing, concentrations were 0.3 µg eq/g in most tissues, except the liver (0.8 µg eq/g) and the small and large intestines (data not shown).

Metabolism. In vitro. L-694,458 underwent metabolism to several products in the presence of monkey and human liver microsomal preparations (fig. 4). The major metabolites were identified by LC-MS/MS and NMR (N-oxide only) analysis, as described in the accompanying paper (21), and were determined to be formed via three primary metabolic pathways: O-demethylenation of the methylenedioxyphenyl group to give the catechol product (designated as M1), N-demethylation of the methylpiperazine group to give M5, and addition of oxygen to the nitrogen of the piperazine ring to form the N-oxide (M6). The N-oxide was the major metabolite in rat liver microsomes, while the catechol was the major metabolite in monkey and human liver microsomes. Three other hydroxylated metabolites, M2, M3, and M4, were observed in rat and monkey but not human incubations and are discussed in the accompanying paper (21). Rat hepatocyte suspensions generated a similar metabolite profile for L-694,458 as rat liver microsomes, while monkey jejunum gave a similar metabolite profile as monkey liver microsomes. After a 4-hr incubation of [³H]L-694,458 (25 µM) with segments of monkey jejunum (0.1 g tissue/ml), about 53% of the radioactivity eluted in the area of the parent, ~10% in the area of M5, and 22% in the area of M1.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   In vitro metabolism of L-694,458 in rat, monkey and human liver microsomes. Incubations of 100 µM compound were carried out at 37°C for 1 hr in the presence of an NADPH-regenerating system. The microsomal protein concentration was 2 mg/ml (rat and monkey) or 1 mg/ml (human). Acetonitrile extracts were analyzed by HPLC on a SB-CN column with UV detection at 220 nm.

In Vivo. L-694,458 was the major radioactive component in plasma, lung, and liver of rats dosed with [³H]L-694,458 orally at 9 mg/kg, contributing to 50-60% of the radioactivity at 2 hr and 25-30% at 6 hr. LC-MS/MS analysis showed the presence of several metabolites, including M1, M5, and M6 (21). Rat bile, feces, and urine contained mostly metabolites, with <5% of the radioactivity eluting in the area of the parent compound (data not shown). More than 40 metabolites were identified from bile by LC-MS/MS analysis (21). These data indicate that in rats, L-694,458 is eliminated by metabolism followed by excretion of the metabolites in feces via the bile.

Identification of the Major Cytochrome P450 Isozyme Responsible for the Metabolism of L-694,458 in Human Liver Microsomes. The major cytochrome P450 isozyme(s) catalyzing the metabolism of L-694,458 in human liver microsomes was identified as belonging to the CYP3A subfamily on the basis of results from three types of studies. These studies investigated i) the effect of isoform-specific P450 inhibitors and/or substrates on the metabolism of L-694,458 in human liver microsomes, ii) the correlation between the rate of metabolism of L-694,458 with different P450 isozyme activities of eight preparations of human liver microsomes, and iii) the capacity of recombinant human P450 isozymes to metabolize L-694,458.

Effect of L-694,458 on Testosterone Metabolism. To investigate the possibility that metabolism of the methylenedioxyphenyl moiety of L-694,458 to the catechol (M1) involves the formation of a reactive intermediate, as has been shown for other methylenedioxyphenyl-containing compounds (23-25), the effect of L-694,458 on testosterone metabolism was determined. Testosterone was used as a model substrate for several cytochrome P450 isozymes, including CYP3A, the major isozyme that metabolizes L-694,458. Pre-incubation of human and rhesus monkey liver microsomes with L-694,458 in the presence of an NADPH-regenerating system resulted in inhibition of the CYP3A activity of the microsomes, as indicated by a decrease in the conversion of testosterone to its 6-hydroxylated metabolite (table 4). Pre-incubation of human liver microsomes with L-694,458 at 10 µM and 100 µM resulted in ~83% inhibition of testosterone 6-hydroxylation at both concentrations without affecting the formation of androstenedione, the other major metabolite. In rhesus monkey microsomes, testosterone 6-hydroxylation was inhibited by 83 and 94% at 10 µM and 100 µM of L-694,458, respectively, while androstenedione formation was actually enhanced substantially.

Cytochrome P450 Difference Spectrum in Microsomes Incubated with L-694,458. Incubation of L-694,458 with rhesus monkey liver microsomes yielded a typical type II optical binding spectrum with a trough minimum at 412 nm and a maximum at ~424 nm (not shown). Addition of an NADPH-regenerating system gave a type III difference spectrum, with peaks at 427 nm and 455 nm, as shown in fig. 5 for the 1-hr incubation mixture.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Type III optical difference spectrum recorded upon incubating L-694,458 with liver microsomes and NADPH.

In Vitro Stability in Blood and Plasma. [³H]L-694,458 was unstable in rat blood and plasma and relatively stable in monkey and human blood and plasma. About 30 and 50% degradation was observed after a 4-hr incubation at 37°C in rat blood and plasma, respectively. At the end of a 24-hr incubation with rat plasma, L-694,458 was almost completely degraded, as shown by HPLC analysis. Parent compound accounted for ~75, 58, and 15% of the total radioactivity in 24 hr incubations with human, monkey, and rat blood, respectively. No degradation was observed when L-694,458-containing rat and monkey plasma samples were stored at 20°C for up to 1 week or 2 months, respectively (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Radiochromatogram of an incubation mixture of [3H]L-694,458 with rat plasma. The compound was incubated at 25 µM with fresh plasma for 4 hr at 37°C. Acetonitrile supernatants were analyzed by HPLC as described in Materials and Methods. The structures of the degradants were determined by MS and NMR spectroscopy.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Low-field region of the two-dimensional 1H-13C correlation spectrum (HMQC) of the degradant D2 along with the corresponding region of the 1H NMR spectrum (top).

	Discussion

Abstract Introduction Results Discussion References

Disposition. L-694,458 exhibited similar pharmacokinetic parameters in rats and rhesus monkeys after iv dosing at 1 and/or 5 mg/kg, with mean values of plasma clearance, terminal half-life and Vd_ss of 26-34 ml/min/kg, ~2 hr, and 4-5 liters/kg, respectively. However, the apparently similar pharmacokinetics of L-694,458 in rats and rhesus monkeys may be serendipitous. On one hand, the higher affinity of L-694,458 for rhesus leukocyte elastase (13) and its lower metabolic stability in monkey liver would be expected to enhance its clearance in monkeys, while on the other hand, its greater instability in rat plasma would result in a higher clearance in rats.

In Vitro Stability. L-694,458 was unstable in rat plasma, degrading to two major products, whose structures are shown in fig. 6. The mechanism for the formation of D1 and D2 is probably similar to that involved in the inhibition of human leukocyte elastase by -lactam compounds (11, 12), namely, an initial attack by a nucleophile on the lactam carbonyl and decarboxylation followed by loss of the phenoxy acetic acid moiety to give D2 or decomposition to the monosubstituted urea D1. D1 could also be formed by cleavage of D2, which was observed in vitro in the presence of acid. Based on the inhibitory effect shown by organophosphorus compounds such as BNPP, it can be postulated that the degradation of L-694,458 in rat plasma is mediated by a type B esterase as the catalyst (26).