Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Tumor necrosis factor-alpha (TNF-), a cytokine produced primarily by activated monocytes and macrophages, is an important mediator of immuno-inflammatory responses and inducer of other pro-inflammatory cytokines (Vassali, 1992). TNF- is produced as a 26-kDa membrane-bound pro-TNF- that undergoes a specific proteolytic cleavage between Ala-76 and Val-77 to release a 17-kDa soluble TNF- (Kriegler et al., 1988). The enzyme that is responsible for this process is TNF- convertase (TACE), a metalloproteinase closely related to the matrix metalloproteinases (MMPs; Gearing et al., 1994; McGeehan et al., 1994; Mohler et al., 1994). MMPs are a large family of Zn²⁺ endopeptidases that include 72- and 92-kDa gelatinases, collagenases, stromelysins 1-3, matrilysin, macrophage metalloelastase, and membrane-bound MMP 1-4 (Birkedal-Hansen et al., 1993). They are expressed in immuno-inflammatory conditions such as rheumatoid arthritis (RA), sepsis, and inflammatory bowel disease and are collectively capable of degrading most connective tissue macromolecules under both physiological and pathological conditions (Buchan et al., 1988; Lewis et al., 1997).

RA is a chronic, progressive inflammatory disease that causes substantial morbidity. Although the precise etiology of RA remains unknown, a great deal has been learned about the immunopathophysiology of the disease in recent years. Proinflammatory cytokines such as TNF-, interleukin (IL)-1, and IL-6 have been suggested to play an important role in the pathogenesis of the RA. In addition to promoting inflammation, these cytokines are capable of directly mediating bone and cartilage destruction, the most prominent feature of RA (Buchan et al., 1988; Feldmann and Maini, 1999). Several lines of evidence have identified TNF- as a key mediator of chronic inflammatory diseases such as RA, inflammatory bowel disease, and sepsis (Espersen et al., 1991; Tracey et al., 1993). The randomized phase II and III clinical trials using a monoclonal antibody to TNF-, infliximab, or the soluble TNF- receptor fusion protein, etanercept, significantly improved the disease activity in RA patients, thus, validating TNF- as a therapeutic target for RA (Moreland et al., 1997; Maini et al., 1998; Lorenz et al., 2000). The inhibition of TACE offers an alternative point of therapeutic intervention and the possibility of a low molecular weight, orally active agent.

Hydroxamic-acid-based MMP inhibitors have been reported to inhibit endotoxin-induced production of TNF- by inhibiting the processing enzyme, TACE (Gearing et al., 1994; Mohler et al., 1994). However, there are very limited reports concerning their selectivity for TACE and oral efficacy in preclinical and clinical models. Ro 32-7315 (Fig. 1) is among the first of a new generation of potent, hydroxamic-acid-containing TACE inhibitors with greater selectivity over the MMP family of enzymes. In addition to the potency and selectivity of this compound, a marked improvement in the in vivo biological profile compared with compounds such as BB 1101 (DiMartino et al., 1997; Barlaam et al., 1999) was observed. In this report, we describe the in vitro and in vivo biological profile of Ro 32-7315, a selective inhibitor of TACE. Furthermore, we considered the feasibility of using changes in LPS-induced TNF- release for the ex vivo evaluation of TACE inhibitor pharmacodynamics. Using this approach, Ro 32-7315 was shown to cause a significant inhibition of TNF- production ex vivo after oral dosing in healthy volunteers. The findings support the contention that this selective TACE inhibitor may have use as a potential oral treatment for inflammatory diseases.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Molecular structure of Ro 32-7315.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals

Ro 32-7315 was synthesized in the Medicinal Chemistry Department of Roche Discovery (Welwyn, UK).

Isolated Enzyme Studies

In Vitro TACE Assay. A recombinant form of TACE, which lacked the transmembrane region and cytoplasmic tail, was used in the in vitro assay. The extracellular domain of TACE was subcloned into pVL1393, and the recombinant baculovirus generated was used to infect Sf9 cells. TACE activity was partially purified from concentrated culture media by Q-Sepharose, concanavalin A Sepharose, and Superdex 75 chromatography. TACE activity was determined by measuring the production of the peptide product VRSSSRTDpa from the peptide substrate ISPLAQAVRSSSRTDpa (Dpa is N-3-(2, 4-dinitrophenyl)-L-2, 3-diaminopropionyl). The assay was carried out in 10 mM Tris-HCL (pH 8.0), 50 mM NaCl, and 2% octylglycoside and at a substrate concentration of 100 µM. After 60 min at 37°C, the reaction was stopped by adding acetic acid to a final concentration of 1 M. The peptide product was separated from the reaction mixture by reverse-phase high performance liquid chromatography, using a 28 to 70% acetonitrile gradient on a C₈ column. The absorbance of the eluate at 360 nm was measured as an index of the amount of product formed.

