Neurotransmitter suppression of the voltage-dependent K+ conductance (M-conductance, gM) is a ubiquitous mechanism for increasing neuronal excitability (Adams et al. 1982). We found that ATP (P2Y)-induced gM suppression in bullfrog sympathetic ganglion (BFSG) neurons involves phospholipase C (PLC) but not the downstream messengers, Ca2+, inositol 1,4,5-trisphosphate or protein kinase C. Because the PLC inhibitor, U73122 (EC50 ~3 µM) promoted a transient increase in gM, we suggested that preservation of membrane levels of phosphatidylinositol 4, 5 bisphosphate (PIP2) may be conducive to M-channel opening (Stemkowski et al. 2002), see also Suh & Hille (2002). The transduction mechanism for ATP-induced gM suppression may therefore involve PLC depletion of PIP2 rather than the production of downstream products. To test this, we used standard whole-cell recording from BFSG neurons isolated from bullfrogs that were humanely killed according to locally and nationally approved protocols (Selyanko et al. 1990). Data are expressed as means ± S.E.M. and compared with Student’s two-tailed t test for paired or unpaired data. The phosphatidylinositol-4-kinase (PtdIns4K) inhibitor, wortmannin (10 µM), significantly slowed the rate of recovery of gM suppression induced by 250 µM ATP. The time for 50 % recovery increased from 19.7 ± 8.3 to 198.1 ± 33.5 s after 5 min in wortmannin, (n = 11, P < 0.0004, paired t test). Because this effect was not seen with LY294002 (10 µM, an inhibitor of phosphatidylinositol 3-kinase) or ML-7 (10 µM, an inhibitor of myosin light chain kinase), the effect of wortmannin probably reflects an action on PtdIns4K. This would impair re-synthesis of PIP2 following its depletion during the action of ATP. Interruption of the lipid cycle at an earlier point, by inhibition of diacylglycerol kinase with R59022 (40 µM), also slowed recovery of ATP responses (half-time of recovery increased from 22.6 ± 6.0 to 56.7 ± 8.1 s, n = 6, P < 0.01, paired t test). This effect required more than one application of agonist to deplete the levels of phospholipid intermediates within the cycle. When a ‘PIP2 neutralising antibody’ (Huang et al. 1998) was included (1:100) in the patch pipette, ATP initially suppressed gM by 85.6 ± 3.1 % but after 25 min of recording, the agonist effectiveness was reduced so that only 42.9 ± 15.2 % suppression was seen (n = 4, P = 0.05, paired t test). Inclusion of a similar concentration of horse serum as a control failed to affect ATP-induced gM suppression. When KCNQ2/3 channels were expressed in tsA201 cells, the resulting ‘M-like’ current ran down to 48.7 ± 4.8 % (n = 21) of control in 5 min. Inclusion of PIP2 (40 µM) in the pipette significantly attenuated rundown to only 78.3 ± 15.7 % of control (n = 10, P < 0.05, unpaired t test). Currents recorded in inside-out, excised patches were reduced by 47.0 ± 6.7 % (n = 5) by 50 µM Al3+ which disrupts PIP2-ion channel interactions (Hilgemann & Ball, 1996). Although the PLC inhibitor U73122 (10 µM) antagonised gM suppression produced by 2 µM muscarine, a similar effect was observed with the inactive isomer, U73343. It was therefore difficult to demonstrate a role for PLC in the effect of muscarine. Despite this, the time for 50 % recovery of gM from muscarine inhibition was consistently increased after 5 min in 10 µM wortmannin. The PIP2 antibody reduced muscarine-induced gM suppression from 87.0 ± 6.3 to 64.0 ± 8.7 % (n = 5, P < 0.002, paired t test). These results are largely consistent with the hypothesis that ATP- and muscarine-induced gM suppression involves PLC depletion of PIP2.
This work was supported by the Alberta Heart and Stroke Foundation and Canadian Institutes for Health Research. C.P.F. received an Alberta Heritage Foundation for Medical Research Studentship. We thank D. McKinnon for gifts of KCNQ2/3 cDNA.