Dithranol (anthralin) is one of the most widely used and effective, albeit empirical, topical treatments for patients with psoriasis (McBride et al. 2003; Swinkels et al. 2001; van de Kerkhof and Franssen 2001; van der vleuten et al. 1996). The molecular basis of its mode of action is still unknown, but it is probably related to the redox activity leading to the production of free radicals, including oxygen radicals (Lambelet et al. 1990; Shroot and Brown 1986). For example, a recent study showed that dithranol accumulated in keratinocyte mitochondria (McGill et al., 2005), induced structural damage, and interfered with the redox status of the endogenous ubiquinone pool at the level of the ubisemiquinone anion. These events lead to increased generation of reactive oxygen species and eventually resulted in apoptosis.#

There are some indications that either or both epidermal proliferation/keratinization and cutaneous inflammation may be crucial in the antipsoriatic effect of dithranol; indeed, the often serious inflammation and the irritative response of the perilesional or uninvolved skin are the most serious limitations to its use (Swinkels et al., 2002). Scratching human skin evokes an inflammatory reaction dependent in part on the sensory nerves (e.g., neurogenic inflammation; see Steinhoff et al., 2003 for references) in the dermis (McGrouther and Ahmad, 1998), while cutaneous inflammation following injury, surgical intervention, contact with chemical irritants, etc. affects the epidermal innervation. The extent of such inflammation on the peripheral nerves has been successfully demonstrated immunohistochemically by the use of the ubiquitin carboxyl-terminal hydrolase L1 (UCH-L1; gene aliases: pan-neuronal marker protein gene product (PGP) 9.5, ubiquitin thiolesterase, PARK5), a cytosolic deubiquitinating enzyme that is highly expressed both in humans and in animals (Doran et al., 1983). For example, the UCH-L1 immunoreactivity was decreased in the oral mucosa of patients with chronic inflammation caused by oral lichen planus (Nissalo et al., 2000), after the intradermal injection of capsaicin (Simone et al., 1998), in leprosy skin (Facer et al., 2000) and around the sweat glands in palmoplantar pustulosis (Hagforsen et al., 2000). In animals, the expression of UCH-L1 was absent in necrotic areas of the colon of rats treated with 2,4,6-trinitrobenzenesulfonic acid to induce experimental colitis (Poli et al., 2001), and its immunoreactivity was decreased in rats with nasosinusitis infected artificially with Staphylococcus bacteria (Ge et al., 2002). Interestingly, psoriasis, an inflammatory skin disorder often treated with dithranol, was reported to be accompanied by either decreased (Johansson et al., 1991) or unchanged (Al'Abadie et al., 1995) UCH-L1 immunoreactivity in the affected skin. In light of these conflicting results, we investigated whether the inflammation elicited by dithranol would affect the UCH-L1 immunoreactivity of the cutaneous nociceptive sensory fibers of the rat orofacial skin.#

The animals treated with dithranol began to rub their perioral area, predominantly with their ipsilateral paw, within 1 h and continued to do so throughout the entire treatment regimen. Reddened perioral swelling was evident on the treated surface after about 3 h. Rats receiving physiological saline did not exhibit paw-related behavioral reactions or perioral inflammation. Dithranol treatment resulted in severe inflammation in the orofacial skin, as evidenced by hematoxylin staining and UCH-L1 immunohistochemistry. The histological signs of the inflammation were already evident after 3 days of treatment (data not shown), but were more severe after 5 days of dithranol treatment; by this time some hair loss also occurred. In the inflamed tissue, hematoxylin staining demonstrated the usual histological signs of severe skin inflammation, e.g., hyperkeratinization, abnormal thickening of the stratum corneum, and acanthosis (Fig. 1C). Corticosteroid treatment reversed these symptoms in the diseased animals (Fig. 1D). As revealed by hematoxylin staining, corticosteroid treatment alone did not have any histological effect (Fig. 1E). When UCH-L1 was visualized, heavy immunoreactivity was seen in the perifollicular fibers in the normal skin (Fig. 1F), whereas the dithranol-treated inflamed skin was characterized by a complete loss of these fibers (Fig. 1H). Topical corticosteroid treatment of the inflamed skin for 5 days almost completely restored the UCH-L1 immunoreactivity in the cutaneous sensory nerves (Fig. 1I), while the same steroid treatment alone did not have any effect on the UCH-L1 immunoreactivity of the neuronal elements of the orofacial skin (Fig. 1J).#

