The generation of antisera specific for the priming phosphorylation sites on protein kinase Calpha (PKCalpha) has permitted analysis of the dephosphorylation of these sites in relation to the down-regulation of the protein. It was demonstrated that these priming sites are subject to agonist-induced dephosphorylation, consistent with inactivation of the protein. Further, the process is shown to be blocked by a PKC inhibitor, indicating a requirement for PKC catalytic activity. This was corroborated by showing that a constitutively active fragment of PKCalpha is able to stimulate the dephosphorylation of wild-type PKCalpha in transfected cells. Consistent with a membrane-traffic event, the process controlled by PKC that leads to dephosphorylation is shown to be temperature-sensitive and to correlate with transient accumulation of PKCalpha on cytoplasmic vesicular structures. It was established that the dephosphorylation of priming sites in PKCalpha is not unique and occurs with other conventional PKC isotypes, demonstrating that this is a general desensitization process for this subclass of kinases. The physiological importance of this desensitization is evidenced by the behaviour of PKCbeta1 in U937 cells, where dephosphorylation of the activation loop site is shown to be a function of cell density.

Dominant negative properties are conferred on protein kinase (PK) Calpha by mutation of the phosphorylation site in the activation loop of the kinase domain. To address the universality and/or specificity of such mutations, analogous alterations were introduced in other members of the PKC family and tested for their effects on the function of co-transfected activated PKC. For all three subclasses of the PKC family, mutations of the predicted activation loop phosphorylation sites resulted in dominant negative properties. These properties were not restricted to the cognate PKC isotypes, but were effective across the different subclasses. For example, two PKCzeta mutants (atypical isotype) inhibited both PKCalpha (classical isotype) and PKCepsilon (novel isotype). For all these mutants, inhibition correlated with an ability to prevent the accumulation of phosphorylated PKCalpha, consistent with the expected mode of action. In the case of the PKCalpha mutant, it was shown that inhibition required the full-length mutant protein. The results provide evidence for the involvement of a common step in the phosphorylation of all PKC isotypes.

In the context of the crosstalk between transmembrane signalling pathways, we studied the loci within the stimulatory receptor/Gs protein/adenylyl cyclase system at which protein kinase C (PKC) exerts regulatory effects in rat prostatic epithelial cells. The treatment of cells with the PKC activator phorbol 12-myristate 13-acetate (PMA) resulted in an impairment of the stimulation of adenylyl cyclase activity in terms of both potency, as seen with both vasoactive intestinal peptide (VIP) and pituitary adenylyl cyclase-activating peptide (PACAP-27), and efficacy, as seen with the beta-adrenergic agonist isoproterenol. This inhibitory effect of PMA could be prevented by cell incubation with pertussis toxin but not with cholera toxin, pointing to a Gi- but not Gs-dependent mechanism. This hypothesis was reinforced by ADP-ribosylation experiments that showed a low extent of alpha i with pertussis toxin but no change of alpha s with cholera toxin, as well as by the observation of the loss of the ability of low Gpp[NH]p doses to inhibit forskolin-stimulated adenylyl cyclase activity (a measure of Gi function) after cell treatment with PMA. However, the phorbol ester did not modify the adenylyl cyclase catalytic subunit, as shown by experiments on direct stimulation of the enzyme by forskolin. Whatever the exact mechanisms, the results support a crosstalk between the PKC and the adenylyl cyclase systems in rat prostatic epithelial cells in terms of an impairment of adenylyl cyclase stimulation, due presumably to phosphorylation of both membrane receptors (coupled to Gs) and Gi protein, but not of Gs protein or the adenylyl cyclase itself.

The properties of protein kinase C (PKC) activity have been studied in cytosolic and membrane fractions from rat and human prostate. Ion exchange chromatography indicated the existence of different PKC isoforms, PKC from rat ventral prostate behaved as a classical Ca(2+)- and phospholipid-dependent enzyme and was activated by 1,2-diacylglycerol as well as by high concentrations of arachidonic acid. PKC activity in the cytosolic fraction was higher and presented different cofactor requirements than that in the membrane fraction. PKC from human benign hyperplastic prostate was also phospholipid dependent, activated by tumor-promotong phorbol esters, and appeared to belong to the group of PKC isozymes which lack Ca2+ sensitivity. Human prostatic PKC activity appeared to be of similar nature in both membrane and cytosolic fractions but the specific activity was higher in the particulate preparation which could be related to the stage of endogenous activation of the enzyme. These results extend previous observations in rat ventral prostate and present evidences on the human counterpart. Forthcoming experiments are needed to establish the exact nature of PKC isozymes and their physiological and pathophysiological role in this gland.

No abstract

No abstract