Materials and Methods

Animals

The following strains of animals were used in these studies: beagle dogs, rhesus monkeys (Macaca mulatta), Dunkin-Hartley/Pirbright-White guinea pigs (Interfauna, Tuttlingen, Germany), HsdWin:NMRI mice (Harlan-Winkelmann, Borchen, Germany), Chbb/Thom rats, and New Zealand White rabbits (bred in the laboratories of Boehringer Ingelheim Pharma KG, Biberach, Germany).

Chemicals

BIIL 284 BS, BIIL 260 CL, BIIL 315 ZW, CGS 25019 C, and LY 293111 were synthesized in the laboratories of Boehringer Ingelheim Pharma KG. Structures of Boehringer substances are shown in Fig. 1. LTB₄ (free acid) was purchased from Paesel GmbH (Frankfurt, Germany). [5,6,8,9,11,12,14,15(n)-³H]Leukotriene B₄, specific activity ca. 200 Ci/mmol, was purchased from Amersham Buchler (Braunschweig, Germany). All other chemicals and reagents were from Merck (Darmstadt, Germany) or Serva (Heidelberg, Germany) unless otherwise specified in the methods.

Cell Line

U937, a monocyte/macrophage differentiated histiocytic cell line of human origin was obtained from the American Type Culture Collection (Rockville, MD) (accession number ATCC CRL 1593). U937 cells were routinely maintained at 37°C in a humidified atmosphere of 5% CO₂, 95% air in RPMI-1640 cell culture medium supplemented with 10% fetal bovine serum (Boehringer-Mannheim, Mannheim, Germany), 2 mM glutamine, and nonessential amino acids. Cultures were split and subcultured every 3 to 4 days to keep a cell density of between 0.3 × 10⁶/ml and 1 × 10⁶/ml.

For dimethyl sulfoxide differentiation, exponentially proliferating cells were harvested by centrifugation and resuspended at 0.3 × 10⁶/ml in the medium specified above, additionally containing 1.25% (w/v) dimethyl sulfoxide. After 4 days the cells differentiated into a line with monocyte-like characteristics (Harris and Ralph, 1985).

Neutrophil Isolation

Human PMNLs were prepared from fresh peripheral blood samples from volunteers not receiving medication essentially as described byRoos and de Boer (1986). In brief, blood withdrawn with citric acid anticoagulant, was mixed 1:1 with a solution containing 140 mmol/l NaCl, 9.2 mmol/l Na₂HPO₄, 1.2 mmol/l NaH₂PO₄, 13 mmol/l sodium citrate, pH 7.4. This mixture was centrifuged (1000g for 20 min at room temperature) over a layer of Percoll (Amersham, Amersham, UK) supplemented with 13 mmol/l sodium citrate and 0.5% bovine serum albumin (BSA). The Percoll had a density of 1.077 g/ml. Contaminating erythrocytes in the neutrophil pellet were subsequently lysed with 3 volumes of a buffer containing 155 mmol/l NH₄Cl, 10 mmol/l KHCO₃, and 0.1 mmol/l EDTA, pH 7.4, for 10 min at 0°C followed by two washing steps (400g, 5 min, 4°C) in PBS. A purity of 98% PMNLs and a vitality of at least 95% (trypan blue exclusion) was routinely achieved. Dog (beagle) and rhesus monkey PMNLs were isolated by essentially the same technique with similar yield. Casein-elicited rat and guinea pig neutrophils were obtained by peritoneal lavage 24 h after intraperitoneal injection of 4 ml of a casein suspension (6 g of casein in 50 ml of isotonic saline). Viability was routinely >98% and purity >85%.

Membrane Preparation from Isolated Neutrophils

The cells, washed in PBS were suspended at ca. 4 × 10⁷cells/ml in the same buffer supplemented with 0.1 mmol/l phenylmethylsulfonyl fluoride (Serva). After equilibration for 30 min at 4°C in a nitrogen cell disruption bomb (1000 psi = 70 bar), cells were disintegrated by slow release into atmospheric pressure. The unbroken cells and cell organelles were removed by centrifugation (1000g, 15 min, 4°C). The membrane fraction obtained by centrifugation at 50,000g (1 h, 4°C) was resuspended in a volume of buffer sufficient to adjust the protein concentration to 5 mg of protein/ml. Membranes were stored in aliquots at −80°C before use.

In Vitro Assays for LTB₄ Function

LTB₄ Receptor Binding.

