Richard J. Griffiths

Impaired Invariant Chain Degradation and Antigen Presentation and Diminished Collagen-Induced Arthritis in Cathepsin S Null Mice

Cathepsins have been implicated in the degradation of proteins destined for the MHC class II processing pathway and in the proteolytic removal of invariant chain (Ii), a critical regulator of MHC class II function. Mice lacking the lysosomal cysteine proteinase cathepsin S (catS) demonstrated a profound inhibition of Ii degradation in professional APC in vivo. A marked variation in the generation of MHC class II-bound Ii fragments and presentation of exogenous proteins was observed between B cells, dendritic cells, and macrophages lacking catS. CatS-deficient mice showed diminished susceptibility to collagen-induced arthritis, suggesting a potential therapeutic target for regulation of immune responsiveness.

Post-capillary venules in the "milky spots" of the greater omentum are the major site of plasma protein and leukocyte extravasation in rodent models of peritonitis.

Intraperitoneal injection of inflammatory agents in the mouse and rat causes plasma protein and leukocyte extravasation into the peritoneal cavity. Following an intraperitoneal injection of zymosan A, the milky spots of the omentum were the only abdominal sites detected where intravenously administered Monastral Blue labeled interendothelial cell gaps responsible for plasma extravasation. In addition, when colored microspheres were intraventricularly administered to quantify blood flow, the omentum was the only abdominal organ which showed an increase in blood flow during zymosan A peritonitis. A combination of light and electron microscopy, plus measurement of myeloperoxidase activity (a marker of neutrophil accumulation) demonstrated that the omental milky spots are the major route through which leukocytes migrate into the peritoneal cavity. Identical structures in the pleura likewise are the sites of protein leakage into the pleural cavity. In contrast, selective sites of protein and cellular extravasation could not be detected in the synovial lining of the inflamed knee joint.

Analgesic activity of 2-amino-4-methylpyridine, a novel NO synthase inhibitor

Inhibitors of NO synthase (NOS) exhibit antinociceptive activity in various systems. The site of action of these compounds is believed to be in the spinal cord and the selective neuronal NOS (NOS I) inhibitor, 7-nitroindazole, possesses analgesic activity at doses which do not significantly affect the other NOS isoforms [1]. 2-Amino-4methylpyridine has recently been identified as a highly potent NOS inhibitor with preferential activity on the inducible isoform, NOS II [2]. However, since 2-amino-4methylpyridine does inhibit NOS I with moderate potency, we thought it would be of interest to determine if it possessed analgesic activity and if so, how the analgesic doses compare with those that inhibit LPS-induced nitrate production, an index of NOS II activity. Intraperitoneal injection of zymosan leads to an inflammatory response characterized by abdominal constrictions, plasma extravasation, leukocyte infiltration and eicosanoid biosynthesis [3]. However, unlike the peritoneal response to acetic acid, injection of zymosan does not cause overt cell death [3] and consequently, may be more relevant to the pathogenesis of inflammatory pain. Therefore, we chose to evaluate the effect of 2-amino-4-methylpyridine on zymosaninduced writhing. In addition, we determined the effect of 2amino-4-methylpyridine on carrageenan-induced hyperalgesia in the rat paw where the drug effects on nociceptive responses can be measured after the inflammatory response has maximally developed.

Tenidap inhibits 5-lipoxygenase product formation in vitro, but this activity is not observed in three animal models.

