Robert I. Norman

Distribution of Kir6.0 and SUR2 ATP-sensitive potassium channel subunits in isolated ventricular myocytes

The subcellular distribution of ATP-sensitive potassium (K(ATP)) channel subunits in rat-isolated ventricular myocytes was investigated using a panel of subunit-specific antisera. Kir6.1 subunits were associated predominantly with myofibril structures and were co-localized with the mitochondrial marker MitoFluor red (correlation coefficient (cc) = 0.63 +/- 0.05). Anti-Kir6.1 antibodies specifically recognized a polypeptide of 48 kDa in mitochondrial membrane fractions consistent with the presence of Kir6.1 subunits in this organelle. Both Kir6.2 and SUR2A subunits were distributed universally over the sarcolemma. Lower-intensity antibody-associated immunofluorescence was detected intracellularly, which was correlated with the distribution of MitoFluor red in both cases (cc, Kir6.2, 0.56 +/- 0.05; SUR2A, 0.61 +/- 0.06). A polypeptide of 40 kDa was recognized by anti-Kir6.2-subunit antibodies in western blots of both microsomal and mitochondrial membrane fractions consistent with the presence of this subunit in the sarcolemma and mitochondria. Similarly, SUR2A and SUR2B subunits were detected in western blots of microsomal membrane proteins consistent with a sarcolemmal localization for these polypeptides. SUR2B subunits were shown in confocal microscopy to co-localize strongly with t-tubules (cc, 0.81 +/- 0.05). Together, the results indicate that Kir6.2 and SUR2A subunits predominate in the sarcolemma of ventricular myocytes consistent with a Kir6.2/SUR2A-subunit combination in the sarcolemmal K(ATP)channel. Kir6.1, Kir6.2 and SUR2A subunits were demonstrated in mitochondria. Combinations of these subunits would not explain the reported pharmacology of the mitochondrial K(ATP) channel (Mol Pharmacol 59 (2001) 225) suggesting the possibility of further unidentified components of this channel.

Characterisation of Kir2.0 proteins in the rat cerebellum and hippocampus by polyclonal antibodies.

Abstract Rabbit polyclonal antibodies were raised to rat Kir2.0 (Kir2.1, Kir2.2 and Kir2.3) inwardly rectifying potassium ion channel proteins. The antibody specificities were confirmed by immunoprecipitation of [35S]-methionine-labelled in vitro translated channel proteins and western blotting. Immunohistochemistry revealed a different patterns of expression of Kir2.0 subfamily proteins in the rat hind-brain (cerebellum and medulla) and fore-brain (hippocampus). Notably, only Kir2.2 protein was detected in the cerebellum and medulla, Kir2.1, Kir2.2 and Kir2.3 proteins were expressed in the hippocampus and immunostaining was not limited to neuronal cell types. Anti-Kir2.1 (fore-brain only) and anti-Kir2.2 (fore- and hind-brain) antibodies showed positive staining in macroglia, endothelia, ependyma and vascular smooth muscle cells. In contrast, anti-Kir2.3 (fore-brain only) immunostaining was limited to neurons, macroglia and vascular smooth muscle. These results indicate that specific regions within the rat fore- and hind-brain have differential distributions of inwardly rectifying potassium ion channel proteins.

Use of site-directed antibodies to probe the topography of the alpha 2 subunit of voltage-gated Ca2+ channels.

Polyclonal antibodies were raised against peptides corresponding to residues 1–15, 469–483 and 933–951 of the rabbit skeletal muscle L‐type calcium channelα2/δ primary translation product, for use as topological probes. Immunocytochemical comparison of the abilities of the antibodies to bind to theα2 and δ subunits in intact and detergent‐permeabilised rat dorsal root ganglion cells enabled the membrane orientation of these regions to be established. The resultant data indicate that the regions containing residues 1–15 and 469–483 of theα2 subunit, and residues 1–17 of the δ subunit, are exposed on the extracellular surface of the membrane, findings consistent with a model that proposesα2 to be entirely extracellular.

