Reduced epidermal cyclic AMP accumulation following prostaglandin stimulation: Its possible role in the pathophysiology of psoriasis

Three decades of psoriasis research: where has it led us?

Psoriasis is a common chronic skin disease. Its pathogenesis has intensively been investigated in the last 3 decades. In the 1970s, the observed increased proliferation of keratinocytes and their altered differentiation were considered to be the most important signs and causes of psoriatic skin lesions. Since the early 1980s, T cells slid into the focus of psoriasis research. It was then postulated that a subpopulation of T cells, so-called T1 cells, and their prominent cytokine interferon-gamma, had a dominant role in the pathogenesis of psoriasis. In the last decade, new data regarding macrophages and dendritic cells and the high therapeutic success of anti-tumor necrosis factor alpha biologics led to the assumption that antigen-presenting cells are important not only in the induction of psoriasis but also in its maintenance. The knowledge gained over the past 3 decades let us postulate that psoriasis is an immunologically induced, overshot, regeneration-like reaction of the skin in which various cells play a dominant role at different stages. This hypothesis is also supported by the very recent discoveries about interleukin (IL)-22, IL-20, and IL-23.

Drugs in exacerbation of psoriasis

Drugs that have been associated with the precipitation or exacerbation of psoriasis include lithium, beta adrenergic receptor blocking agents, and antimalarials. The withdrawal of corticosteroids has been reported to activate pustular psoriasis. Nonsteroidal anti-inflammatory drugs, such as indomethacin, have recently been reported to exacerbate psoriasis, although additional well-controlled studies are still needed. Drugs used for treatment of psoriasis will sometimes cause a flare because of irritation, phototoxicity, or hypersensitivity reaction resulting in a Koebner phenomenon. Because psoriasis is a very complex disease and its activity is often unpredictable, clinical studies on adverse drug effects on psoriasis have been difficult to conduct. This review evaluates clinical, histologic, and biochemical evidence in the literature for drug-associated onset or exacerbation of psoriasis.

Characterization of alpha- and beta-adrenergic agonist stimulation of adenylate cyclase activity in human epidermal keratinocytes in vitro

Adrenergic receptors are responsible for selective recognition and binding of catecholamines and may in turn have an effect on epidermal cell growth and maturation via adenosine-3',5'-monophosphate (cAMP). Using endogenous catecholamines and drugs specific for alpha- and beta-receptor subtypes, we have characterized the adrenergic receptor coupled to adenylate cyclase in cultured human epidermal keratinocytes. The relative potency order of stimulation of adenylate cyclase was: isoproterenol greater than epinephrine much greater than norepinephrine. The predominant adrenergic receptor is of the beta 2-subtype, as also confirmed by use of the selective antagonists propranolol, butoxamine, and atenolol. No evidence of alpha-adrenergic receptor mediation of adenylate cyclase was noted with the alpha 2-specific agonist, clonidine. Phenylephrine, the alpha 1-specific agonist, affected cAMP formation but this response could not be totally inhibited with prazosin, suggesting an unknown mechanism of action. Human keratinocytes retained the same beta-adrenergic receptor potency order properties throughout growth and maturation.

Effects of trypsin on the cyclic AMP system of the pig skin

Trypsin increased cyclic AMP levels of the pig skin. This effect was markedly potentiated by the cyclic nucleotide phosphodiesterase inhibitor theophylline. Without theophylline this trypsin-induced cyclic AMP accumulation was transient and the maximal accumulation was noted by 5 min. Soybean trypsin inhibitor inhibited this trypsin-induced cyclic AMP accumulation. After the trypsin treatment, marked acantholysis was noted histologically and the decreased responsiveness to other adenylate cyclase stimulators was seen. The decrease of the epinephrine response was most marked and that of histamine response was much less. Both low and high Km cyclic nucleotide phosphodiesterase activities were decreased by the trypsin treatment. However, at 5 min incubation time, when the increase in cyclic AMp level was most marked, the decrease in the phosphodiesterase activities was minimal. Trypsin seems to reveal its action through the proteolytic activation of adenylate cyclase system of the skin.

Inhibition of epidermal adenyl cyclase by lithium carbonate

An in vitro floating system was used to investigate the effect of lithium carbonate (Li2CO3) on the activity of adenyl cyclase in normal pig epidermis. Li2CO3 decreased the responsiveness of adenyl cyclase to stimulation by histamine, adenosine monophosphate (AMP) and epinephrine. This abnormality is similar but not identical to the previously described impaired responsiveness of adenyl cyclase to epinephrine and PGE2 in psoriatic plaques compared to normal adjacent skin. Involved and uninvolved skin from a psoriatic on lithium therapy demonstrated decreased responsiveness to in vitro stimulation by epinephrine, histamine and adenosine when compared to skin from psoriatics who were not on lithium therapy. These results are consistent with the observation that lithium therapy worsens psoriatic lesions.

