The Impact of ATP-Sensitive K؉ Channel Subtype Selectivity of Insulin Secretagogues for the Coronary Vasculature and the Myocardium Ulrich Quast, Damian Stephan, Susanne Bieger, and Ulrich Russ

Insulin secretagogues (sulfonylureas and glinides) in- crease insulin secretion by closing the ATP-sensitive K؉ ␤ nsulin secretagogues are widely prescribed in the channel (KATP channel) in the pancreatic -cell mem- brane. K channels subserve important functions also treatment of type 2 diabetes. They close the ATP- ATP ϩ in the heart. First, KATP channels in coronary myocytes sensitive K channel (KATP channel) in the mem- contribute to the control of coronary blood flow at rest Ibrane of the pancreatic ␤-cell, thereby depolarizing and in hypoxia. Second, K channels in the sarco- ATP the cell and triggering insulin secretion. KATP channels are lemma of cardiomyocytes (sarcKATP channels) are re- gated by intracellular nucleotides with ATP inducing chan- quired for adaptation of the heart to stress. In addition, nel closure and MgADP channel opening. The ␤-cell is the opening of sarcK channels and of K channels ATP ATP special in that physiological changes in plasma glucose in the inner membrane of mitochondria (mitoKATP chan- nels) plays a central role in ischemic preconditioning. change the intracellular ATP and ADP concentrations such that the channel opens and closes; hence, the channel Opening of sarcKATP channels also underlies the ST- segment elevation of the electrocardiogram, the pri- functions as the glucose sensor in this cell (1–3). KATP mary diagnostic tool for initiation of lysis therapy in channel subtypes are found in many cell types. The acute myocardial infarction. Therefore, inhibition of generation of mice in which the genes for the KATP channel cardiovascular KATP channels by insulin secretagogues subunits were deleted have shed new light on the diverse is considered to increase cardiovascular risk. Electro- functions of the KATP channels in various tissues in physiological experiments have shown that the secreta- physiological and pathophysiological conditions (1). In gogues differ in their selectivity for the pancreatic over brain, K channels are involved in actions as diverse as the cardiovascular K channels, being either highly ATP ATP the control of glucose homeostasis and the regulation of -selective (ϳ1,000؋; short sulfonylureas such as nateg neuronal excitability in hypoxia (1); however, the insulin ;linide and mitiglinide), moderately selective (10–20؋ long sulfonylureas such as glibenclamide [glyburide]), secretagogues do not cross the blood-brain barrier easily or essentially nonselective (<2؋; repaglinide). New enough to affect these channels at therapeutic plasma binding studies presented here give broadly similar levels (4). In several vascular beds, the KATP channel in the results. In clinical studies, these differences are not yet vascular myocytes is involved in the regulation of vessel taken into account. The hypothesis that the in vitro tone; opening is triggered in particular by stimuli increas- selectivity of the insulin secretagogues is of importance ing cAMP (5). In general, the channel opens in ischemia for the cardiovascular outcome of diabetic patients with and hypoxia when the ADP-to-ATP ratio increases; this coronary artery disease needs to be tested. Diabetes 53 clamps the cell at the Kϩ equilibrium potential and brings (Suppl. 3):S156–S164, 2004 the cell to rest.

KATP channels are composed of two types of subunits, inwardly rectifying Kϩ channels (Kir6.x) and sulfonylurea receptors (SURx), arranged as tetradimeric complexes,

