The conjunctival epithelium is intrinsically different from the corneal epithelium in vivo, but sometimes can transform into an epithelium morphologically indistinguishable from the latter after healing of a total corneal epithelial defect. It remains unclear whether this morphologic transformation represents a process of extrinsic modultation or transdifferentiation of intrinsically divergent epithelium. In air-lifted organotypic cultures, rabbit conjunctival epithelial cells lost goblet cell differentiation and were stratified to the same extent as corneal epithelial cells, resembling the above in vivo morphologic transformation. Paired expression of K3 (64 kD) and K12 (55 kD) keratins has been regarded as a marker for corneal-type differentiation. Immunoblot analysis by monoclonal antibody AE5 revealed that K3 keratin was expressed by both submerged or air-lifted corneal and conjunctival cultures with or without 3T3 fibroblasts in collagen gel. In contrast, K12 keratin was expressed only by air-lifted corneal cultures with 3T3 fibroblasts using monoclonal antibody AK2 and two epitope-specific antibodies to N- and C- terminal oligopeptides deduced from the mouse K12 gene. This finding was also confirmed by Northern hybridization with a rabbit K12 cDNA probe. The expression of K12 keratin was more delayed than that of K3 keratin in air-lifted corneal cultures. This dissociated expression of these two keratins resembles that noted in vivo in the stem cell-containing limbal region. These results suggest that morphologic transformation of the conjunctival epithelium represents extrinsic environment modulation, and that differential expression of K12 but not K3 keratin signifies corneal epithelial differentiation.

Evidence that an adenyl cyclase system is present in all mammalian epidermis is reviewed. This adenyl cyclase is stimulated by at least two separate types of chemicals: catecholamines, which act at a beta-adrenergic receptor site, and prostaglandins of the E series, which act at a separate site. In the psoriatic lesion, the response to these stimulators, especially to the catecholamines, is reduced. Despite this lack of response to external agents which elevate cyclic AMP, the concentration of cyclic AMP within the epidermis of the psoriatic lesion is no lower than in noninvolved skin. How cyclic nucleotides act to control cell proliferation and cell differentiation remains unclear.

Slices of human skin obtained with a keratome were pre-incubated with [3H]adenine to label the ATP pool from which cyclic AMP was subsequently formed. The accumulation of radioactive cyclic AMP was measured as an index of adenyl cyclase activity. The data showed that both the ability to incorporate [3H] into ATP and adenyl cyclase activity were significantly lower in psoriatic plaques than in uninvolved skin of the psoriatic patients, or in normal skin of control subjects. The response of adenyl cyclase to the stimulation of 3.3 muM adrenaline was less than five fold in psoriatic plaques as compared to twelve to thirty-two fold in the uninvolved skin. The response to the stimulation of prostaglandin E2 (5 mug/ml) showed no significant difference between the plaque and normal skin. The adenyl cyclase activity in uninvolved skin of psoriatic patients appeared normal. Propranolol (10 muM) blocked the stimulatory effect of adrenaline but not that of PGE2 in normal skin. These results suggest that the adenyl cyclase system of the skin has different regulatory sites for adrenaline and PGE2 and that the enzyme is defective in the epidermis of the psoriatic plaque, especially at the adrenaline regulatory site.

No abstract

No abstract