Competition studies with intact human PMNLs or isolated PMNL membranes were performed with 0.3 nM [³H]LTB₄ in a total volume of 0.5 ml. The assay mixtures, containing radioligand, cells, or membranes and the appropriate dilutions of the test substances (10⁻⁵–10⁻¹¹ mol/l) in PBS, were incubated at 0°C (ice bath) in case of vital cells or at 22°C (membranes) for the times indicated. The incubation was terminated by rapid filtration through Whatman GF/C filters. Filters were washed three times with ice-cold incubation buffer (10 mmol/l HEPES, pH 7.4, supplemented with 145 mmol/l NaCl, 1 mmol/l MgCl₂, 5 mmol/l KCl, 0.5 mmol/l Na₂HPO₄, 6 mmol/l glucose, 0.1% bovine serum albumin). Radioactivity was measured by liquid scintillation counting (Beckman, München, Germany) after adding 4 ml of Opti Fluor (Packard, Meriden, CT) to dried filters. Each assay was performed in triplicate and the assays were repeated as indicated in the tables. Specific binding was defined as total binding minus nonspecific binding determined in the presence of 0.1 μmol/l LTB₄.

Measurement of Cytoplasmic Calcium Concentrations upon LTB₄ Stimulation of PMNLs.

Freshly isolated PMNLs were incubated 1 h at 37°C with 4 μmol/l Fura-2 AM (Molecular Probes, Eugene, OR) in PBS buffer (140 mmol/l NaCl, 9.2 mmol/l Na₂HPO₄, 1.2 mmol/l NaH₂PO₄), and then centrifuged at 300g, resuspended in PBS plus test compound (or vehicle control), and transferred to a stirred cuvette held at room temperature in an LS 5 spectrofluorimeter (PerkinElmer, Norwalk, CT). The intensities of fluorescence (F) at the emission wavelength 510 nm were measured first at an excitation wavelengths of 340 nm and then at 380 nm and the ratio (R = F₃₄₀/F₃₈₀) of Fura-2 fluorescence calculated. LTB₄ was then added to a final concentration (unless otherwise indicated) of 100 nmol/l and the changes in F₃₄₀, F₃₈₀, and thus R measured every 3 s. R is proportional to the intracellular cytosolic Ca²⁺ concentrations. Absolute Ca²⁺ concentrations could be calculated as described by Minta et al. (1989). For each individual test sample, a baseline value of R, R_prestim, was determined before LTB₄ stimulation. Ca²⁺ entry inhibiting potencies of test agents were quantified by integrating measurements of R (R_prestim set zero) over the 1st min after addition of LTB₄.

Neutrophil Chemotaxis.

PMNLs were isolated essentially as previously described. Percoll (Percoll density 1.077) was layered under citrate-buffered blood from different healthy donors. For sedimentation, blood was mixed with ammonium chloride buffer and allowed to stand for 15 min at room temperature. The pelleted PMNLs where washed and chemotaxis was measured as described by Lippert et al. (1998), with minor modifications. A 96-well microchemotaxis chamber was used with a 96-well microtiter plate (Costar, Cambridge, MA) as lower chamber. The microtiter plate was filled with 50 nM LTB₄ and the respective test compounds in the concentrations from 0.1 to 30 nmol/l in a final volume of 350 μl in PBS (with 0.1% BSA, 1 mM CaCl₂, 1 mM MgCl₂). The upper chambers were separated by a polycarbonate filter (8-μm pores; Nucleopore, Pleasanton, CA) and filled with 300 μl of 1 × 10⁶ PMNLs in PBS, including test compounds. After incubation (120 min at 37°C), the microtiter plate, including filter membrane was centrifuged at 1300g, the supernatant aspirated, and sedimented cells lysed with 100 μl of 0.25% Triton-X 45 at room temperature for 30 min.

Migrated cells were quantified as myeloperoxidase activity as increase of optical density of the oxidation product of the substrateo-dianisidine dihydrochloride per minute at 450 nm over 5 min (Bradley et al., 1982). IC₅₀ was derived graphically. Myeloperoxidase activity has previously been reported to correlate with neutrophil numbers.

Specificity control studies were also carried out in which the stimulator LTB₄ (final concentration 30 nM) was replaced with formyl-methionyl-leucyl-phenylalanine (30 nM), interleukin-8 (3 nM), C5a (5 nM), platelet-activating factor (100 nM), tumor necrosis factor-α (0.01 nM), or phorbol-12-myristate-13-acetate (100 nM).

In Vitro LTB₄-Induced PMNL CD11b (Mac-1) Expression.

Venous blood samples were incubated 20 min at 37°C with 10 μl of LTB₄ (final concentration 40 nM) and indicated concentrations of respective test compounds. After incubation of 30 min at 4°C with saturating concentration of a fluorescein isothiocyanate-labeled monoclonal anti-CD11b antibody (Bear-1; Coulter, Krefeld, Germany), red blood cells were lysed by formic acid and white blood cells were fixed with paraformaldehyde using a Q prep automated device (Coulter). The degree of fluorescent staining was determined on a Coulter Epics XL-MCL flow cytometer. Mouse FITC-IgG₁ lacking anti-CD11b activity was used as the isotype control. Fluorescence was quantified within the granulocyte gates, defined by forward and sideward light scatter, and the mean channel fluorescence ratio was calculated as a measure of Mac-1 expression.