(Kir6.x/SURx)4 (Fig. 1) (rev. in 1–3,6). The Kir6.x subunits form the pore of the channel. SURs function as regulatory subunits. They are members of the ATP-binding cassette From the Department of Pharmacology and Toxicology, Medical Faculty, University of Tu¨ bingen, Tu¨ bingen, Germany. protein superfamily, carry binding sites for nucleotides, Address correspondence and reprint requests to Ulrich Quast, Department and exhibit ATPase activity. In addition, SUR is endowed of Pharmacology and Toxicology, Medical Faculty, University of Tu¨ bingen, Wilhelmstra␤e. 56, D-72074 Tu¨ bingen, Germany. E-mail: ulrich.quast@uni- with the binding sites for the sulfonylureas and the KATP tuebingen.de. channel openers (Fig. 1). Both Kir6.x and SURx are Received for publication 12 March 2004 and accepted in revised form 1 June 2004. encoded by two genes; this, together with alternative This article is based on a presentation at a symposium. The symposium and splicing of SUR, gives rise to various combinations of the publication of this article were made possible by an unrestricted educa- Kir6.x/SURx, which account for the differences in the tional grant from Servier. ϩ biophysical and pharmacological properties of the K KATP channel, ATP-sensitive K channel; Kir channel, inwardly rectifying ATP ϩ K channel; mitoKATP channel, KATP channel in the inner membrane of channels in the various tissues. Of particular importance mitochondria; sarcKATP channel, KATP channel in the sarcolemma of cardio- myocytes; SUR, sulfonylurea receptor. for the following discussion are the KATP channels in © 2004 by the American Diabetes Association. pancreatic ␤-cells (Kir6.2/SUR1), in cardiomyocytes

S156 DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 U. QUAST AND ASSOCIATES

FIG. 1. Structure of KATP channels and its subunits. The binding regions of SUR for the insulinotropes and KATP channel openers are indicated (see text for details). CL, cytosolic loop; NBF, nucleotide binding fold.

(Kir6.2/SUR2A), and in vascular myocytes (Kir6.1/SUR2B) openers, diazoxide has been found to affect the KATP chan- (1,3,6). nels in the ␤-cell and in the vascular myocyte at similar The Kir6.x subtypes differ mainly in their single-channel concentrations (6), and, recently, openers with excellent ϩ conductance (ϳ30 vs. 80 pS in high symmetrical K selectivity for SUR1 over SUR2 have been synthesized (8). solution for Kir6.1 and Kir6.2, respectively) and in their In addition to these KATP channels in the cell membrane, sensitivity to block by ATP. Kir6.1 is only weakly sensitive there is evidence for a KATP channel in the inner mem- (K ϳ1 mmol/l), whereas for Kir6.2, the K is ϳ100 ␮mol/l i i brane of mitochondria (the mitoKATP channel). This chan- (1). The sensitivity of Kir6.2 to inhibition by ATP is nel was discovered in 1991 in electrophysiological increased by coexpression with SUR. In the Kir6.2-contain- experiments; however, despite intensive effort, its molec- ing channels, inhibition by ATP is mediated by ATP ular identity remains unknown (9,10). This channel has binding to Kir6.2, whereas channel activation by MgADP is been proposed to play a central role in ischemic precon- conferred by SUR (1–3,6). The SUR subtypes, SUR1 and ditioning (i.e., the mechanism by which brief ischemic SUR2, give rise to several isoforms due to alternative episodes “condition” the heart to better survive subse- splicing; of major importance are SUR2A and SUR2B, which differ in the last exon (i.e., by 44 amino acids) (6). quent longer periods of ischemic insult) (9,10). Generally, SUR1 has a higher affinity for sulfonylureas and In the following sections, we will first consider the glinides than the SUR2 isoforms, whereas SUR2 has a selectivity of representative insulin secretagogues for the K channel in the ␤-cell over that in the cardiomyocyte higher affinity for the KATP channel openers (1,6). There are ATP notable exceptions to both rules. The selectivity ranking of and present new binding studies dealing with this ques- the sulfonylureas and glinides for the pancreatic over the tion. Then the role of the KATP channels in the (human) cardiac KATP channels is detailed in Table 1; in addition, a coronary vascular bed and in cardiac myocytes is dis- sulfonylthiourea, HMR 1883 (HMR 1098), that is highly selec- cussed. In the last section, we examine the available tive for SUR2 has been synthesized (7). Regarding the clinical evidence that therapy with insulin secretagogues

TABLE 1 ͓3 ͔ Inhibition of H glibenclamide binding to recombinant KATP channels subtypes by sulfonylureas and glinides