Abstract.Objective and Design: The effect of tenidap on the metabolism of arachidonic acid via the 5-lipoxygenase (5-LO) pathway was investigated in vitro and in vivo.¶Materials and Treatment: In vitro (cells). Arachidonic acid (AA) stimulated rat basophilic leukemia (RBL) cells; A23817 activated neutrophils (human rat, and rabbit), macrophages (rat), and blood (human). In vitro (enzyme activity). RBL-cell homogenate; purified human recombinant 5-LO. In vivo: Rat (Sprague-Dawley) models in which peritoneal leukotriene products were measured after challenge with zymosan (3 animals per group), A23187 (11 animals per group), and immune complexes (3-5 animals per group), respectively.¶Methods: 5-Hydroxyeicosatetraenoic acid (5-HETE) and dihydroxyeicosatetraenoic acids (diHETEs, including LTB4) were measured as radiolabeled products (derived from [14C]-AA) or by absorbance at 235 or 280 nm, respectively, after separation by HPLC. Radiolabeled 5-HPETE was measured by a radio-TLC analyser after separation by thin layer chromatography (TLC). Deacylation of membrane bound [14C]-AA was determined by measuring radiolabel released into the extracellular medium. 5-LO translocation from cytosol to membrane was assessed by western analysis. Rat peritoneal fluid was assayed for PGE, 6-keto-PGF1, LTE4 or LTB4 content by EIA and for TXB2 by RIA.¶Results: Tenidap suppressed 5-LO mediated product production in cultured rat basophilic leukemia (RBL-1) cells from exogenously supplied AA, and in human and rat neutrophils, and rat peritoneal macrophages stimulated with A23187 (IC50, 5–15 M). In addition, tenidap was less potent in inhibiting the release of radiolabeled AA from RBL-1 cells (IC50, 180 M), suggesting that the decrease in 5-LO derived products could not be explained by an effect on cellular mobilization of AA (i.e., phospholipase). Tenidap blocked 5-hydroxyeicosatetraenoic acid (5-HETE) production by dissociated RBL-1 cell preparations (IC50, 7 M), as well as by a 100,000 × g supernatant of 5-LO/hydroperoxidase activity, suggesting a direct effect on the 5-LO enzyme itself. In addition, tenidap impaired 5-LO translocation from cytosol to its membrane-bound docking protein (FLAP) which occurs when human neutrophils are stimulated with calcium ionophore, indicating a second mechanism for inhibiting the 5-LO pathway. Surprisingly, tenidap did not block the binding of radiolabeled MK-0591, an indole ligand of FLAP, to neutrophil membranes. Although its ability to inhibit the cyclooxygenase pathway was readily observed in whole blood and in vivo, tenidaps 5-LO blockade could not be demonstrated by ionophore stimulated human blood, nor after oral dosing in rat models in which peritoneal leukotriene products were measured after challenge with three different stimuli. The presence of extracellular proteins greatly reduced the potency of tenidap as a 5-LO inhibitor in vitro, suggesting that protein binding is responsible for loss of activity in animal models.¶Conclusions: Tenidap inhibits 5-lipoxygenase activity in vitro both directly and indirectly by interfering with its translocation from cytosol to the membrane compartment in neutrophils. A potential mechanism for the latter effect is discussed with reference to tenidaps ability to lower intracellular pH. Tenidap did not inhibit 5-LO pathway activity in three animal models.

New drugs for the treatment of inflammatory, respiratory and immunological diseases

The 10 ~ National Conference of the Inflammation Research Association concluded with what has now become the traditional finale: a symposium devoted to disclosures of clinical candidates that are in Phase I or beyond. The presentations this year covered a wide spectrum, from small molecules to antibodies and diseases that included rheumatoid arthritis, chronic obstructive pulmonary disease and transplant rejection. The session began with an overview of the development of a p38 MAP kinase inhibitor from Vertex (VX-745) given by Dr. Frank Salituro. Interestingly, this compound entered the clinic only 28 months after the start of the program. The scientific rationale for this enzyme as a drug target is based on the ability of inflammatory stimuli to activate the enzyme, and by the discovery of a series of pyridinyl imidazole compounds at SmithKlineBeecham that inhibited inflammatory cytokine production and were subsequently shown to be inhibitors of p38. The key pharmacological properties of these inhibitors are the ability to inhibit production of both IL-1 and TNF, raising the prospect of enhanced efficacy compared to biological agents that neutralize the biological activity of only one of these cytokines. Dr. Salituro described the discovery of a novel template by high-throughput screening (ICs0 of 4 pM versus p38a). A better understanding of how inhibitors interact with the ATPbinding site of the enzyme was achieved by X-ray crystallography; this facilitated the development of potent inhibitors based on the HTS lead. The clinical candidate, VX-745, inhibits p38ot with an IC50 of 10 nM but is 20-fold less potent versus the closely related family member p38fl and is inactive versus a panel of other kinases. VX-745 inhibits TNF/IL1 release from human peripheral blood monocytes with an IC50 of 50 nM and has a plasma half-life of 4.5 h in the rat. Unlike first generation p38 inhibitors, VX-745 neither inhibits cytochrome P450 isozymes nor does it induce these enzymes. It is an effective inhibitor of both the clinical signs and tissue destruction in rodent models of arthritis. Based on these properties it was selected for clinical development and has completed Phase I studies in humans. A pilot study in a small number of rheumatoid arthritis patients has also been completed and a larger Phase II study in this patient population will begin in the latter part of 2000.