In Vivo Assays for LTB₄ Function

Neutropenia.

LTB₄-induced neutropenia was assessed after substance administration in monkeys, rats, and guinea pigs. The technique was essentially similar in all species. Two catheters were laid, through one of which LTB₄ in saline/2% ethanol was injected, from the other of which blood was withdrawn into EDTA-Microvettes (Sarstedt, Germany) immediately prior and 30 s after LTB₄ administration. LTB₄-induced neutropenia was assessed immediately before administration of drugs by oral gavage and at the stated time points thereafter. Neutrophils of guinea pigs were estimated manually by counting total cells and estimating the proportion of neutrophils from cytocentrifuge preparations stained with May-Grunwald-Giemsa. Neutrophils in rats and monkeys were estimated automatically using a Technicon H.1E cell counter together with multispecies software (Miles Diagnostics version 3.0). This not only counted the cells but also differentiated them according to light scattering characteristics and peroxidase staining.

Rat neutropenia.

Catheters were laid in the left carotid artery and jugular veins under anesthesia with 100 mg/kg ketamine hydrochloride (Ketavet; Parke Davis, Berlin, Germany) and 4 mg/kg xylazine (Rompun; Bayer, Leverkusen, Germany). LTB₄ was administered as a bolus at a dose of 1 μg/kg in a volume of 1.0 ml/kg.

Guinea pig neutropenia.

As for rats, except that the LTB₄ dose was 0.1 μg/kg. Animals were only used if the proportion of neutrophils in the baseline blood sample lay between 20 and 60% of total leukocytes.

Monkey neutropenia. Catheters were laid in the left and right cubital veins under a short-acting anesthetic (ketamine hydrochloride 7 mg/kg). This anesthesia was repeated for each LTB₄ challenge. LTB₄ was administered as a bolus at a dose of 30 ng/kg in a volume 0.1 ml/kg.

For statistical analysis, the neutrophils measured 60 s after administration of LTB₄ were expressed as percentage of reduction related to the neutrophils measured immediately before LTB₄ administration. A percentage of inhibition, INH_i, of this reduction at each dose and time point was calculated individually for each animal. To calculate the ED₅₀ value with 95% confidence interval for the results with rats and guinea pigs the logarithms of dose and the inhibition values were fitted to a sigmoidal curve using the method of least squares and the following equation:INHi=100−1001+10SL×(LDOSEi−LED) where LDOSE_i is logarithm of the dose administered to animal i, LED is the logarithm of the ED₅₀, and SL is the slope parameter The control groups were included using 10 pg/kg as a value for dose.

In the monkey neutropenia studies, only three doses of CGS25019 C or LY 293111 were tested, and here the logarithms of dose and the inhibition values were fitted by simple linear regression for calculation of ED₅₀. Seven dose levels of BIIL 284 were available, but for consistency the same method of calculating the ED₅₀ was applied using only those doses that described the linear portion of the dose-response curve.

To calculate duration of action of drug (e.g., data from Fig. 7) the percentage of inhibition INH_i was fitted to an exponential function beginning with the time that caused the maximum inhibition. The method of least squares and the following equation were used:INHi=c×e−b×time where b and c are parameters of the exponential function, and time is time between administration of drug and administration of the challenge dose of LTB₄.

• Download figure
• Open in new tab
• Download powerpoint

Figure 7

Time course of the effect of BIIL 315 ZW on LTB4-induced neutropenia in guinea pigs. Data between 16 and 144 h after intravenous administration of BIIL 315 (0.03 mg/kg) are fitted to an exponential function. Each point represents the mean, with standard error of the mean, of measurements on six individual guinea pigs.

Ex Vivo LTB₄-Induced PMNL CD11b (Mac-1) Expression.

BIIL 284 was given orally to eight fasted male monkeys at doses of 0.01, 0.03, 0.1, and 0.3 mg/kg. A washout time of minimally 14 days was allowed between the individual doses, which were given according to a randomization scheme. All eight monkeys received all four doses and vehicle as control. Venous blood samples were taken into EDTA-Monovettes before, and 0.5, 1, 2, 3, 5, 7, and 24 h after oral dosing by gavage of fasted rhesus monkeys. Duplicate 90-μl aliquots of fresh blood were incubated 20 min at 37°C with 10 μl of LTB₄ (final concentration 40 nM) or vehicle, and then subsequently incubated 30 min at 4°C with saturating concentration of an FITC-labeled monoclonal mouse antibody (Bear-1; Coulter), which reacted with both human and monkey CD11b antigens. Mouse IgG1-FITC was used as isotype control. Subsequent processing was as described above.

Neutrophil Migration into Mouse Ear.

