Fig. 4. Coexpression of a truncated construct of MARK2 kinase with TRESK prevents activation of the channel by PMA. (A) The average amplitudes of basal TRESK currents, the currents after the application of PMA (100 nM), and the currents after the final administration of ionomycin (light blue, purple, and yellow columns, respectively, as I1, I2, and I3 in Fig. 2A) are represented. Three groups of oocytes expressing human TRESK, injected with half amount of TRESK cRNA, or coexpressing TRESK with MARK2Δ are illustrated (three triads of columns, respectively, as indicated below the triads). (B) Relative activations of TRESK currents by PMA (purple, as I2/I1 in Fig. 2A) and the final activations by ionomycin (0.5 μM) after PMA (yellow, as I3/I1 in Fig. 2A) were calculated for the three groups from the data shown in (A). Note that the activation by PMA was diminished in the presence of MARK2Δ (compare the purple columns); however, the effect of ionomycin was not influenced by the coexpression of the kinase (yellow columns). In this experiment, a double amount of TRESK cRNA was coinjected with MARK2Δ to obtain the same basal K+ current amplitude as in the control (TRESK) group. In a previous similar experiment, when equal amounts of TRESK cRNA were injected, the coexpression of MARK2Δ also prevented the activation of TRESK by PMA, although the basal K+ current was reduced in the cells coexpressing MARK2Δ compared with the control group (not shown). Note that the average K+ current after the application of PMA is in the same range in the TRESK (0.5×) as in the TRESK (2×) + MARK2Δ group, verifying that the degree of TRESK activation by PMA was independent of the current amplitudes in this experiment. ***P < 0.001, MANOVA followed by Tukey HSD test; ns, not significant.