SUR Ki (nmol/l) Selectivity Secretagogue subsite Kir6.2/SUR1 Kir6.2/SUR2A pancreas/heart* Glibornuride A 360 (330–400) 6,800 (6,200–7,400) 19 (17–21) Glibenclamide AϩB 0.45 (0.31–0.65) 6.2 (5.6–6.8)† 14 (10–20) Glimepiride AϩB 0.58 (0.53–0.60) 11 (10–12) 18 (15–22) Meglitinide B 3,200 (2,800–3,600) 830 (660–1,000) 0.26 (0.20–0.35) (Ϫ)AZ-DF 265 B 3.7 (2.9–4.9)‡ 25 (23–28) 6.8 (5.1–9.6) Repaglinide B 0.72 (0.60–0.87) 1.5 (1.4–1.7) 2.1 (1.7–2.6) Nateglinide A 350 (330–360) 10,000 (8,500–12,000) 30 (25–35) ف Data are Ki values (95% CIs). Experiments were performed as described in Fig. 3. Hill coefficients were 1. Ki values are followed by the ⌬pK ⌬ ϭ Ϫ 95% CI in parentheses. *Calculated as 10 with pK pKi(Kir6.2/SUR1) pKi(Kir6.2/SUR2A), taking propagation of errors into account. †For the recombinant vascular KATP channel, a value of 5.8 (5.2–6.2) nmol/l was obtained (20). ‡Ki value of high-affinity component, which ϭ comprised 87% of specific binding; the low-affinity component (13%) gave Ki,2 180 nmol/l.

DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 S157 CARDIAC KATP CHANNELS AND INSULINOTROPES

–NHCO–) (Fig. 2). These compounds bound to SUR1 with ϳ1,000ϫ higher affinity than the short-chain sulfonylureas, and it was hypothesized that the binding pocket of gliben- clamide comprises two parts (or subsites): site A, which accommodates the old (short-chain sulfonylurea) part of the compound, and site B for the new carboxamido part (12). This “two-sites hypothesis” of glibenclamide binding was supported by the finding that meglitinide, the carbox- amido-part of glibenclamide with the negative charge provided by benzoic acid (Fig. 2), also blocked the chan- nel, albeit with lower potency. New glinides in clinical use or development are repaglinide (a benzoic acid deriva- tive), nateglinide (a D-phenylalanine derivative), and mitiglinide (a 3-phenylpropionic acid derivative) (Fig. 2). The sulfonylureas and glinides have in common the nega- tive charge and the central phenyl ring, suggesting that the binding sites for a type A ligand such as tolbutamide and for a type B ligand such as meglitinide overlap to accom- modate these parts (12). The two-sites hypothesis of glibenclamide binding to SUR1 is now supported by structure-function analysis of cloned SURs. Exploiting the potency difference between tolbutamide blocking the Kir6.2/SUR1 and Kir6.2/SUR2A channels, Ashfield et al. (13) showed that transmembrane segments 14–16 were important for high-affinity block (Fig. 1). In particular, exchange of one amino acid, Ser 1237, in the cytoplasmic loop linking transmembrane segments 15 and 16, by Tyr (which is the corresponding amino acid in SUR2) abolished the block by tolbutamide, rendered the block by glibenclamide readily reversible, and abolished high-affinity [3H]glibenclamide binding to SUR1(S1237Y). However, the block by the benzoic acid derivative meglitinide was unaltered (13). This led to the hypothesis that the short-chain sulfonylureas bound to the region surrounding S1237 of SUR1 (i.e., site A) and that the bulkier Tyr residue in SUR2 resulted in steric hindrance of binding to this site; the binding site of meglitinide was assumed to be in another region of SUR (13). In the FIG. 2. Structure of sulfonylureas and glinides. The compounds are photoaffinity labeling studies leading to the cloning of aligned according to the position of the negative charge and in accor- 125 dance with their classification as type A, A ؉ B, or B ligands. Binding SUR1, it was found that [ I]glibenclamide labeled the site A contains in the cytosolic loop 8 in position 1237 Ser, which, in NH -terminal portion of SUR (rev. in 6), and Mikhailow et SUR2 corresponds to Tyr 1206 (mouse numbering). For drugs on the 2 market, the year of the market launch is given; for the other com- al. (14) showed that in SUR1, the third cytosolic loop pounds, the year of the first publication is given. (CL3), which links the transmembrane domains 0 and 1 (Fig. 1), was essential for [3H]glibenclamide binding. This may affect the cardiovascular outcome of type 2 diabetic suggested that CL3 was (an essential part of) the binding patients. region for the carboxamido part of glibenclamide (i.e., site B) (2,3). The region accommodating the negative charge THE SUR BINDING SITE(S) FOR SULFONYLUREAS AND remains unknown. GLINIDES Comparing the block of wild-type (Kir6.2/SUR1) and In the sulfonylureas (for selected structures, see Fig. 2), mutant (Kir6.2/SUR1(S1237Y)) channels by sulfonylureas and glinides has led to a classification of the secretagogues the nitrogen next to the sulfoxide is acidic with pKa values ranging from 5.3 (glibenclamide and tolbutamide) to 6.3 as interacting with sites A, B, or A ϩ B (Fig. 2). Most (glimepiride). Sulfonylureas act from the inside of the cell; interestingly, the glinides fall into two classes—either type they cross the cell membrane in the undissociated form, B ligands, like the benzoic acid derivatives meglitinide (13) but it is the negatively charged form that binds to SUR and repaglinide (15,16), or type A ligands, like nateglinide (11). The first-generation sulfonylureas were the short- (15,17) and mitiglinide (18). Molecular modeling has chain sulfonylureas such as tolbutamide, glibornuride, and shown that nateglinide resembles tolbutamide, whereas gliclazide. The second generation comprises the long- repaglinide resembles the carboxamido part of gliben- chain sulfonylureas such as glibenclamide and clamide; however, the piperidine ring of repaglinide pro- glimepiride, with a new aromatic moiety attached to the trudes out of the common pharmacophore (15). This short chain via a carboxamido group (–CONH–or residue could possibly make contacts with a region of SUR