Adult female HsdWin:NMRI mice received compound or vehicle control p.o. 30 min before challenge with 250 ng (5 μl) of LTB₄ applied in acetone to each side of the left ear under light anesthesia. The animals were killed with ether 6 h later, and a biopsy (diameter 8 mm) was punched from both the left ear and the right (untreated) ear. Acetone treatment alone had no effect. To assess the increase of PMNL in the left ear compared with the contralateral ear, tissue samples were homogenized in 1 ml of 0.5% hexadecyl-trimethyl-ammonium-bromide dissolved in 0.05 M phosphate buffer, pH 6.0, using a tissue homogenizer (IKA-Ultraturrax T5; Janke & Kunkel, Staufen/Breisgau, Germany) at 30,000 rpm for 15 s. After centrifugation (16,000g, 5 min) the supernatants were frozen until processing for spectrophotometric myeloperoxidase determination (see above). The dose groups were compared with the corresponding control groups in a Wilcoxon test calculating exact p values by means of a permutation test procedure. An inhibition of LTB₄-induced accumulation of neutrophils related to the concurrent control group was defined as follows:INHi=100×(YC−Yi)/(YC) where Yi is accumulation of neutrophils (myeloperoxidase) of animal i, and YC is mean accumulation of neutrophils of the concurrent control group.

To calculate the ED₅₀ value with 95% confidence interval for accumulation of neutrophils into the ear the logarithms of dose and the inhibition values were fitted to a sigmoidal curve using the method of least squares and an equation analogous to that described in the section above dealing with the neutropenia studies.

Statistics

All statistical analyses was performed with the software product SAS (SAS Institute, Cary, NC), version 6.08.

Results

In Vitro Models

Effect of BIIL 284 and Metabolites on LTB₄ Binding to Human LTB₄ Receptors.

Due to its prodrug character, BIIL 284 does not bind to the LTB₄ receptor with high affinity, as shown on freshly isolated human granulocytes and membranes thereof.

BIIL 260, the active metabolite formed from the prodrug BIIL 284 and its glucuronide conjugate BIIL 315, however, bind to the LTB₄ receptor on human granulocytes or cell membranes thereof (Table 1) with high affinity in monophasic displacement curves. TheirK_i values of 1.7 and 1.9 nmol/l are similar to the affinity of LTB₄, which also displaced [³H]LTB₄ in monophasic displacement curves from LTB₄receptors on vital PMNLs with K_i = 2.1 ± 0.93 nmol/l and K_i = 1.1 ± 0.11 nmol/l on PMNL membranes (Fig.2). The receptor binding system used thus reflects interactions with the high-affinity sites of the LTB₄ receptors. For monocytes (human monocytic U937 cells), similar K_i values were calculated for BIIL 260 and BIIL 315 in LTB₄binding assays compared with human PMNLs (Table2). Scatchard plots (Fig.3) of binding of [³H]LTB₄ to human PMNL receptors in the presence of no, 5, or 20 nmol/l BIIL 260 or 1 or 5 nmol/l BIIL 315 were essentially linear and calculated values ofB_max were essentially similar at each of the drug concentrations tested, whereas apparentK_D increased with increasing drug concentration (Table 3). From the combined results from [³H]LTB₄ displacement and saturation experiments, it can be concluded that the interaction of the BIIL 284 active metabolites BIIL 260 and BIIL 315 with the LTB₄ receptor is saturable, reversible, and competitive.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Inhibition of [³H]LTB₄binding to human PMNL membranes by BIIL 284, its metabolites, and LTB₄. Competition studies were performed with 0.3 nmol/l [³H]LTB₄ in the presence and absence of various concentrations of the substances indicated. Incubation at 0°C for 2 h was terminated by filtration. Specific binding was defined as total binding minus nonspecific binding determined in the presence of 0.1 μmol/l LTB_4. Data are means with standard error of the mean. Each assay was performed in triplicate and the assays were repeated at least three times.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

[³H]LTB₄ saturation curve and Scatchard plots of [³H]LTB₄ binding to human PMNL membranes. Scatchard plots of [³H]LTB₄ binding to human PMNL receptors in the absence or presence of BIIL 260 (5 or 20 nmol/l) or BIIL 315 (1 or 5 nmol/l). PMNL membranes were incubated with increasing concentrations of [³H]LTB₄ (0.01–5 nmol/l) for 2 h. Data shown are representative of four experiments each performed in triplicates. Saturation curves were fitted to a one-site model using the EASY FIT software. Scatchard plots were fitted with GraphPad Prism software.

Receptor binding assays using 27 different human non-LTB₄ receptors and their specific ligands for determining specificity of the active metabolites were performed. No relevant binding was found for any of the receptors determined (data not shown).

Species Differences in LTB₄ Receptor Binding of BIIL 260 and Reference Compounds.