In Vitro MMPs Assay. The inhibitory activity of Ro 32-7315 against a number of human MMPs was determined using a methodology published by Knight et al. (1992). The IC₅₀ values for Ro 32-7315 were determined for collagenase 1 (using recombinant catalytic domain MMP-1, final assay concentration 800 pM), collagenase 2 (MMP-8, final assay concentration of 250 pM), collagenase 3 (using recombinant catalytic domain MMP-13, final assay concentration of 20 pM), stromelysin (MMP-3, final assay concentration of 1 nM), gelatinase A (using recombinant catalytic domain MMP-2, final assay concentration of 300 pM), gelatinase B (MMP-9, final assay concentration of 250 pM), metalloelastase (MMP-12, final assay concentration of 300 pM), and matrilysin 1 (MMP-7, final assay concentration of 35 pM) using the fluorogenic substrate, Mca-Pro-Leu-Gly-Leu-Dap-Ala-Arg-NH2 (2 mM; Bachem, Saffron Walden, UK).

Cell-Based Studies

LPS-Induced Cytokine Release from THP-1 Cells. THP-1 cells (5 × 10⁵/ml, American Tissue Culture Collection) in RPMI medium containing 20 mM HEPES buffer, 10% fetal calf serum, and antibiotics were aliquoted (200 µl) into 96-well plates. The cells were incubated for 30 min at 37°C, 95% humidity, and 5% CO₂ with vehicle [0.8% dimethyl sulfoxide (DMSO) final concentration; BDH, Poole, UK] or Ro 32-7315 (10⁵ to 10⁹ M final concentration) before the addition of LPS (2 µg/ml final concentration; Salmonella typhimurium; Sigma-Aldrich, St. Louis, MO). After further incubation for 3 h at 37°C (95% humidity and 5% CO₂), the plates were centrifuged (1200 rpm, 4 min, MSE 2000) to sediment the cells. Aliquots of the supernatant were taken for the estimation of TNF- and IL-8 levels by ELISAs, selective for the human cytokines (R & D Systems, Abingdon, UK). The inter- and intra-assay coefficients of variations were, respectively: TNF- assay, 6.4%, 4.8%; IL-8 assay, 7.8%, 5.9%.

LPS-Induced TNF- Release from Rat and Human Whole Blood. Rat blood collected in EDTA (1.6 mg/ml potassium EDTA; Sarstedt, Numbrecht, Germany) was preincubated with either vehicle (0.5% DMSO, final concentration) or Ro 32-7315 (10⁵ to 10⁹ M, final concentration) at 37°C for 10 min. This was followed by incubation (37°C) of the whole blood with LPS (20 µg/ml, final concentration, Escherichia coli serotype, 0111:B4, Sigma-Aldrich) for 2 h. The blood samples (200-µl aliquots) were centrifuged (4000 rpm, 2 min, Jouan A14), and the plasma was snap frozen and stored at 20°C until rat TNF- levels were determined by ELISA (R & D Systems). The inter- and intra-assay coefficients of variation for this assay were 8.2 and 9.5%, respectively.

5

9

In Vivo Studies

LPS-Induced TNF- Release in Rats. Age-matched, male rats (Wistar, Charles River Breeding Laboratories) were pretreated orally with vehicle (10% succinylated gelatin, 10 ml/kg) or Ro 32-7315 (10-60 mg/kg). Thirty minutes after pretreatment, the rats were given i.p. injections of LPS (100 µg in PBS; E. coli serotype, 0111:B4; Sigma-Aldrich). The rats were bled (via tail vein) at 0 h (pre-LPS challenge), 0.5, 1, 2, 3, and 6 h, and the blood samples were collected into EDTA (potassium EDTA). The blood samples were centrifuged (4000 rpm, 2 min, Jouan A14), and the plasma was collected and frozen at 20°C until plasma TNF- and Ro 32-7315 levels were determined.