S158 DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 U. QUAST AND ASSOCIATES

3 FIG. 3. Inhibition of [ H]glibenclamide binding to Kir6.2/SUR1 and Kir6.2/SUR2A KATP channels by glibenclamide, nateglinide, and repaglinide in intact HEK cells at 37°C. For transient transfection, the mouse clones were used with the exception of SUR1, which was the rat clone. Inhibition curves are presented as percent total binding (% BTOT) and are the mean of three to four individual experiments. Nonspecific binding was determined in the presence of glibenclamide (100 nmol/l, Kir6.2/SUR1) or P1075 (100 ␮mol/l, Kir6.2/SUR2A), respectively. Data were analyzed using the logistic form of the Hill equation. IC50 values were corrected for the presence of the radioligand according to Cheng-Prusoff to give the ϳ Ki values listed in Table 1 (20,21). The correction amounted to factors of 5 and 1.6 for Kir6.2/SUR1 and Kir6.2/SUR2A, respectively. The corrected inhibition curves are represented by the dotted curves and the pKi values are indicated.

(or Kir6.2) not reached by glibenclamide or the other HEK cells was inhibited by glibenclamide, with an IC50 compounds. value of 43 nmol/l (value at 37°C) (20). Regarding the binding step, data for the interaction of

KATP CHANNEL SUBTYPE SELECTIVITY secretagogues with Kir6.2/SUR1 are available (2); how-

