Lipoidal anti-inflammatory prodrugs as local targeting agents.

/3-Adrenergic agonists have been used extensively and successfully in the treatment of reversible airways diseases such as asthma (Djurup, 1981). This form of therapy has proven to be particularly beneficial when the /3stimulant drug is targeted mechanically by aerosol administration directly to the defective tissue, namely the lung. Oral administration of the /3-agonist has proven less satisfactory as cardiac and metabolic side-effects have been encountered along with peripheral muscular tremor, which have limited the usefulness of this basically effective therapy. In an attempt to extend the range of diseases amenable to treatment by 8-stimulants, the targeting of such agents to the skin was investigated. As /3-agonists are known to possess anti-inflammatory and anti-proliferative properties, a therapeutic benefit in inflammatory dermatites of an acute or chronic nature, and in inflammatory proliferative diseases such as psoriasis, could be anticipated. It has been recognized for many years that cyclic AMP is the 'second messenger' involved in the transmission of the primary signal due to /3-adrenoceptor stimulation and that cyclic AMP is intimately involved in the modulation of inflammatory reactions (Bourne et al., 1974). Thus in the model in vitro of the immediate hypersensitivity reaction in the human basophil, histamine secretion is triggered by the combination of antigens with specific immunoglobulin E antibody bound to the cell membrane. /3-Adrenergic catecholamines, arachidonic acid metabolites and histamine itself, agents which stimulate the accumulation of intracellular cyclic AMP by activating adenylate cyclase, prevent antigen-induced histamine release. The human polymorphonuclear neutrophil when triggered by contact with immune complexes releases lysosomal hydrolases. This enzyme release is inhibited by agents which raise intracellular cyclic AMP levels, namely prostaglandins and /3-adrenergic agonists. A cellmediated immune response, as elicited by T-lymphocytes derived from the spleens of mice immunized with allogeneic cells, which has been shown to effect cytolysis of cells bearing the appropriate alloantigen, can be inhibted by agents which raise intracellular cyclic AMP levels. Blymphocytes, as a result of antigenic stimulation, differentiate into antibody-secreting plasma cells. /3-Adrenergic catecholamines, histamine and arachidonic acid metabolites as well as methylxanthines, dibutyryl cyclic AMP and cholera endotoxin, inhibit either the production or secretion of antibody in this humoral immune response reaction. A number of recent clinical dermatological observations supporting this general thesis include: the successful treatment of atopic dermatitis with caffeine, a phosphodiesterase inhibitor (Kaplan et al., 1976, 1978), the use of isoprenaline in psoriasis (Das et al., 1978), and the use of various agents which modulate cyclic AMP in the treatment of urticaria (Keahey & Greaves, 1980). Cyclic AMP is predominantly located in the cytoplasm of the germinative or basal cells of the epidermis. In the skin, as in other tissues, the formation of cyclic AMP from ATP is catalysed by a membrane-bound adenylate cyclase, and cyclic AMP is hydrolysed to 5'AMP by cyclic nucleotide phosphodiesterase, which may be present in multiple forms in the skin. Cyclic AMP-dependent protein kinase is thought to mediate all the effects of cyclic AMP in the epidermis and has been demonstrated to exist in different forms in monkey, adult and neonatal skin. Normal human epidermis contains approx. 0.24.4 pmol of cyclic AMP/ pg of DNA. Values for psoriatic involved and uninvolved tissues have been determined mainly by Voorhees et al., (1973), but also by Harkonen et al. (1974) and Yoshikawa et al. (1975), and suggest that there is an abnormality in the cyclic AMP system in psoriasis, with a decrease in the lesional tissue levels of cyclic AMP associated with the disease process. Treatment of the psoriatic lesions with the topical use of the phosphodiesterase inhibitor papaverine (Stawiski et al., 1975) and also with L-dihydroxyphenylalanine and fenfluramine, agents which raise intracellular cyclic AMP levels, supports the general hypothesis. Cyclic GMP is also present in the dermis in much smaller amounts, ranging from 5 to 7 fmol/pg of DNA, and shows a statistically highly significant two-fold increase in the lesional epidermis. Against this background the search for a topically active /3-adrenergic agonist for use in skin diseases was undertaken. The prodrug approach was considered, as the prototype agonist isoprenaline showed poor bioavailability, potency and duration of action when applied topically. The required agent should produce no untoward cardiovascular side-effects when applied topically and have significantly improved chemical stability, as compared with isoprenaline, in a cream or gel formulation. The /3-adrenergic agents synthesized were evaluated in the mouse-ear croton-oil-induced inflammation model (Glenn et al., 1978; Green et al., 1984, and references therein) as a primary screen. This screen was deemed suitable as the rank order of potency for the corticosteriods is identical to that found for these agents in the more definitive vasoconstriction-forearm blanching screens for topical anti-inflammatory agents. Isoprenaline was inactive in this screen; moderate activity was observed for salbutamol and high activity for clenbuterol (Table 1). Non-steriodal anti-inflammatory agents were inactive at reasonable dose levels. As a result of such testing and further chronic administration studies, two compounds, ICI 125,436 (111) and 116,258 (VII), were selected for further anti-inflammatory evaluation. The duration of the anti-inflammatory effect for these compounds was determined by preadministration of drug followed, after various time intervals, by the croton oil. In this way predosing by up to 4 h had no significant effect on the anti-inflammatory potency of salbutamol, ICI 116,258 and 125,436. In cross-ear experiments, where the drug was administered to one ear and the croton oil to the other ear, a systemic IC,, was determined. Salbutamol was found to penetrate the skin fairly rapidly with an IC,, of 297pg/ear after application to the contralateral ear, this dose being only three times that required to produce an IC,, in the standard crotonoil screen. Both ICI 125,436 and 116,258 showed higher ratios of systemic to topical anti-inflammatory doses and were therefore considered to be safer. Further evaluation of these compounds in the mouse contact sensitivity to 4-ethoxy-methylene-2-phenyl-oxazolone (oxazolone) screen (Evans et al., 1971) confirmed their anti-inflammatory activity, even though their potency in this screen, as compared with the croton-oil-induced inflammation screen, was lower. The ability of this class of compound to affect another important facet of the inflammatory process was exemplified by their activity in inhibiting the release of /3-glucuronidase, a lysosomal