Adjuvant-Induced Arthritis Model. Female AHH/R from rats bred in-house were caged in groups of five and provided with water and a pelleted diet ad libitum. One group of rats was nominated as control and was not dosed or given adjuvant, the other groups were injected with 0.1 ml of adjuvant into the subplantar surface of the right hind paw. Adjuvant was prepared by homogenizing Mycobacterium tuberculosis (heat killed, human strains C, DT, and PN; Central Veterinary Laboratory, Weybridge, UK) in liquid paraffin BP to a final concentration of 5 mg/ml. The groups of adjuvant-injected rats were dosed intraperitoneally (twice daily, from day 0 to 14) either with vehicle (2 ml/kg Gelofusine; AAH Pharmaceuticals, Ruislip, UK) or Ro 32-7315 at doses of 2.5, 5, 10, and 20 mg/kg. The volumes of hind paws (both left and right-hand paws) were determined by water plethysmography on days 0, 2, 5, 7, 9, 12, and 14 postadjuvant injection. The secondary swelling response (from day 9 to14) was calculated for both the injected (right) and noninjected (left) hind paw. On day 14, the last day of the experiment, the degree of swelling of the nose, ears, fore paws, noninjected hind paw, and tail was assessed visually and scored as: none = 0, mild = 1, moderate = 2, or severe = 3 as described by Birchall et al. (1994). A further analysis of the noninjected hind paw was made to determine the extent of movement of the paw by measuring the angle of flexion and extension, and the difference was expressed as joint mobility. Serial blood samples (in potassium EDTA) were taken from representative animals (n = 3/group, via the tail vein) on day 13 to obtain a detailed plasma kinetic profile for Ro 32-7315.

Ex Vivo Studies in Human Volunteers

LPS-Induced TNF- Release in Healthy Human Volunteers. In a double blind, randomized, placebo-controlled healthy volunteer study, subjects received single oral doses of either Ro 32-7315 (450 mg, n = 6) or placebo (n = 2). Blood samples (10 ml via venipuncture into heparinized vacutainers, 10 U/ml) were collected at the following times, 0-, 0.5-, 1-, 2-, 4-, 6-, and 8-h postdose administration. Additional blood samples up to 48 h were taken for detailed pharmacokinetic profiling of Ro 32-7315. The heparinized whole blood (500 µl) was incubated with LPS (2.7 µg/ml, final concentration; S. typhimurium) or PBS (Invitrogen, Carlsbad, CA) control for 2 h at 37°C. After incubation, the whole blood was mixed and centrifuged (12,500 rpm, 1 min, Jouan A14), and the plasma was snap frozen and stored at 20°C until determination of TNF- Levels by ELISA.

