Pharmacological comparison of native mitochondrial K(ATP) channels with molecularly defined surface K(ATP) channels.

Many mammalian cells have two distinct types of ATP-sensitive potassium (K(ATP)) channels: the classic ones in the surface membrane (sK(ATP)) and others in the mitochondrial inner membrane (mitoK(ATP)). Cardiac mitoK(ATP) channels play a pivotal role in ischemic preconditioning, and thus represent interesting drug targets. Unfortunately, the molecular structure of mitoK(ATP) channels is unknown, in contrast to sK(ATP) channels, which are composed of a pore-forming subunit (Kir6.1 or Kir6.2) and a sulfonylurea receptor (SUR1, SUR2A, or SUR2B). As a means of probing the molecular makeup of mitoK(ATP) channels, we compared the pharmacology of native cardiac mitoK(ATP) channels with that of molecularly defined sK(ATP) channels expressed heterologously in human embryonic kidney 293 cells. Using mitochondrial oxidation to index mitoK(ATP) channel activity in rabbit ventricular myocytes, we found that pinacidil and diazoxide open mitoK(ATP) channels, but P-1075 does not. On the other hand, 5-hydroxydecanoic acid (5HD), but not HMR-1098, blocks mitoK(ATP) channels. Although pinacidil is a nonselective activator of expressed sK(ATP) channels, diazoxide did not open channels formed by Kir6.1/SUR2A, Kir6.2/SUR2A (known components of cardiac sK(ATP) channels) or Kir6.2/SUR2B. P-1075 activated all the K(ATP) channels, except Kir6.1/SUR1 channels. Glybenclamide potently blocked all sK(ATP) channels, but 5HD only blocked channels formed by SUR1/Kir6.1 or Kir6.2 (IC(50)s of 66 and 81 microM, respectively). This potency is similar to that for block of mitoK(ATP) channels (IC(50) = 95 microM). In addition, HMR-1098 potently blocked Kir6.2/SUR2A channels (IC(50) = 1.5 microM), but was 67 times less potent in blocking Kir6.1/SUR1 channels (IC(50) = 100 microM). Our results demonstrate that mitoK(ATP) channels closely resemble Kir6.1/SUR1 sK(ATP) channels in their pharmacological profiles.