Metabolic aspects of psoriasis: the basis for specific therapy

Psoriasis, like diabetes, is a multifactorial genetic disease with complex interactions of deranged metabolism. Factors that affect epidermal differentiation and proliferation include cyclic nucleotide interactions, polyamine metabolism, cell surface--cytoskeleton interactions, and arachidonic acid--prostaglandin cascade. Evidence indicates that pharmacologic manipulation of the so-called critical metabolic systems not only may help us to understand the pathophysiology of psoriasis but also may yield improved treatments for the disease.

Effect of prostaglandins on cyclic nucleotide levels in cultured keratinocytes

Ginea pig ear epidermal cells (keratinocytes) were established in primary cultures using trypsin, and treated in their proliferative phase of growth with prostaglandins E1, D1, F1 alpha, E2, D2, or F2 alpha. This phase is induced by the addition of retinoic acid during cell plating. Intracellular content of cAMP and cGMP was measured by radioimmunoassay at various times after treatment. Maximum stimulation of cAMP levels was observed with PGD2, smaller increases with PGE2 and relatively transient rises with PGF2 alpha which were of low significance, but confirm earlier data. Similar results were observed with PGD1, PGE1, and PGF1 alpha with smaller increases. The effects of D and E PGs were biphasic. Significant increases in cGMP were immediately observed with PGD2 and PGE2. With PGF2 alpha, maximum cGMP levels were noted after some delay. All PGs tested showed some effect in elevating cyclic nucleotides in keratinocytes. The most striking result was the increase in cAMP on PGD2 treatment.

The 2', 3'-cyclic AMP 3'-phosphohydrolase in normal and psoriatic epidermis

The activity of the enzyme 2', 3'-cyclic AMP 3'-phosphohydrolase is significantly greater in the 'involved' psoriatic skin than in the 'uninvolved' psoriatic skin or in skin samples taken from persons having clinically normal skin. Although the physiological function of this enzyme is not established, it is possible that, besides being associated with myelin, it may also play some role in cell proliferation and maturation, probably at the membrane level.

Cyclic nucleotide-phosphodiesterase in the uninvolved and involved skin of psoriasis

In the present study we have compared cyclic nucleotide-phosphodiesterase activities and affinity of phosphodiesterase for substrates (Km) in enzyme preparations obtained from the involved and uninvolved skin of psoriatic patients. With crude skin homogenates we consistently obtained two Km values (high and low) for both the involved and uninvolved, and both Km values were nearly identical between the involved and uninvolved. The same conclusion is also drawn from the Km determinations with partially purified preparations. Cyclic AMP-phosphodiesterase activities with crude homogenates showed no statistically significant differences between the involved and the uninvolved skin. However, when cyclic AMP- and cyclic GMP-phosphodiesterase activities were compared with a highly sensitive assay method in "pure" epidermal samples, which were microdissected free from stratum corneum, dermis and skin appendages, the involved skin contained 40% more activity of the low Km enzyme and 100% more of the high Km enzyme of both cyclic AMP- and cyclic GMP-phosphodiesterase. It is suggested that this may be due to a higher proportion of germinative cells in the lesional epidermis.

Cyclic AMP accumulation in psoriatic skin: differential responses to histamine, AMP, and einephrine by the uninvolved and involved epidermis

Using the uninvolved and involved skin from psoriatic patients, we investigated the effects of histamine and AMP (or adenosine) in vitro on the intracellular cyclic AMP levels. Both agents activated adenylate cyclase of the uninvolved and involved resulting in the accumulation of cyclic AMP. Without a cyclic nucleotide phosphodiesterase (PDE) inhibitor, these responses were biphasic and the maximal accumulation was observed in 5 min. With the PDE inhibitor both responses were markedly potentiated and high levels of cyclic AMP were observed for more than 20 min. The response to histamine by the involved skin was much greater than that by the uninvolved. The degree of the response to adenosine was approximately equal. In accordance with our previous work, the response to epinephrine by the involved skin was much less than that by the uninvolved. Thus adenylate cyclases of involved skin from psoriatic patients exhibit a markedly diminished response to epinephrine while at the same time exhibiting a markedly enhanced response to histamine. This precludes the possibility that the unresponsiveness to epinephrine can be due to a generalized inability of the epidermal psoriatic plaque cell to make a functioning cell membrane.

HLA C and D antigens associated with psoriasis

The frequency of two newly defined HLA antigens, HLA CT7 and DMA, was found to be greatly increased in patients with psoriasis. These two antigens and those previously found to be associated with the disease, B13, BW16 and BW17, frequently occurred together. The disease may be primarily associated with HLA DMA, or with the HLA haplotypes CT7--B13/W16/W17--DMA.

The relationship between cyclic adenosine 3', 5' - monophosphate and biochemical events in rat skin after the induction of epidermal hyperplasia using hexadecane

Sequential changes in skin metabolism have been studied in a model system of epidermal hyperplasia and hyperkeratinization induced by the application of n-hexadecane to shaved rat skin. The epidermal accumulation of glycogen typical of the hyperplastic response has been correlated with an increase in glycogenesis and a decrease in glycogenolysis. DNA synthesis was increased by 6 h after the start of hexadecane treatment and reached a maximum after one day. The concentration of skin cyclic AMP fell immediately after hexadecane application and subsequently rose to give a prolonged increase. Use of the combined topical application of hexadecane and the anti-inflammatory drugs triamcinolone acetonide, hydrocortisone and indomethacin showed that the hexadecane-induced changes in DNA synthesis and glycogen metabolism were linked to the initial fall in cyclic AMP concentration. The significance of the biphasic change in cyclic AMP levels is discussed as a possible system of control for the development and maintenance of hyperplasia.