Endothelin-I and angiotensin II inhibit arterial voltage-gated K+ channels through different protein kinase C isoenzymes.

AIMS Voltage-gated K+ (Kv) channels of arterial smooth muscle (ASM) modulate arterial tone and are inhibited by vasoconstrictors through protein kinase C (PKC). We aimed to determine whether endothelin-1 (ET-1) and angiotensin II (AngII), which cause similar inhibition of Kv, use the same signalling pathway and PKC isoenzyme to exert their effects on Kv and to compare the involvement of PKC isoenzymes in contractile responses to these agents. METHODS AND RESULTS Kv currents recorded using the patch clamp technique with freshly isolated rat mesenteric ASM cells were inhibited by ET-1 or AngII. Inclusion of a PKCepsilon inhibitor peptide in the intracellular solution substantially reduced inhibition by AngII, but did not affect that by ET-1. Kv inhibition by ET-1 was reduced by the conventional PKC inhibitor Gö 6976 but not by the PKCbeta inhibitor LY333531. Selective peptide inhibitors of PKCalpha and PKCepsilon were linked to a Tat carrier peptide to make them membrane permeable and used to show that inhibition of PKCalpha prevented ET-1 inhibition of Kv current, but did not affect that by AngII. In contrast, inhibition of PKCepsilon prevented Kv inhibition by AngII but not by ET-1. The Tat-linked inhibitor peptides were also used to investigate the involvement of PKCalpha and PKCepsilon in the contractile responses of mesenteric arterial rings, showing that ET-1 contractions were substantially reduced by inhibition of PKCalpha, but unaffected by inhibition of PKCepsilon. AngII contractions were unaffected by inhibition of PKCalpha but substantially reduced by inhibition of PKCepsilon. CONCLUSION ET-1 inhibits Kv channels of mesenteric ASM through activation of PKCalpha, while AngII does so through PKCepsilon. This implies that ET-1 and AngII target Kv channels of ASM through different pathways of PKC-interacting proteins, so each vasoconstrictor enables its distinct PKC isoenzyme to interact functionally with the Kv channel.

Phenylephrine preconditioning involves modulation of cardiac sarcolemmal K(ATP) current by PKC delta, AMPK and p38 MAPK.

Preconditioning of hearts with the α(1)-adrenoceptor agonist phenylephrine decreases infarct size and increases the functional recovery of the heart following ischaemia-reperfusion. However, the cellular mechanisms responsible for this protection are not known. We investigated the role of protein kinase C ε and δ (PKCε and PKCδ), AMP-activated protein kinase (AMPK), p38 MAPK (p38) and sarcolemmal ATP-sensitive potassium (sarcK(ATP)) channels in phenylephrine preconditioning using isolated rat ventricular myocytes. Preconditioning of ventricular myocytes with phenylephrine increased the recovery of contractile activity following metabolic inhibition and re-energisation from 30.1±1.9% to 66.5±5.2% (P<0.01) and increased the peak sarcK(ATP) current activated during metabolic inhibition from 32.1±1.8 pA/pF to 46.0±5.0 pA/pF (P<0.05), which was required for protection. Phenylephrine preconditioning resulted in a sustained activation of PKCε and PKCδ, and transient activation of AMPK, which was dependent upon activation of PKCδ but not PKCε. P38 was also activated by phenylephrine preconditioning and this was blocked by inhibitors of PKCε, PKCδ or AMPK. Inhibition of PKCδ, AMPK or p38 was sufficient to prevent the increase in current, suggesting that these kinases are involved in modulation of sarcK(ATP) channel current by phenylephrine preconditioning. However, whilst inhibition of AMPK and p38 prevented the protection from phenylephrine preconditioning, PKCδ inhibition paradoxically had no effect. The increase in sarcK(ATP) current induced by phenylephrine preconditioning requires PKCδ, AMPK and p38 and may contribute to the observed improvement in contractile recovery.