Plasma Drug Level Determination. Concentrations of Ro 32-7315 were measured in plasma samples (rat and human) using a specific high performance liquid chromatographic-mass spectrometric method. The proteins in plasma (20 µl) were precipitated by the addition of 80 µl of methanol containing the trideutero form of Ro 32-7315 as an internal standard. After centrifugation and dilution with water (1:1), the resulting supernatant was injected (40-100 µl) onto an isocratic chromatography system consisting of a C₁₈ column (5 µm) and precolumn with mobile phase (65% methanol containing 2.5 mM ammonium formate, pH 3.0), which was run at a flow rate of 1 ml/min. The retention times for Ro 32-7315 and the isotopic internal standard were both approximately 2 min. After reduction by postcolumn splitting (4:1), the flow was directed into the mass spectrometer (PE Sciex API3+, PerkinElmer) via a TurboIonSpray interface. The precursor and product ions of Ro 32-7315 (m/z 454.5/421.5 Da) and the internal standard, D₃-Ro 32-7315 (m/z 457.2/454.2 Da) were monitored, and unknown concentrations were determined by means of calibration standards. The assay sensitivity was at least 5 ng/ml. Mean drug levels are reported as nanograms per milliliter plasma.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   LPS-induced TNF- production in male Wistar rats after oral administration of Ro 32-7315 or vehicle. Rats were injected with LPS (i.p.) 30 min after oral dosing with Ro 32-7315 or vehicle. Blood samples were collected over the 6-h duration, plasma was separated, and the TNF- levels were determined using ELISA, as described under Materials and Methods. Data shown are mean ± S.E.M. (n = 6-9/group). Figures in parentheses are mean plasma Ro 32-7315 levels (nanograms per milliliter, at 1-h time point). *, p < 0.05, represents a significant reduction in plasma TNF- levels compared with vehicle group at the same time points, based on Student's t test. , vehicle; , 10 mg/kg; , 20 mg/kg; , 30 mg/kg; ×, 60 mg/kg.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Dose-dependent effect of Ro 32-7315 on noninjected hind paw inflammation, as assessed by changes in paw volumes (from day 9 to 14) in AHH/R rats with adjuvant arthritis Ro 32-7315 or vehicle was administered i.p. (twice daily) throughout the experiment (from day 0 to 14). The changes in paw volumes are shown as mean ± S.E.M. (n = 10-20/group). The values in parenthesis are percentage reduction in paw volume as compared with the vehicle-treated arthritic rats. *, p < 0.05, represents a significant difference compared with the vehicle group, based on Student's t test.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   LPS-induced TNF- release in whole blood derived from healthy subjects receiving Ro 32-7315 (, n = 6) or placebo (, n = 2). The TNF- levels are shown as mean. One way, repeat measures analysis of variance showed a significant within subject trend (Df 30, F = 5.03, p < 0.001). Post hoc analyses against the predose data revealed that the 0.5-, 1-, and 2-h means were significantly different; **, p < 0.01.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibition of TACE, MMPs, and LPS-Induced TNF- Release in Vitro. The inhibitory potency of Ro 32-7315 was determined against a recombinant form of TACE using a peptide substrate, ISPLAQAVRSSSRTDpa. For selectivity evaluation, Ro 32-7315 was also tested against a number of MMPs. The TNF- inhibition in cells was determined in LPS-induced THP-1 cells and in rat and human whole blood; the results are summarized in Table 1. Ro 32-7315 demonstrated some selectivity for TACE over other human MMPs (minimum of 2-fold selectivity over MMP 12, metalloelastase, to a maximum of 96-fold selectivity over MMP 1, collagenase 1; Table 1).

View this table: [in this window] [in a new window]   TABLE 1 Inhibitory potency of Ro 32-7315 (nanomolar) against recombinant TACE, selected human MMPs, and LPS-induced TNF- release in THP cells and whole blood Ro 32-7315 was serially diluted from an initial concentration of 10 mM for IC50 determination. A constant vehicle concentration (0.5% or 0.8% DMSO) was maintained in the final assay mixture.

5

8

LPS-Induced TNF- Release Studies in Wistar Rats. Figure 2 shows the time course of inhibition of TNF- release by of Ro 32-7315 dosed orally after an i.p. injection of LPS. Administration of LPS resulted in a transient release of TNF- in vivo, which peaked between 1 and 2 h, after which time the plasma TNF- levels declined to baseline (<0.5 pg/ml) by 6-h post-LPS injection. Oral pretreatment with Ro 32-7315 (20-60 mg/kg) resulted in a significant (p < 0.05, compared with vehicle control) dose-dependent reduction in LPS-induced systemic TNF- release. In the same model, the plasma drug levels were also monitored over the 6-h period (Fig. 2, in parentheses). The inhibition observed was associated with a high plasma exposure of Ro 32-7315 (maximal concentration, C_max, of 218 ng/ml and 405 ng/ml, with 30 and 60 mg/kg dose, respectively). These levels are approximately 4- and 8-fold higher, respectively, than the in vitro rat whole-blood IC₅₀.

Adjuvant-Induced Arthritis Model. Adjuvant induced two phases of inflammation in the AHH/R rats, as assessed by increases in hind paw volume. The primary swelling phase (days 0-4) occurred in the injected hind paw, and this represents an acute inflammatory response (not shown). The secondary phase (days 9-14) is a systemic immune response and affects both left (noninjected) and right paws and the nose, ears, forepaws, and tail (Birchall et al., 1994). Ro 32-7315 (administered i.p. at 2.5, 5, 10, and 20 mg/kg, twice daily) caused a significant dose-dependent inhibition of adjuvant-induced secondary paw swelling in the noninjected paws (42, 71, 83, and 93% reduction, respectively, as compared with the vehicle group, n = 10-20/group; Fig. 3). Furthermore, the reduced paw swelling was associated with significant improvements in the lesion score and degree of joint mobility in the paws (Table 2).