These immunohistochemical findings were supported by Western blot analysis. An UCH-L1-immunoreactive band of the expected size (approximately 24 kDa) was detected from control and treated tissues (Fig. 2). The intensity of the bands was similar for the control, corticosteroid-treated and dithranol- and steroid-treated samples, while the average intensity of the dithranol-treated samples was only 39.5% of the control values.#

The UCH-L1 gene is a member of a gene family whose products remove ubiquitin from ubiquitinated cellular proteins by hydrolyzing small C-terminal adducts of ubiquitin to generate ubiquitin monomers, thereby preventing them from targeted degradation by the proteasome-dependent pathways. The expression of UCH-L1 is highly specific to neurons and to cells of the diffuse neuroendocrine system and their tumors (Doran et al., 1983). The protein is abundant in the nervous system (1–2% of the total protein content of the brain; Wilkinson et al., 1989), where its function is still not completely understood. In vitro, however, UCH-L1 catalyzes the hydrolysis of C-terminal ubiquityl esters and amides (Larsen et al., 1998); this activity is presumed to be critical for cytoplasmic protein degradation. The degradation of proteins via the ubiquitin pathway involves two successive steps: a conjugation of multiple ubiquitin moieties to the substrate, and degradation of the tagged protein by a downstream 26S proteasome complex with the release of ubiquitins. Among the key proteins degraded by the ubiquitin-proteasome system are those involved in the control of inflammation, cell cycle regulation, gene expression and development, and differentiation (Ciechanover et al. 2000; Evans 2005).#

Recent proteomic analysis provided evidence that full-length UCH-L1 is a major target of oxidative damage (Choi et al., 2004) as UCH-L1 can be extensively modified by carbonyl formation and methionine and cysteine oxidation. Thus, impairment of the UCH-L1 ubiquitin hydrolase activity may be an important contributor to the neurodegeneration associated with the accumulation of ubiquitinated proteins and inflammation (Li et al., 2004). In that study, the molecular mechanisms linking inflammation with neurodegeneration were investigated in neuronal cultures. Treatment with certain prostaglandins, which are mediators of inflammation, reduced the viability and increased the levels of ubiquitinated proteins in the neuronal cells. The presence of intracellular aggregates containing ubiquitinated proteins in some of the treated cells indicated that these aggregates can form independently of proteasome inhibition.#

Dithranol undergoes a complex chemical transformation after topical application. The generation of oxygen free radicals is responsible for both the antipsoriatic and inflammatory effects of the drug (Willis et al., 2001). The experimental data suggest that the oxidative stress generated at the site of dithranol treatment alters the expressions of dermal chemokines and other cytokines, resulting in the recruitment of inflammatory cells (Lange et al., 1998). As dithranol is a powerful tool for the generation of reactive oxidative species in the skin, it is possible that the observed loss of UCH-L1 immunoreactivity in inflamed skin is due to the extensive oxidative damage to the UCH-L1 protein. Previous studies have demonstrated a decrease in UCH-L1 immunoreactivity in several forms of inflammation elicited by bacterial infection (Ge et al., 2002), capsaicin treatment (Simone et al., 1998) or skin diseases of unknown origin, such as lichen planus or lichenoid reactions (Nissalo et al., 2000). Our results show that dithranol-induced inflammation is also a condition that powerfully affects the immunohistochemically detectable amounts of UCH-L1 as it completely eliminates the UCH-L1 immunoreactivity in the nerve fibers in the rat orofacial skin in a corticosteroid-reversible manner. The fact that UCH-L1 immunoreactivity was not completely abolished in the samples used for Western analysis can be explained by (1) the presence of contaminating small amounts of tissue perhaps less affected by dithranol, and (2) the different accessibility of the epitopes in histological sections and denaturing gel electrophoretic circumstances.#

Beneficial effects of steroids (estrogens, catecholestrogens, phytoestrogens, dehydroepiandrosterone, 7α-hydroxy-dehydroepiandrosterone, etc.) in preventing oxidative stress-associated tissue injury have been observed in different experimental models (Gagne et al., 2003; Munoz-Castaneda et al., 2006; Pelissier et al., 2006), and the effectiveness of antioxidants against dithranol-associated toxicities have been successfully demonstrated (Lange et al. 1998; Swinkels et al. 2001 2003). Our results show that corticosteroid treatment ameliorates dithranol-induced inflammation, as it attenuates the oxidative damage to UCH-L1, which leads to the reappearance of its immunohistochemically detectable form.#