Due to the prodrug character of BIIL 284, i.e., the masking of the benzamidine moiety, no high-affinity binding of BIIL 284 to the LTB₄ receptor was measured. BIIL 260 was particularly active at the human LTB₄ receptor (on PMNLs,K_i = 1.7 ± 0.72 nmol/l; on U937 cells, K_i = 1.2 ± 0.33 nmol/l) and the monkey LTB₄ receptor (on PMNLs,K_i = 1.8 ± 0.33 nmol/l, Table2). BIIL 260 displayed considerably lower affinity to LTB₄ receptors on rat PMNLs (K_i = 6.0 + 3.4 nmol/l) and even more so in the guinea pig (K_i = 25 + 5.7 nmol/l) and the dog, K_i = 44 + 18 nmol/l). Essentially the same pattern was found for its glucuronide conjugate BIIL 315. The benzamidine compound CGS 25019 C (not a prodrug) as well as LY 293111 displayed a comparable pattern of higher affinities to human and monkey receptors and reduced affinities to guinea pig and dog LTB₄ receptors. The rank order of affinities human ≈ monkey ≥ rat ≫ guinea pig > dog was similar for BIIL 260 and BIIL 315.

Effect of BIIL 284 and Metabolites on LTB₄-Induced Elevation of Intracellular Calcium Levels in Human PMNLs (Fura-2).

For a functional evaluation of the LTB₄inhibitory potency of BIIL 284 BS metabolites in vitro, LTB₄-induced Ca²⁺transients of vital human PMNLs were studied.

Intracellular Ca²⁺ increase is an early and sensitive hallmark of cellular activation. Ca²⁺transients were dose dependently induced by LTB₄in PMNLs. Maximal Ca²⁺ responses at 0.3 μmol/l LTB₄ were dose dependently inhibited by BIIL 284 metabolites. BIIL 260 and its glucuronide BIIL 315 inhibited LTB₄-induced intracellular Ca²⁺ release in human neutrophils with IC₅₀ values of 0.82 and 0.75 nmol/l, respectively. No agonistic activity of either BIIL 260 or BIIL 315 was found on intracellular Ca²⁺ levels. In the presence of 0.1% BSA, the inhibitory potency of BIIL 260 and BIIL 315 was reduced 6- to 8-fold due to protein binding. Similar results were obtained for the reference LTB₄ antagonists CGS 25019 C and LY 293111 (Table 4). LY 293111 appears to lose more potency by protein binding (factor of about 9) than CGS 25019 C (factor of about 2). Functional inhibition of LTB₄-induced cell activation in the presence of protein is thus achieved in the same concentrations as those found inhibitory for LTB₄receptor binding.

LTB₄-Induced Chemotaxis of Human PMNLs.

BIIL 260 and BIIL 315 potently inhibited LTB₄-induced chemotaxis of human PMNLs with IC₅₀ values of 2.90 and 0.65 nmol/l. BIIL 260, BIIL 315, and CGS 25019 C (IC₅₀ of 1 nmol/l) were about equipotent, whereas LY 293111 (66.2 nmol/l) was about 50-fold less potent (Table 5).

Functional inhibition of LTB₄-induced cell activation monitored by intracellular Ca²⁺release and inhibition of LTB₄-induced chemotaxis were achieved by comparable concentrations of BIIL 260, BIIL 315, and CGS 25019 C. LY 293111 appears to loose more functional potency by protein binding.

In Vivo Models

Inhibition of LTB₄-Induced Neutropenia.

BIIL 284 dosed orally as well as intravenously was effective in inhibiting LTB₄-induced neutropenia in all species tested (rat, guinea pig, monkey). Maximum inhibition was observed between 5 and 7 h after oral administration in monkeys and guinea pigs, and between 3 and 5 h in rats. The compound was extremely potent in all species tested. In rats, the ED₅₀ 3 h after oral was 0.019 mg/kg, with 95% confidence limits 0.009 to 0.041 mg/kg. ED₅₀ values at this time point of the literature reference substances CGS 25019 C and LY 293111 were 1.7 and 12.2 mg/kg, respectively. In rhesus monkeys, the ED₅₀ 7 h after dosage was 0.0041 mg/kg, with 95% confidence limits 0.0027 to 0.0068 mg/kg. ED₅₀ values at this time point of the literature reference substances CGS 25019 C and LY 293111 were 3.4 and 5.0 mg/kg, respectively. In monkeys, even at the first measurement time point (after 1 h) BIIL 284 had an ED₅₀ of 0.0119 mg/kg, with confidence limits 0.0082–0.0197 mg/kg (reference LY 293111 ED₅₀ at this time point 3.0 mg/kg). Figure4 compares BIIL 284 dosed at 0.03 mg/kg and reference substance CGS 25019 C and LY 293111, both dosed at 3 mg/kg p.o. over a series of time periods.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Inhibition of LTB₄-induced neutropenia by oral BIIL 284 in comparison to the reference compounds CGS 25019 C and LY 293111 in monkeys. Each column is the mean, with standard error of the mean, of measurements on four monkeys that were repeatedly challenged with LTB₄ at the time points specified after oral dosage with drug (0.3 mg/kg).