LPS-Induced TNF- Release in Healthy Human Volunteers. Stimulation of TNF- release from monocytes in human blood has been studied both in vivo and in vitro after LPS challenge (Mohler et al., 1994; DiMartino et al., 1997). The production of human TNF- has been documented in vivo both clinically in sepsis patients and experimentally, after healthy volunteers received systemic LPS (Tracey et al., 1993; Dekkers et al., 1999). The majority of reported in vitro studies used primary cells in culture or tumor lines (Gearing et al., 1994; DiMartino et al., 1997). Although these systems have provided valuable information, they cannot take account of the complexities that can occur in whole blood such as protein binding, compartmentalization, and possible interactions between different cell types. We have used whole blood as a system to assess the production of human TNF-. In the present study we have measured inhibition of LPS-induced TNF- release in blood taken from healthy volunteers receiving an oral dose of Ro 32-7315 (450 mg). An initial series of experiments was undertaken to study the baseline and the influence of circadian changes on LPS-induced TNF- release in whole blood taken from individuals of the same population who were not dosed with Ro 32-7315. For these studies, sequential blood samples (over 24 h) were taken from healthy volunteers (as described under Materials and Methods), and the whole blood was stimulated with LPS in vitro. These studies demonstrated that over a 24-h period, there were no significant changes in the LPS-induced TNF- release profile (mean levels of 5140 pg/ml and 6930 pg/ml over 24 h for female and male volunteers, respectively; not shown).

View larger version (9K): [in this window] [in a new window]   Fig. 5.   Correlation between drug concentration (plasma drug levels, nanograms per milliliter) and ex vivo efficacy (as assessed by whole-blood TNF- release) after oral administration of Ro 32-7315 in healthy subjects. The figure shows individual concentration-effect data pairs from six individual volunteers receiving 450 mg of Ro 32-7315 and assessed at seven time points between 0- and 8-h postdose. The data were fitted to a four parameter logistic function (r = 0.84).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study show that Ro 32-7315 is a potent inhibitor of the recombinant TACE (IC₅₀ 5.2 nM) and inhibits LPS-induced TNF- release from a monocytic cell line and in rodent and human whole blood (IC₅₀ values of 350, 110, and 2400 nM, respectively). In experimental models, Ro 32-7315 (administered orally and i.p.) significantly reduced systemic TNF- release and the adjuvant-induced immuno-inflammatory response in a dose-dependent manner. An important finding of the present study is the demonstration of oral bioavailability and the associated efficacy, as assessed by inhibition of ex vivo TNF- release, in healthy human subjects. This is the first such demonstration for a selective TACE inhibitor.

In the present study, a considerably higher concentration of Ro 32-7315 (11-500 nM range) was required to inhibit selected MMPs compared with recombinant TACE. Of the selected MMPs studied, Ro 32-7315 was most potent against metalloelastase (MMP12, IC₅₀ = 11 nM) and least active against collagenase 1 (MMP1, IC₅₀ = 500 nM). Numerous potent MMP inhibitors have been described over the past few years, the majority of which incorporate an hydroxamic acid group, as the zinc binding ligand. This approach has resulted in a number of clinical candidates such as Marimastat, BB 1101, AG 3340, and GW 947 (Gearing et al., 1994; Barlaam et al., 1999). However, these molecules tend to show broad-spectrum inhibition of MMPs (IC₅₀ values for inhibition of MMP-1 were 5, 10, and 8.2 nM, respectively; 50- to 100-fold more potent than Ro 32-7315) and TACE (IC₅₀ values against TACE are 3.8, 0.2, and 5.5 nM, respectively; Barlaam et al., 1999). Thus, Ro 32-7315 demonstrates a more selective profile for TACE than these compounds.