British Inflammation Research Association Meeting: Metalloproteinases and Inflammatory Diseases, The Fielder Centre, Hatfield, UK, 18 April 1997

This was an extremely well attended meeting of the British Inflammation Research Association which attracted people from as far affield as Australia and the USA. The meeting was organized by Kevin Bottomley and John Nixon together with their colleagues from Roche Discovery in Welwyn Garden City. The large turn out was a tribute to the excellent program that Kevin and John had put together and was very timely in view of the recent advances in this field of research. The programme was chaired by Professor Gill Murphy (Strangeways Laboratories, Cambridge, UK) who in her opening remarks exhorted the audience to come up with stimulating questions and not just try to indulge in industrial espionage (always a challenge for those of us in the pharmaceutical industry). The audience for the most part responded to the challenge very well and Professor Murphy was always ready with a question to start things off on the right track. The programme got underway with a high-tech presentation from Dr Neera Borkakoti (Roche Discovery Welwyn, Welwyn Garden City, UK) who gave a presentation entitled ‘‘Zinc Endoproteases: A Structural Superfamily’’. This superfamily can be divided into three broad classes, the aminopeptidases, the endopeptidases and the carboxypeptidases. 3-D structures exist for several members of each class together with inhibitors bound into the active site. The endopeptidase family, to which the matrix metalloproteinases belong, are a particularly rich source of 3-D structures with 11 proteins having been published. Dr Borkakoti pointed out that proteins with minimal primary sequence homology may have a very similar 3-D structure. This was illustrated (with the aid of stereo slides) by superimposing the structure of astacin on MMP-1. The two proteins have 15% homology at the amino acid level but a very similar 3-D structure. When the conformation of inhibitors bound in the active site is compared it is clear that hydroxamate inhibitors of MMP-1 and thermolysin bind to the zinc in the active site in a very similar conformation but the detail of the other interactions that the inhibitors make with the protein differ considerably because of the architecture of the enzymes. Dr Borkakoti went on to argue that homology models can provide useful information about the 3-D structure of a new protein before a structure is available. As an example she used the structure of the snake venom protein adamalysin, which has 50% homology to TNF converting enzyme (TACE) and displayed images of the homology derived structure of TACE with a modelled hydroxamate inhibitor bound. It will be extremely interesting to learn how accurate this type of model building is when the structure of TACE is solved. The second talk was given by Dr Dave Bradshaw also from Roche Discovery Welwyn. The subject of the talk was ‘‘Cartilage Protective Agent: A novel therapy for Arthritis’’ and described the discovery and development of a collagenase inhibitor for use in rheumatoid and osteoarthritis. Cartilage protective agent (Ro 32-3555) is an hydroxamate collagenase inhibitor (Ki for MMP-1 and MMP-13 of 3 nM) with lower potency versus gelatinase and stromelysin. The compound is orally active and prevents collagen but not proteoglycan degradation in a sponge/ cartilage implant model (ED50 10 mg.kg-1 b.i.d.). CPA also prevents cartilage degradation in models of inflammatory arthritis e.g. P.acnes induced monoarthritis in rats and rabbits (illustrated with an impressive video of MRI images taken through the knee joint) and in a spontaneous model of OA, the STR/ORT mouse. It has excellent pharmacokinetic properties and is well tolerated in early phase I studies in normal volunteers. Phase II studies in RA patients are scheduled for later this year. The theme of how metalloproteinase inhibitors are developed was taken up in the next talk by Dr Alan Drummond (British Biotechnology, Cowley, UK) entitled ‘‘Matrix Metalloproteinase Inhibitor Sensitive Processes in Biology: Relevance to Cancer and Inflammatory Disease’’. The first part of Dr Drummond’s talk described the progress of the hydroxamate inhibitor marimastat for the treatment of cancer. The evidence that led to the development of a broad Inflamm. res. 46 (1997) 376–378 q Birkhäuser Verlag, Basel, 1997