In both monkeys (Fig. 4) and rats (Fig.5), a dose of 0.3 mg/kg BIIL 284 BS p.o. achieved an inhibition greater than 50% between 1 and 7 h. An exponential function was fitted to the values from rats, measured at and after the time point (3 h) of maximum inhibition of neutropenia. This analysis gave a time of 8.0 h (95% confidence limits 6.8–9.3 h) for the degree of inhibition to be halved.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5

Inhibition of LTB₄-induced neutropenia by oral BIIL 284 in rats. Each column is the mean, with standard error of the mean, of measurements on at least six different rats, each rat receiving LTB₄ only once, at the time point specified after oral dosage with drug (0.3 mg/kg).

In guinea pigs, neutropenia was 50% inhibited between 5 and 16 h after dosage at 0.5 mg/kg p.o., but a slightly slower time of onset was observed. Figure 6 compares the duration of action of BIIL 284 dosed at 0.5 mg/kg and of reference substances CGS 25019 C and LY 293111, both dosed at 1 mg/kg p.o. Oral BIIL 284 protected against LTB₄-induced neutropenia for 16 h, whereas both reference substances had a duration of protective activity for no longer than 3 h. The active metabolite of BIIL 284, BIIL 315 (formed by removal of the ethoxycarbonyl protecting group and glucuronidation), was also tested in guinea pigs and was also highly active and long-acting. The ED₅₀ after intravenous injection (2 min before LTB₄) was 0.0095 mg/kg (95% confidence limits 0.0063–0.0143 mg/kg). A dose of 0.03 mg/kg i.v. was used to determine the duration of action of this metabolite. An exponential function was fitted, beginning at the last time at which inhibition was maximal (16 h postadministration) and ending at 145 h (Fig.7). The fall from total to 50% inhibition of neutropenia required 55 h.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6

Inhibition of LTB₄-induced neutropenia by oral BIIL 284 (0.5 mg/kg p.o.; ▴) in comparison to the reference compounds CGS 25019 C (1 mg/kg p.o.; ▪) and LY 293111 (1 mg/kg p.o.; ●) in guinea pigs. Each point is the mean, with standard error of the mean, of measurements on at least six guinea pigs, each receiving LTB₄ only once, at the time point specified after oral dosage with drug.

Inhibition of LTB₄-Induced Accumulation of Neutrophils in the Mouse Ear.

BIIL 284 and CGS 25019 C were able to inhibit LTB₄-induced mouse ear inflammation dose dependently after p.o. administration. The calculated ED₅₀ values were 0.0082 (0.0067–0.010) mg/kg p.o. for BIIL 284 and 5.3 (4.29–6.54) mg/kg p.o. for CGS 25019 C. LY 293111 showed no dose-dependent effect, however, doses of 5 and 10 mg/kg p.o. demonstrated significant (p = 0.007 andp = 0.019, respectively) anti-inflammatory effects in this model.

In Vivo Inhibition of LTB₄-Induced Inhibition of Mac-1 Expression on ex Vivo Granulocytes.

Orally dosed BIIL 284 was highly effective in inhibiting LTB₄-induced Mac-1 expression on ex vivo neutrophils in monkeys. Maximum inhibition was observed between 5 and 7 h after oral administration. An ED₅₀ of 0.05 mg/kg was calculated in monkeys (Fig. 8). A dose of 0.3 mg/kg p.o. was used to determine the duration of action of BIIL 284 and found to completely inhibit LTB₄-induced Mac-1 expression over 24 h (Fig. 9).

• Download figure
• Open in new tab
• Download powerpoint

Figure 8

Inhibition of ex vivo LTB₄-induced Mac-1 expression by oral BIIL 284 in monkeys—dose response. Each point is the mean, with standard error of the mean, of measurements from eight different monkeys, each having blood drawn and ex vivo stimulated by addition of LTB₄ at 5 h after oral dosage with drug. A washout time of minimally 14 days was allowed between the individual doses, which were given according to a randomization scheme. All eight monkeys received all four doses and vehicle as control.

• Download figure
• Open in new tab
• Download powerpoint

Figure 9

Inhibition of ex vivo LTB₄-induced Mac-1 expression by oral BIIL 284 in monkeys as a function of time. Each point is the mean, with standard error of the mean, of measurements from eight different monkeys, each having repeatedly blood drawn and ex vivo stimulated by addition of LTB₄ at the time point specified after oral dosage with drug (0.3 mg/kg p.o.).