Bacterial LPS administration is known to induce a rapid TNF- synthesis from inflammatory cells. This process occurs through rapid up-regulation of TNF- mRNA expression (Vassali, 1992). LPS-induced TNF- release was used further to study the in vitro and in vivo biological profile and the mechanism of Ro 32-7315. In the monocytic cell lines, THP-1, and rat and human whole blood, Ro 32-7315 reduced LPS-induced TNF- release in a dose-dependent manner, with IC₅₀ values of 350, 110, and 2400 nM, respectively. In the same experiments Ro 32-7315 had no effect on the release of IL-8. LPS-induced release of TNF- and other cytokines proceeds through the activation of the transcriptional activator NF-B (Guha and Mackman, 2001). That Ro 32-7315 does not inhibit IL-8 release indicates that it does not inhibit TNF- production at the transcriptional level. Furthermore, it was evident that the high intrinsic potency of Ro 32-7315 against recombinant TACE did not translate to the THP-1 cell and whole-blood assays (21- to 461-fold reduction in potencies between the enzyme and the cellular assays). A similar profile has been reported for MMP inhibitors such as BB 1101 and GW 9471 (Gearing et al., 1994, Barlaam et al., 1999). A number of explanations for this reduction in potency in whole-blood cell system versus isolated enzyme have been advanced (Barlaam et al., 1999). It has been reported that all these agents bind extensively to plasma proteins present in the cell-based assays. For example experiments using isolated human mononuclear cells treated with LPS in the presence of human serum (1%) resulted in a 40-fold reduction in activity (Barlaam et al., 1999). Furthermore, the reduced potency in the cellular assays observed in the present study could be due to the fact that the enzyme, TACE, is located in a submembrane compartment (Schlondorff et al., 2000).

In the in vivo experimental model, Ro 32-7315 demonstrated potent inhibitory activity against LPS-induced systemic TNF- release after an oral administration (ED₅₀ of 25 mg/kg). To evaluate the anti-arthritic activity, we used an adjuvant-induced arthritis model that has been shown to be sensitive to anti-TNF- therapy, Tenefuse (TNF receptor, p55 fusion protein; M. Brewster, personal communication). Ro 32-7315 administered at doses of 2.5, 5, 10, and 20 mg/kg (i.p., twice daily from day 0 to 14) significantly reduced adjuvant-induced secondary paw swelling in a dose-dependent manner. At a dose of 20 mg/kg (twice daily), greater than 90% inhibition was observed. The reduced paw swelling was accompanied with improved joint mobility and reduced adjuvant-induced lesions. Furthermore, the efficacy demonstrated in the immuno-inflammatory model was associated with a high plasma exposure of Ro 32-7315. In comparison, hydroxamic-acid-containing inhibitors of MMP such as BB 1101, have been reported to produce anti-arthritic activity in adjuvant-induced model, however, the degree of inhibition demonstrated was modest (maximal inhibition of 40-56%; DiMartino et al., 1997). The reduced efficacy was attributed to their poor pharmacokinetic profile and the reduced potency for the TNF- processing enzyme.

The hypothesis that TNF- plays an important role in the pathogenesis of chronic inflammatory disease such as RA has been confirmed by recent reports on the clinical efficacy of monoclonal antibody, infliximab, and a fusion protein of soluble TNF receptors linked to human immunoglobulin, TNFRp75:Fc, Enbrel and TNFRp55:Fc, Lenercept (Moreland et al., 1997; Maini et al., 1998; McKay et al., 1998). These biologicals significantly reduced signs and symptoms and improved the sense of well being in a large number of RA patients after parenteral administration. The result of anti-TNF treatment on the progression of damage to cartilage, bone, and other connective tissue components in active RA patients remains to be established. Although in collagen-induced arthritis in DBA/1 mice, joint protection has been reported (Joosten et al., 1996). Moreover, the positive clinical results obtained with these parenterally administered anti-TNF- biologicals provide an impetus to develop orally effective TNF- inhibitors. The efficacy of Ro 32-7315 was further evaluated in healthy human volunteers by studying the ex vivo, LPS-induced TNF- release profile and the associated plasma drug exposure. In this study, a high exposure of Ro 32-7315 was observed in human subjects, and this was associated with a significant reduction (up to 69%) in the LPS-induced whole-blood TNF- release.

These findings suggest that selective inhibitors of TACE, such as Ro 32-7315, may provide an important oral therapy for chronic inflammatory diseases. An investigation of the effect of Ro 32-7315 on TNF- release in patients with rheumatoid arthritis would be relevant, because they may respond differently to healthy volunteers. Thus, in rheumatoid subjects, there is evidence of enhanced release of TNF- after incubation of either whole blood (Zangerle et al., 1992) or isolated monocyte preparations with LPS (Leirisalo-Repo et al., 1995). The magnitude of LPS-induced release of TNF- in whole blood is under genetic control (Louis et al., 1998), and there is evidence that certain polymorphisms of the TNF gene may underlie rheumatoid arthritis (Verweij, 1999).