Inhibiting Inflammation in Rheumatoid Arthritis

http://www.100md.com 《新英格兰医学杂志》

     Targeting signal-transduction molecules, especially kinases, is a daunting task. First, these enzymes are often widely distributed in many tissues and contribute to a vast array of cellular processes essential for survival. Second, truly selective inhibitors can be difficult to synthesize because the target enzyme is often structurally similar to others and many such enzymes share substrates such as ATP. Two recent reports by a single group show how one can overcome these hurdles through a combination of serendipity and inventiveness: the researchers used a novel inhibitor of phosphatidylinositol-3'-kinase {gamma}

    (PI3K{gamma}) to treat mouse models of the inflammatory diseases rheumatoid arthritis¹ and systemic lupus erythematosus.²

    The PI3Ks have long been considered attractive candidates for therapeutic intervention. By catalyzing phosphorylation of the 3-hydroxy position of inositol in the phosphoinositide lipids, they regulate diverse cellular functions such as glucose transport, cell survival and proliferation, cytoskeletal rearrangement, and the production of oxygen radicals.³ These heterodimers are divided into three groups on the basis of their subunit composition. The PI3Ks composed of {alpha}

    and {beta}

    subunits are ubiquitous and are activated by many extracellular stimuli and cellular stresses.

    Here is the serendipity: the expression of PI3K{gamma}

    is far more restricted; it is found primarily in hematopoietic cells. Even more fortuitously, the stimuli that activate PI3K{gamma}

    are comparatively few and include chemokines (cytokines that mobilize and activate leukocytes) and other chemoattractant proteins, such as the C5a fragment of complement, that signal through G-protein–coupled receptors. PI3K{gamma}

    is an attractive therapeutic target for immune and inflammatory diseases because it is expressed mainly in leukocytes and regulates the transport of cells into inflammatory sites after activation of chemokine receptors. Because many chemokine and chemoattractant receptors signal through PI3K{gamma}

    , one can potentially suppress numerous overlapping pathways at once.

    The validity of this concept was supported by studies of a mouse model of rheumatoid arthritis. Mice lacking the PI3K{gamma}

    gene, as compared with wild-type mice, had significantly less synovial inflammation and joint destruction, on the basis of both the clinical extent of arthritis and a histologic evaluation of the joint. This particular model relies on the ability of systemically administered antibodies against collagen to bind cartilage and activate complement locally and is independent of the activity of T and B cells.⁴ The release of chemotactic factors such as C5a in the joint after complement fixation and activation of Fc receptors on resident cells recruits circulating blood leukocytes to the joint and leads to synovitis (Figure 1A).

    Next comes inventiveness: using rational drug design, Camps and colleagues developed novel PI3K{gamma}

    inhibitors with a degree of selectivity, as determined by in vitro assays using monocytes and macrophages.¹ One compound completely arrested disease progression in the standard, collagen-induced model of arthritis (Figure 1B). This model differs from the one used in the PI3K{gamma}

    -knockout mice because the mice were immunized with type II collagen rather than given premade antibodies against type II collagen. Therefore, it tests the effect of a therapeutic agent on the entire immune response, from initial antigen recognition to the late effects of autoantibody-induced damage to tissues. Because the drug was administered after the onset of clinical arthritis, the primary effect was probably on the ablation of signaling induced by chemokines, rather than the activation of T and B cells, which occurs very early in the model and is responsible for the production of autoantibody.

    Although this study design does not permit one to differentiate the relative contributions of innate and adaptive immunity, it is closer to the real-world clinical situation in which patients with established rheumatoid arthritis need to be treated. Administration of the PI3K{gamma}

    inhibitor also improved survival in the MRL–lpr strain, a mouse model of lupus, and longevity in the treated mice rivaled that among animals given a potent glucocorticoid. Although perhaps not predictive of results in patients,⁵ these models provide preclinical proof-of-concept that selective inhibition of PI3K{gamma}

    may suppress inflammation while leaving the other, more widely distributed isoforms untouched. This approach would theoretically have a reduced risk of adverse effects.

    All these findings portend well for the eventual implementation of a strategy involving inhibitors of pivotal kinases in clinical practice. Several caveats must be kept in mind. The beneficial effect of the complete loss of PI3K{gamma}

    in the knockout-mouse model was relatively small, as compared with the effect of the small-molecule PI3K{gamma}

    inhibitor, in the standard collagen-induced model of arthritis. This could be due to an effect on T and B cells in the latter model. In addition, the compound has partial rather than complete selectivity over other PI3K isoforms and could potentially inhibit these enzymes at high doses. The long-term safety of this approach has yet to be established; the duration of treatment even in the mouse models of chronic disease was only a fraction of the time needed to assess the effect on host defenses and homeostasis. It is also uncertain whether PI3K{gamma}

    blockade will have unanticipated adverse effects if the protein is expressed at low levels in other nonhematopoietic tissues. Despite these questions, the studies represent a paradigm for modern drug development. Molecular biology and biochemistry established the hierarchy of signaling molecules; rational drug design produced an inhibitor of a pivotal molecule in the hierarchy. However, we must not lose sight of the fact that mouse models only imperfectly mimic the human condition. Clinical trials will be the ultimate test of the hypothesis that PI3K{gamma}

    is a therapeutic target for uniquely human diseases such as rheumatoid arthritis.

    http://www.100md.com/html/DirDu/2006/07/03/02/46/77.htm