Discussion

BIIL 284, its metabolite BIIL 260 (formed by removal of the ethoxycarbonyl protecting group), and its major metabolite BIIL 315 (formed by removal of the protecting group and glucuronidation) had potent in vitro and in vivo LTB₄ antagonistic properties. Due to the prodrug character of BIIL 284, i.e., the masking of the benzamidine moiety, no high-affinity binding of BIIL 284 to the LTB₄ receptor was measured, but both metabolites had high affinity (K_i less than 2 nM) to the LTB₄ receptor on U937 cells or on isolated human neutrophils or membranes thereof. The reference compound CGS 25019 C, also a benzamidine, had an affinity for the human LTB₄ receptor of the same order as that of the BIIL 284 metabolites, whereas the affinity of LY 293111 (5.5 nM on neutrophilic granulocytes, 24 nM on U937 cells) was somewhat less. Our calculated affinity constants of LY 293111 for the human LTB₄ receptor are of the same order as the value reported in the literature (25 nM, Marder et al., 1995). The high affinity of the glucuronide BIIL 315 for the LTB₄receptor was surprising in view of the bulkiness of the carbohydrate group, and the fact that many other glucuronides have considerably less affinity for their receptors than the parent substance. It supports drug-receptor models in which the 4-hydroxyphenyl moiety of the benzamidine is held away from the receptor binding site.

For in depth analysis we used neutrophil membranes as source of human LTB₄ receptors. BIIL 260 and BIIL 315 (like LTB₄ itself) displaced [³H]LTB₄ in monophasic displacement curves. Experimental points were fitted well by the law of mass action-based program EASY FIT (Schittkowski, 1994). Fitting was optimal using a one receptor/two ligand model, which accords with interactions with a single high-affinity site of the LTB₄ receptor. [³H]LTB₄ saturation experiments without and in the presence of BIIL 260 or BIIL 315 directly demonstrated the saturable, reversible and competitive nature of their interaction with the LTB₄ receptor. Both LTB₄ receptor concentrations and apparentK_D values calculated with EASY FIT compared well with results from graphical analysis according toScatchard (1949). Scatchard analysis revealed LTB₄ receptor concentrations (B_max values) essentially unchanged in the presence of two concentrations of either BIIL 260 and BIIL 315. However, due to reduced binding levels and consequently increasing variability of low radioactivity measures in the presence of high concentrations of antagonist an appreciable variability was found in the B_max concentrations determined. In contrast, increasing values of the apparentK_D values for [³H]LTB₄ were calculated in the presence of BIIL 260 and BIIL 315. These findings indicate a competitive interaction of the ligands LTB₄, BIIL 260, and BIIL 315 with the human LTB₄ receptor.

In further studies (not detailed in this article) no evidence was obtained of relevant binding to any other receptor apart from the LTB₄ receptor. From the combined binding results, the specific, reversible, and competitive nature of the interaction of the BIIL 284 active metabolites BIIL 260 and BIIL 315 with the LTB₄ receptor can be concluded.

Interpretation of comparisons of antagonist potencies from in vivo models in different species requires characterization of LTB₄ receptor binding properties in these species. The rank order of affinities, i.e., human ≈ monkey ≥ rat ≫ guinea pig > dog was similar for the benzamidines BIIL 260, BIIL 315, and CGS 25019 C, and the carboxylic acid derivative LY 293111. We are cautious in extrapolating from species comparisons of affinity to species differences in the in vivo situation because in addition to LTB₄ receptor affinity, LTB₄ receptor kinetic properties (i.e., receptor off-dissociation rates), and species-specific pharmacokinetic properties are major determinants of efficacy in vivo. However, the high affinity of BIIL 260 and BIIL 315 at the human receptor encourages us to expect even higher potency of the LTB₄antagonists studied in humans, when potency is projected from pharmacological models in guinea pigs, rats, or dogs. About similar potency can be expected from results in monkey models.

One functional consequence of LTB₄ receptor binding is the immediate induction of intracellular calcium transients. Intracellular Ca²⁺ increase is an early and sensitive hallmark of cellular activation (Richter et al., 1990). BIIL 260 and its glucuronide BIIL 315 dose dependently inhibited LTB₄-induced intracellular Ca²⁺ release in human neutrophils. Maximal Ca²⁺ responses at 0.3 μmol/l LTB₄ were inhibited with IC₅₀ values of 0.82 and 0.75 nM, respectively. In the presence of 0.1% bovine serum albumin, the inhibitory potency of BIIL 260 and BIIL 315 was reduced 6- to 8-fold due to protein binding (BIIL 260 Cl, IC₅₀ = 6.6 nM; BIIL 315 ZW, IC₅₀ = 4.3 nM). No agonistic activity of either BIIL 260 or BIIL 315 in intracellular Ca²⁺ levels could be found. Of the reference molecules LY 293111 appears to loose more potency by protein binding (factor of about 9) than CGS 25019 C (factor of about 2). Under these conditions, which may reflect more the physiological situation of PMNLs in blood, inhibition of cell activation is achieved at comparable concentrations to those found inhibitory for LTB₄ receptor binding. The IC₅₀ for inhibition of LTB₄-induced calcium mobilization in human neutrophils by reference substance LY 293111 in the presence of BSA, as reported here (12 nM) is not very different from the value (20 nM) reported by Marder et al. (1995).

