Fig a and b illustrates
Fig. 3a and b illustrates a slow inhibition by PMA in 2 μM Ca2+. Moreover, channel inhibition was reversed by addition of PKC19–31, a peptide inhibitor specific for PKC, as shown in Fig. 3c. In three experiments, Im was reduced to 14.5±0.6% of the control value by PMA to increase to 207.6±73.6% by addition of PKC19–31. Similar effects were observed using calphostin C (1 μM) as the PKC inhibitor.
The PMA analog 4α-phorbol 12,13-didecanoate (4αPDD), which is uneffective on PKC, did not inhibit channel activity and did not maintain the channel inhibition previously induced by PMA. In three experiments, Im was reduced to 11.1±3.2% of the control value by PMA and restored to 83.2±13.4% by substitution of PMA with 4αPDD.
Channel inhibition was also obtained by substituting PMA with the structurally unrelated PKC activator OAG (see Fig. 4a and b), and the activity was resumed upon the addition of PKC19–31 (data not shown).
Fig. 5 shows how different calcium concentrations allow or counteract channel inhibition by a PKC activator.
The PMA-induced down-modulation of the mean current is mainly due to a decrease of the opening frequency. Samples of activity recorded from a patch containing a single channel are displayed in Fig. 6. Table 1 summarizes the values of opening frequency, mean channel open and closed times and single channel current, measured in the experiment illustrated in Fig. 6. The bursting activity of these AICAR indicates the existence of at least two closed states, PMA mainly affected the long closures.
Finally, the effects of concurrent activation of PKA and PKC on channel activity was studied. Channels which had been previously down-modulated by OAG were also perfused with a PKA-activating solution without theophylline, as shown in Fig. 4. It can be seen that the inhibition induced by OAG was then removed by the simultaneous activation of PKA. Reversing the order of presentation of protein kinase activations did not seem to alter the specific effects of the two kinases on channel activity. Fig. 7 illustrates how channel activity, up-modulated by PKA, can be reversibly depressed by the simultaneous application of OAG.
Moreover, the PKC inhibitor chelerythrine removed channel inhibition induced by OAG during PKA activation, as shown in Fig. 8. The relative potency of the two modulations was not further investigated because some findings suggested the occurrence of a variable basal channel inhibition in the absence of PKC activators. In fact, when applied alone, PKC19–31 attenuated the activity run-down following excision (data not shown). Furthermore, as indicated, addition of PKC19–31 after PMA, besides removing channel inhibition, often raised channel activity beyond the control level.
Discussion The structurally unrelated PKC activators PMA and OAG produced a similar inhibition of IKCa channels of human erythrocytes when applied in the inside–out configuration. The reduction in the patch current was associated with a clear-cut decrease in the opening frequency with minor changes in the mean channel open time and no change in unitary current. The PMA analog 4αPDD, which is known to have no effect on PKC, did not mimic the effect produced by PMA. Addition of PKC19–31, which acts as a pseudosubstrate maintaining the enzyme PKC in its inactive form , removed either PMA or OAG inhibition, often resulting in higher activity than in control conditions. Similar effects were observed using calphostin C, a potent and irreversible PKC inhibitor that affects the phorbol-ester binding site , or chelerythrine, which interacts with the catalytic domain of PKC by competitive inhibition of adenosine thriphosphate binding . These findings rule out the hypothesis that the observed effects are due to PKC activators per se  and indicate that activation of an endogenous PKC was responsible for Gardos channel down-modulation. Four isoforms of PKC have been detected in human red blood cells α, ζ , μ and ι . The α-isoform is probably involved in the effects on Gardos channels reported here, because channel inhibition was induced by both OAG and PMA. The short latency of channel inhibition measured in some patches at low calcium concentration (see the example illustrated in Fig. 2), as well as the consistency of the mechanism found in most of the cell-free membrane patches, make it unlikely that a complicated biochemical cascade is responsible for the PKC-induced channel inhibition. Although our results cannot allow us to identify the molecular target of phosphorylation, they are consistent with a direct modulation of the channel itself or of an associated regulatory protein.