Inflammation Research Association Meeting: The Puzzle Posed by Cytokines Lacking a Signal Sequence, New York Academy of Sciences, USA, 22 May 1997

The purpose of this meeting was to focus on the pathway by which cytokines that lack the characteristic signal peptide present in the majority of secreted proteins are exported from the cell. The major focus was inevitably on IL-1 (for which the organizers make no apologies!) but new data on IL-18, a cytokine that seems to have many features in common with IL-1 was also presented. The first speaker was David Chaplin (Washington University, USA) and the title of his presentation was ‘‘IL-1 b release by apoptosis – Relevance in vivo?’’ David began by describing studies on the synthesis of IL-1 in murine macrophages. The precursor forms of IL-1 a andb accumulate inside the cell after stimulation with LPS and very little of the processed forms are detected extracellularly. Agents that inhibit the passage of proteins through the endoplasmic reticulum had no effect on IL-1 release. The amount of IL-1 released correlated with the amount of LDH present in the extracellular medium. This observation prompted experiments in which cells were deliberately injured by stimuli designed to induce either necrosis or apoptosis. Cells that had undergone necrosis released IL-1 a and b as the precursor forms whereas cells undergoing apoptosis released the processed 17kDa forms of the cytokine. David then went on to describe his studies with IL-1b deficient mice. These mice have no obvious developmental defects and fail to produce IL-1 b. However, IL-1a release is normal. The sensitivity of these mice to lethality induced by LPS is similar to wild type mice, haematopoesis is normal and there are no gross defects in resistance to bacterial and viral infection. However, in models of contact sensitivity a role for IL-1b was uncovered. At low sensitizing doses of the hapten TNCB (but not oxazolone) the delayed type hypersensitivity response was reduced. Local injection of IL-1b into the skin at the time of sensitization could restore the response. Interestingly, IL-1 a was unable to duplicate this effect. The induction of MHC class II expression in the skin after application of the antigen was reduced in IL-1b deficient mice as was the amount of antigen transported to local lymph nodes. It is not clear which cell type in the skin is required for production of IL-1 b in response to antigen. Keratinocytes can make both IL-1 a and b but in vitro most of the IL-1b is in the pro-form. The mechanisms for stimulating release is not clear. Langerhans cells, the major antigen presenting cell in the skin, are also capable of producing IL-1 but again the stimuli required for release are not known. Skin sections from antigen sensitized mice were subjected to TUNEL staining to see if apoptosis was occurring during the period that IL-1 is required. Strong staining was observed in both the epidermis and dermis, particularly around sebaceous glands. The relationship between cell death and IL-1 release was developed further in the second presentation by Chris Gabel (Pfizer Central Research, Groton, CT, USA) entitled ‘‘Stimulus Coupled IL-1 Posttranslational Processing: An Ionically Regulated Mechanism’’. In human monocytes LPS induced IL-1b synthesis is uncoupled from release. In freshly isolated monocytes LPS induces the synthesis of large amounts of intracellular IL-1 b but<20% is released as mature 17kDa IL-1b. In monocytes aged in culture for several days large amounts of intracellular IL-1 b can be induced by LPS but in this case none at all is released. Addition of ATP to both cell populations causes the rapid release of up to 80% of the IL-1, the majority in the processed form. This is associated with death of the cell. A similar enhancing effect of ATP is observed in human blood and in the mouse in vivo, either after LPS injection into the peritoneal cavity or into sub-cutaneously implanted tissue chambers. In the human monocyte the effect of ATP is selective for IL-1 since both IL-6 and IL-1ra are secreted in response to LPS alone and do not require a second stimulus. In addition to ATP other means of inducing efficient processing and release of IL-1 include cytolytic T cells, certain potassium ionophores and bacterial toxins. These treatments have in common the fact that they induce depolarization and cell death. If potassium is substituted for sodium in the medium, IL-1 release is blocked. This suggests that it is not depolarization per se that initiates Inflamm. res. 46 (1997) 379–381 q Birkhäuser Verlag, Basel, 1997