For a functional evaluation of the LTB₄inhibitory potency of BIIL 284 active metabolites in vitro, chemotaxis of vital human PMNL was evaluated. BIIL 260 and BIIL 315 potently inhibited LTB₄-induced chemotaxis with IC₅₀ values of 2.90 and 0.65 nM. BIIL 260 and CGS 25019 C were equipotent to BIIL 315 under the test conditions with albumin, whereas LY 293111 was about 100-fold less potent.

We are cautious in comparing drug concentrations required to inhibit LTB₄ effects in binding studies, in functional studies in tissue culture, and in whole blood, not least because of the importance of protein binding in influencing the drug concentrations required. However, LTB₄ receptor binding, inhibition of LTB₄-induced cell activation as monitored by intracellular Ca²⁺ release, and functional inhibition of LTB₄-induced chemotaxis do seem to be achieved at comparable concentrations of BIIL 260 and BIIL 315. This indicates an LTB₄receptor-mediated mechanism of activity for the active principle of the prodrug BIIL 284, i.e., BIIL 260 and its glucuronide BIIL 315.

In addition to our in vitro studies, we examined the in vivo activity of BIIL 284 in several species. LTB₄-induced activation of neutrophils increases their stickiness to endothelial cells by expression of adhesion molecules such as CD11b. The inhibition of this transient neutropenia after intravenous LTB₄ injection served as a measure of LTB₄ antagonism. BIIL 284 dosed orally as well as intravenously was effective in inhibiting LTB₄-induced neutropenia in all species tested (rat, guinea pig, and monkey). Maximum inhibition was observed between 5 and 7 h after oral administration of solubilized BIIL 284. The compound was extremely potent in all species tested with ED₅₀ of 0.019 mg/kg in rat and 0.0041 mg/kg in rhesus monkeys. ED₅₀ values (3 h postdosing) of the literature reference substances CGS 25019 C and LY 293111 were 1.7 and 12.2 mg/kg, respectively, in rats and 3.4 and 5.0 mg/kg, respectively, in monkeys. Our values for the doses of reference compounds required to inhibit neutropenia confirm those reported in the literature. Thus, CGS 251019 C given 3 h before LTB₄ challenge was reported by Raychaudhuri et al. (1995) to inhibit LTB₄-induced neutropenia in rats with an ED₅₀ of 2 mg/kg. These workers also reported ED₅₀ values for inhibition of arachidonic acid-induced mouse ear neutrophil influx (myeloperoxidase) by CGS 251019 C. The reported values of 1.2 mg/kg for the early (1.5-h) inflammation and 7.7 mg/kg for the 18-h inflammation compare with an ED₅₀ measured here of 5.3 mg/kg (measured 6 h after LTB₄ administration). The in vivo potency of BIIL 284 compared with the reference substances was found to be unexpectedly high, considering their relative in vitro potency. BIIL 315 was also highly active in vivo. The ED₅₀ of BIIL 315 in guinea pig neutropenia after intravenous injection was 0.0095 mg/kg. The duration of action of this metabolite was very long, requiring 55 h for the fall from total to 50% inhibition at a dose of 0.03 mg/kg i.v. This very long duration of action was also found for orally administered BIIL 284. In guinea pigs, BIIL 284 dosed at 0.5 mg/kg protected against LTB₄-induced neutropenia for 16 h, whereas both reference substances dosed at 1 mg/kg p.o. had a duration of 50% protective activity for no longer than 3 h. It is likely that it is the long half-life of the metabolite that explains the long duration of action of BIIL 284. In monkeys, BIIL 284 dosed at 0.3 mg/kg completely protected for 7 h, whereas CGS 25019 C, dosed at 3 mg/kg p.o., had fallen to 50% inhibition after 6 h. LY 293111, also dosed at 3 mg/kg p.o., inhibited neutropenia maximally about by 50% in this model at this time. When the dose of LY 293111 was raised to 10 mg/kg inhibition was virtually complete over a period of 5 h. LY 293111 (10 mg/kg) was the dose reported by Allen et al. (1996) to significantly block bronchoalveolar neutrophilia in rhesus monkey produced by LTB₄ inhalation.

Mac-1 inhibition on ex vivo LTB₄-stimulated PMNLs confirmed the high potency and long duration of action found in neutropenia models. Maximum inhibition was observed between 5 and 7 h after oral administration of BIIL 284 in Tylose suspension with an ED₅₀ of 0.05 mg/kg. A single oral dose of 0.3 mg/kg p.o. completely inhibited LTB₄-induced Mac-1 expression over 24 h.

Inhibition of ex vivo Mac-1 expression on PMNLs has also been used in clinical trials as surrogate marker for CGS 25019 C (Morgan et al., 1995) and LY 293111 (Marder et al., 1996). It will be useful in clinical trials of BIIL 284. We believe that with BIIL 284 we have available a specific, potent, and long-acting LTB₄ antagonist suitable for a clinical proof of concept in a variety of inflammatory diseases, including chronic obstructive pulmonary disease, arthritis, and cystic fibrosis.