The KATP channel subtype selectivity of the insulin secre- ever, little is known about the binding to the cardiovascu- tagogues has been extensively studied in electrophysiolog- lar channels. Due to the lower affinity of these channels for ical experiments mostly using recombinant channels glibenclamide, binding studies were generally performed expressed in Xenopus oocytes (rev. in 2); some data from using the opener [3H]P1075 as the radioligand. However, isolated organs are also available (19). Among the glinides, [3H]P1075 senses the binding of glibenclamide to SUR2 by nateglinide (17) and mitiglinide (18) were shown to be negative allosteric interactions and, in such experiments, ϳ1,000ϫ selective for the Kir6.2/SUR1 over the Kir6.2/ the true affinity of SUR2 for glibenclamide (and probably SUR2 channels. This exquisite selectivity is in line with the for the other secretagogues, too) is underestimated classification of these secretagogues as type A ligands and (20,21). For these reasons we have studied here the is in sharp contrast to the benzoic acid derivatives, meg- binding of selected sulfonylureas and glinides to the litinide and repaglinide, which are type B ligands and do Kir6.2/SUR1 and Kir6.2/SUR2A channels, using [3H]gliben- not discriminate much between the channel subtypes clamide as the radioligand. Experiments were performed (2,16). For repaglinide, half-maximal inhibitory concentra- at 37°C in intact HEK cells transiently expressing the tion (IC50) values of 7.4, 8.7, and 10 nmol/l were reported respective channels as described earlier (20,21). for inhibition of the channels formed by Kir6.2 with SUR1, Figure 3 shows the inhibition of [3H]glibenclamide bind- SUR2A, and SUR2B, respectively, in azide-treated intact ing to the channels by glibenclamide, nateglinide, and Xenopus oocytes; in inside-out patches in the absence of repaglinide; the results for these and the other compounds nucleotides, IC50 values of 5.6, 2.2, and 2.0 nmol/l were are listed in Table 1. Figure 4 compares the affinities of the obtained for inhibition of the respective channels (16). An secretagogues for the channels. For the four compounds intermediate selectivity was found for the long-chain sul- binding to sites A (glibornuride and nateglinide) and A ϩ fonylureas as type A ϩ B ligands (rev. in 2). For gliben- B (glibenclamide and glimepiride), correlation analysis ϭ Ϯ clamide, IC50 values of 0.13, 45, and 42 nmol/l were gave a straight line with slope 1 at a distance of d 1.29 determined in isolated patches from COS cells transfected 0.07 below the line of identity (Fig. 4). This gives a mean with Kir6.2 ϩ SUR1, 2A, and 2B, respectively; in macro- selectivity (calculated as 10Ϫd) of 19 (14–27) of these patches from Xenopus oocytes. IC50 values were 4 and 27 compounds for the pancreatic over the KATP channels in nmol/l for the Kir6.2/SUR1 and/SUR2A channels, respec- the sarcolemma of cardiomyocytes (sarcKATP channels). tively (2). The whole-cell current passing through the Generally, these results agree with the selectivity ratios recombinant vascular channel Kir6.1/SUR2B expressed in determined electrophysiologically, with the exception of

DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 S159 CARDIAC KATP CHANNELS AND INSULINOTROPES

channel has the same composition as that in vascular myocytes (i.e., Kir6.1/SUR2B). Under resting conditions, the coronary vasculature of several species responds to glibenclamide with vasocon- striction and a reduction of coronary blood flow (5). The importance of KATP channels for the regulation of coro- nary tone is increased in various pathological states (e.g., in hyperlipidemic pigs [26] and in dogs with alloxan- induced diabetes [27], left ventricular hypertrophy [28], or pacing-induced heart failure [29]). To knock out the vas- cular channel, both Kir6.1Ϫ/Ϫ (25) and SUR2Ϫ/Ϫ mice (30) were generated. Both phenotypes showed spontaneous ischemic ST elevations in the electrocardiogram, arterio- venous (AV) blocks of various degrees, and, finally, asys- tole (i.e., sudden cardiac death) (25,30) due to transient coronary vasospasms (30). This phenotype resembles Prinzmetal (vasospastic) angina in humans (25,30), and it suggests that, in humans, loss-of-function mutations in the Kir6.1 or SUR2B genes may genetically predispose for Prinzmetal angina. In patients with coronary artery dis- ␮ FIG. 4. Comparison of the affinities (pKi values) of secretagogue ease, infusion of glibenclamide (40 g/min) in a coronary binding to Kir6.2/SUR1 and Kir6.2/SUR2A channels. Data are from conduit artery decreased vessel diameter by 7.2%, in- Table 1. The broken line represents the line of identity (no selectivity); the solid line gives the correlation line for the four compounds binding creased coronary resistance by 28%, and reduced flow by to sites A (glibornuride [GLIB] and nateglinide [NAT]) and A؉B 18%; higher doses were not given in order to avoid (glibenclamide [GBC] and glimepiride [GLIM]) (F). The correlation confounding effects on insulin levels and blood glucose ؍ had a slope of 1.06 ؎ 0.04 and a mean distance of d (0.999 ؍ line (r ؊1.29 ؎ 0.07 from the line of identity. Pure type B ligands are (31). However, another study showed that at therapeutic represented by the open circles (E). doses of glibenclamide (50 ␮g/kg i.v.), insulin levels were increased but coronary blood flow and vasodilator re- sponse to adenosine and papaverine were unchanged in the relatively low ratios for the pure type A ligands, poststenotic and angiographically normal coronary arter- glibornuride and nateglinide. For glibenclamide, the bind- ies of patients with coronary artery disease (32). ing study gave a selectivity ratio of 14, which is in During exercise, coronary blood flow increases. In ex- agreement with that obtained for the comparison with the ercising swine and dog, adenosine, NO, and glibenclamide- ϭ recombinant vascular KATP channel, Kir6.1/SUR2B (Ki sensitive KATP channels are major players in recruiting 5.8 nmol/l) (20). For repaglinide, the lack of selectivity coronary reserve. In various species (5), including the found in the electrophysiological studies (16) is confirmed human (33), the reactive hyperemia following transient (Table 1, Fig. 4). Regarding the three type B ligands cardiac ischemia and hypoxia is strongly glibenclamide Ϫ [repaglinide, meglitinide, and ( )AZ-DF 265], Fig. 4 shows sensitive. In pressurized human coronary arterioles dis- that they lie near, above, or below the line of identity. This sected from right atrial appendages, hypoxia hyperpolar- indicates that ligands for the B site can exhibit some ized the vascular smooth muscle cells and induced selectivity for either the cardiac or the pancreatic channel vasodilation in an endothelium-independent manner; glib- with selectivity ratios varying from 0.26 for meglitinide enclamide (1 ␮mol/l) completely reversed the hyperpolar- ϫ Ϫ (i.e., 4 cardioselective) to 6.8 for ( )AZ-DF 265. The ization and reduced the vasodilator response by 65% (33). relative cardioselectivity of meglitinide makes a modest In coronary arterioles from patients with type 1 and type 2 contribution to the exquisite selectivity (Ͼ10,000) of HMR diabetes, the responses to hypoxia and to the KATP channel 1883 for the cardiovascular over the pancreatic KATP opener aprikalim were greatly attenuated (33). Under the channels (7). proviso that these observations in atrial arterioles are representative for the coronary resistance vessels in gen- FUNCTION OF KATP CHANNELS IN THE CORONARY eral, these results indicate that the vascular KATP channel VASCULAR BED IN NORMOXIA AND ISCHEMIA plays a major role in the autoregulation of coronary blood The functionally important KATP channel in vascular myo- flow in the human heart in hypoxia and that the function of cytes is Kir6.1/SUR2B. In several vascular beds, it is this channel is impaired in patients with type 1 or type 2 opened by local or circulating vasodilator transmitters (5), diabetes (33). and its opening leads to vasorelaxation by a variety of pathways (22). In addition, there is a K channel in ATP K CHANNELS IN CARDIAC MYOCYTES: FUNCTION IN endothelial cells consisting of SUR2B ϩ Kir6.1 and/or ATP Kir6.2 (23). Opening of this channel hyperpolarizes the THE ADAPTIVE RESPONSE TO STRESS AND ROLES IN endothelial cell and increases Ca2ϩ entry (24), which in ISCHEMIA turn promotes the synthesis of nitric oxide (NO) and other Experiments with the Kir6.2(Ϫ/Ϫ) mouse have shown that vasorelaxant factors and increases the permeability of the functional sarcKATP channels (Kir6.2/SUR2A) are required vascular wall (23,24). That pinacidil (a KATP channel for adaptation of the heart to stress (34). In severe opener) did not decrease blood pressure in the Kir6.1 ischemia and hypoxia, the decrease in the ratio of ATP to knockout mouse (25) argues that the endothelial KATP ADP (rev. in 35) and the increase in oleoyl-coenzyme A