SUR2A C‐terminal fragments reduce KATP currents and ischaemic tolerance of rat cardiac myocytes

C‐terminal fragments of the sulphonylurea receptor SUR2A can alter the functional expression of cloned ATP‐sensitive K+ channels (KATP). To investigate the protective role of KATP channels during metabolic stress we transfected SUR2A fragments into adult rat cardiac myocytes. A fragment comprising residues 1294–1358, the A‐fragment, reduced sarcolemmal KATP currents by over 85% after 2 days (pinacidil‐activated current densities were: vector alone 7.04 ± 1.22; and A‐fragment 0.94 ± 0.07 pA pF−1, n= 6,6, P < 0.001). An inactive fragment (1358–1545, current density 6.30 ± 0.85 pA pF−1, n= 6) was used as a control. During metabolic inhibition (CN and iodoacetate) of isolated myocytes stimulated at 1 Hz, the A‐fragment delayed action potential shortening and contractile failure, but accelerated rigor contraction and increased Ca2+ loading. On reperfusion, A‐fragment‐transfected cells also showed increased intracellular Ca2+ and the proportion of cells recovering contractile function was reduced from 40.0 to 9.5% (P < 0.01). The protective effect of pretreatment with 2,4‐dinitrophenol, measured from increased functional recovery and reduced Ca2+ loading, was abolished by the A‐fragment. Our data are consistent with a role for KATP channels in causing action potential failure and reduced Ca2+ loading during metabolic stress, and with a major role in protection by preconditioning. The effects of the A‐fragment may arise entirely from reduced expression of the sarcolemmal KATP channel, but we also discuss the possibility of mitochondrial effects.

Proximal C-terminal domain of sulphonylurea receptor 2A interacts with pore-forming Kir6 subunits in KATP channels

Functional KATP (ATP-sensitive potassium) channels are hetero-octamers of four Kir6 (inwardly rectifying potassium) channel subunits and four SUR (sulphonylurea receptor) subunits. Possible interactions between the C-terminal domain of SUR2A and Kir6.2 were investigated by co-immunoprecipitation of rat SUR2A C-terminal fragments with full-length Kir6.2 and by analysis of cloned KATP channel function and distribution in HEK-293 cells (human embryonic kidney 293 cells) in the presence of competing rSUR2A fragments. Three maltose-binding protein-SUR2A fusions, rSUR2A-CTA (rSUR2A residues 1254-1545), rSUR2A-CTB (residues 1254-1403) and rSUR2A-CTC (residues 1294-1403), were co-immunoprecipitated with full-length Kir6.2 using a polyclonal anti-Kir6.2 antiserum. A fourth C-terminal domain fragment, rSUR2A-CTD (residues 1358-1545) did not co-immunoprecipitate with Kir6.2 under the same conditions, indicating a direct interaction between Kir6.2 and a 65-amino-acid section of the cytoplasmic C-terminal region of rSUR2A between residues 1294 and 1358. ATP- and glibenclamide-sensitive K+ currents were decreased in HEK-293 cells expressing full-length Kir6 and SUR2 subunits that were transiently transfected with fragments rSUR2A-CTA, rSUR2A-CTC and rSUR2A-CTE (residues 1294-1359) compared with fragment rSUR2A-CTD or mock-transfected cells, suggesting either channel inhibition or a reduction in the number of functional KATP channels at the cell surface. Anti-KATP channel subunit-associated fluorescence in the cell membrane was substantially lower and intracellular fluorescence increased in rSUR2A-CTE expressing cells; thus, SUR2A fragments containing residues 1294-1358 reduce current by decreasing the number of channel subunits in the cell membrane. These results identify a site in the C-terminal domain of rSUR2A, between residues 1294 and 1358, whose direct interaction with full-length Kir6.2 is crucial for the assembly of functional KATP channels.