Does ultraviolet-evoked prostaglandin formation protect skin from actinic cancer?

The role of increased prostaglandin activity which results from exposure of human skin to ultraviolet light is uncertain. Evidence from the effects of prostaglandins on epithelial cells in vitro suggests that prostaglandin E reduces proliferative activity in the epidermis, probably through a cyclic-A.M.P.-dependent mechanism, thus reducing the vulnerability of epidermis to the mutagenic effects of ultraviolet exposure.

Synthesis of prostaglandins by psoriatic skin

The biosynthesis of prostaglandins (PG) in biopsies from 9 patients with psoriasis was studied. The involved as well as the uninvolved psoriatic skin showed a statistically significant decrease of the ability to synthesize PG's. In PGE1-equivalents the concentration (mean +/- S.E.M.) was 4.41 +/- 0.48 ng/g wet weight in the psoriatic lesion, 5.41 +/- 0.64 ng/g wet weight in uninvolved psoriatic skin in the presence of exogenous arachidonic acid in the incubation medium as compared with 9.02 +/- 1.59 in normal human skin. When the skin was incubated without excess of exogenous precursor acid the activity formed was similarly significantly lower in psoriatic skin as compared with normal skin. A disturbed balance between E and F PG synthesis was not demonstrated, which might have accounted for the postulated altered intracellular ratio of cyclic AMP to cyclic GMP in psoriatic skin.

The effects of prostaglandins E1 and F2alpha on epidermal growth

The effects of prostaglandins E1 and F2alpha (PGE1 and PGF2alpha) on the growth of guinea pig epidermis have been studied in vivo. The prostaglandins were injected intradermally into guinea pigs, and tritiated thymidine was injected intraperitoneally prior to sacrifice. The autoradiographic labelling indices (L.I.) were assessed and a significant increase was found 4 h after injection of PGE1 -an effect which lasted for up to 72 h. PGF2alpha injections had no significant effect on the L.I.

Effect of prostaglandins on 3H-thymidine uptake into human epidermal cells in vitro

Prostaglandins A2, B1, E1, E2, F1alpha and F2alpha were added to cultures of human epidermal cells (keratinocytes) for 24 hours at 37 degrees C, and the effects on 3H-thymidine uptake into DNA was measured. At 70 mu/ml all prostaglandins tested except PGF2alpha inhibited the uptake of 3-thymidine greater than 50%. However, at 35 mug/ml, PGA2 and PGB1 were the only two prostaglandins to show significant inhibition, 96% and 51% respectively. At 17.5 mug/ml only PGA2 caused substantial inhibition, 68%. In order to determine if the PGA2 action was mediated by membrane receptors propranolol, phentolamine, metiamide and prostynoic acid were added in conjunction with PGA2. None of the above receptor antagonists were able to reduce the PGA2-induced inhibition of 3H-thymidine uptake. These results indicate that the pre-incubation of human keratinocytes with prostaglandins for 24 hours results in a decrease of 3H-thymidine incorporation in DNA. The precise mechanism of action is unknown at this time.

Increased concentrations of nonesterified arachidonic acid, 12L-hydroxy-5,8,10,14-eicosatetraenoic acid, prostaglandin E2, and prostaglandin F2alpha in epidermis of psoriasis

Lesional epidermis of psoriasis has a probable reduction in the cyclic AMP/cyclic GMP ratio. This altered ratio may in part be responsible for the characteristic glycogen storage, rapid cell proliferation, and reduced differentiation in lesional epidermis. The concentrations of prostaglandins E2 and F2alpha, free arachidonic acid, and 12L-hydroxy-5,8,10,14-eicosatetrawnoic acid in specimens of uninvolved and involved epidermis of psoriasis were measured with deuterium-labeled carriers and multiple ion analysis. Snap frozen specimens contained: 1.4 +/- 0.4 mug/g (wet weight) of arachidonic acid in uninvolved in contrast to 36.3 +/- 16.7 mug/g in involved epidermis (P = 0.015); less than 0.05 +/- 0.01 mug/g of hydroxyeicosatetraenoic acid in uninvolved in contrast to 4.1 +/- 1.9 mug/g in involved epidermis (P = 0.015); 23.6 +/- 5.0 ng/g of prostaglandin E2 in uninvolved in contrast to 33.1 +/- 5.7 ng/g in involved epidermis (P less than 0.01); and 21.0 +/- 4.4 ng/g of prostaglandin F2alpha in uninvolved in contrast to 39.0 +/- 5.9 ng/g in involved epidermis (P less than 0.01). The arachidonic acid and hydroxyeicosatetraenoic acid levels in involved epidermis were strongly correlated (r = 0.97). The increased levels of arachidonic acid and 12L-hydroxy-5,8,10,14-eicosatetraenoic acid in involved epidermis may have diagnostic and pathophysiological importance.