KATP channels mediate the β2-adrenoceptor agonist-induced relaxation of rat detrusor muscle

Abstract We propose that ATP-sensitive K + (K ATP ) channels are normally inactive but involved in β 2 -adrenoceptor stimulated relaxation of the rat bladder. Spontaneous detrusor muscle contractions were unaffected by glibenclamide (K ATP channel blocker) but were reduced when pinacidil (K ATP channel opener) concentrations exceeded 10 −5 M. Inhibition by β 2 -adrenoceptor agonist clenbuterol (10 −6 M) of 1 Hz electrical field stimulated contractions was abolished by glibenclamide (10 −6 M). Glibenclamide (10 −6 M) decreased forskolin-induced relaxation (10 −9 –10 −4 M) in bladder muscle stimulated with 1 Hz electrical field. In the presence glibenclamide (10 −6 M) or myristoylated protein kinase A inhibitor (2×10 −6 M), clenbuterol (10 −9 –10 −5 M) failed to inhibit bladder contraction in response to 1 Hz electrical field stimulation. Therefore, K ATP channel opening and the subsequent hyperpolarization of cell membranes in response to β 2 -adrenoceptor activation is mediated by raised cyclic-AMP levels and activation of protein kinase A. This counteracts ATP-stimulated depolarization in bladder muscle, thereby reducing cell contraction.

Cyclic AMP-vepenvent protein kinase phosphorylates residues in the C-terminal domain of the cardiac L-type calcium channel α1 subunit

The molecular basis of the regulation of cardiac L-type calcium channel activity by cAMP-dependent protein kinase (cA-PK) remains unclear. Direct cA-PK-dependent phosphorylation of the bovine ventricular alpha1 subunit in vitro has been demonstrated in microsomal membranes, detergent extracts and partially purified (+)-[3H]PN 200-110 receptor preparations. Two 32P-labeled phosphopeptides, derived from cyanogen bromide cleavage, of 4.7 and 9.5 kDa were immunoprecipitated specifically by site-directed antibodies against the rabbit cardiac alpha1 subunit amino acid sequences 1602-1616 and 1681-1694, respectively, consistent with phosphorylation at the cA-PK consensus sites at Ser(1627) and Ser(1700). No phosphopeptide products consistent with phosphorylation at three other C-terminal cA-PK consensus phosphorylation sites (Ser(1575), Ser(1848) and Ser(1928)) were identified using similar procedures suggesting that these sites are poor substrates for this kinase. Ser(1627) and Ser(1700) may represent sites of cA-PK phosphorylation involved in the physiological regulation of cardiac L-type calcium channel function.

Reduced effectiveness of HMR 1098 in blocking cardiac sarcolemmal KATP channels during metabolic stress

ATP-sensitive K(+) (K(ATP)) channels are involved in ischemic cardioprotection induced by preconditioning (IPC), though the relative role of sarcolemmal (sK(ATP)) and mitochondrial (mitoK(ATP)) channels remains controversial. The sK(ATP)-selective sulphonylthiourea HMR 1098 has often been reported to be without effect on ischemic cardioprotection, suggesting minimal involvement of sK(ATP). Since some sulphonylureas show reduced potency under conditions of metabolic stress, we used patch clamp to assess the ability of HMR 1098 to block sK(ATP) currents of adult rat ventricular myocytes activated by metabolic inhibition (MI, NaCN+iodoacetate). In contrast to the prototype sulphonylurea glibenclamide, HMR 1098 (10 muM) was without effect on sK(ATP) currents, and also did not inhibit MI-induced action potential shortening. However, HMR 1098 blocked sK(ATP) current induced by the K(ATP) opener pinacidil (IC(50)=0.36+/-0.02 muM), and reversed pinacidil-induced action potential shortening. In inside-out patches, block by HMR 1098 was relieved by increasing MgADP concentrations (1-100 muM). HMR 1098 inhibited pinacidil-activated recombinant Kir6.2/SUR2A channels with a similar IC(50) (0.30+/-0.04 muM), but was less effective when channels were activated by low intracellular ATP. HMR 1098 displaced binding of the pinacidil analogue [(3)H]P1075 to native cardiac membranes with a biphasic inhibition curve. Our results show that HMR 1098 becomes a much less effective inhibitor of sK(ATP) during metabolic stress, and suggest that the lack of effect of HMR 1098 on ischemic cardioprotection reported in some studies may represent loss of block by the drug under these conditions rather than a lack of involvement of sK(ATP) channels.