S160 DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 U. QUAST AND ASSOCIATES

ester levels (36) lead to a massive opening of sarcKATP analyzed (39). In patients treated with “sulfonylureas of all channels. Simultaneously, a variety of tissue hormones kinds,” the ischemia-induced ST elevation was reduced (among them adenosine) are released, which triggers the and these patients were significantly less likely to meet the process of ischemic preconditioning. This mechanism was standard ECG criteria for admission to thrombolytic ther- first linked to the opening of the sarcKATP channel, later to apy (39). The small number of patients did not allow the the opening of the mitoKATP channel, and then to the authors to take into account the differing selectivity ratios opening of both channels (9,10). In the following, the of the insulin secretagogues for the KATP channel subtypes physiological consequences of the opening of these chan- (nor the different plasma half-lives of the drugs). In an nels are discussed with emphasis on the human heart. accompanying commentary, Brady and Jovanovic (40) pointed out several limitations of the study. In the light of sarcKATP CHANNELS AND ADAPTATION OF THE this new evidence, however, they recommended to discon- HEART TO STRESS tinue sulfonylurea treatment in type 2 diabetic patients Until recently, it was thought that under physiological with suspected acute coronary syndrome and to infuse insulin instead, if necessary, until myocardial ischemia conditions, sarcKATP channels had no function since the high ATP concentrations in the healthy cadiomyocyte kept was ruled out (40); this opinion is expressed frequently them locked in the closed state. This notion changed (see CLINICAL PERSPECTIVES). dramatically when it was found that Kir6.2(Ϫ/Ϫ) mice The second point of concern is that functional sarcKATP were much less tolerant to physical and sympathetic stress channels are required for the adaptation of the heart to stress. In addition, recent experiments in mice have clearly than wild-type mice (34). The sarcKATP channel–deficient mice performed poorly in the treadmill exercise test. shown that opening of the sarcKATP channel plays a Under vigorous ␤-adrenergic stimulation, these animals central role in ischemic preconditioning, at least in this died from arrhythmia; cardiomyocytes showed a dimin- species (1,10). ished reduction in action potential duration, Ca2ϩ over- load, and myocardial contraction bands (34). Such ISCHEMIC PRECONDITIONING: sarcKATP AND phenomena may not be restricted to the mouse with its mitoK CHANNELS high heart rate (Ͼ600 bpm); in humans, defective regula- ATP Prolonged cardiac ischemia leads to the death of cardio- tion of channel activity due to mutations in SUR2A has myocytes by necrosis and apoptosis. Ischemic precondi- been linked to the occurrence of dilated cardiomyopathy tioning (i.e., the protection that a short ischemic event, or (37). stimulus, affords against a prolonged ischemic insult and subsequent reperfusion) occurs in two temporal windows. OPENING OF sarcK CHANNELS IN CARDIAC ATP The early phase lasts 1–3 h after the stimulus (classic HYPOXIA AND ISCHEMIA: PATHOPHYSIOLOGICAL AND preconditioning); the delayed phase (second window) of CLINICAL IMPLICATIONS protection appears ϳ18 h later and lasts for up to 3 days The opening of sarcKATP channels in cardiac hypoxia and (9,10). The triggers for the two phases of ischemic precon- ischemia reduces action potential duration and clamps the ditioning are essentially the same: release of adenosine, cardiomyocyte at the potassium equilibrium potential, bradykinin, norepinephrine, and opioids. NO, however, is rendering the cell unexcitable. Whereas this may salvage more effective in triggering the delayed response. Great ATP and preserve the structural integrity of the cell, it also progress has been made in delineating the complex signal- increases the electrical heterogeneity of the heart and ing chains that mediate the two windows of precondition- promotes reentry arrhythmias (35). In addition, a pro- ing (9,10). Work in Kir6.2Ϫ/Ϫ mice demonstrated that longed opening of sarcKATP channels leads to the accumu- opening of the sarcKATP channel (perhaps via protein lation of extracellular Kϩ in the ischemic zone, depolarizes kinases C [PKC] and an adenosine pathway [10]) is a the cell, and induces cytotoxic Ca2ϩ entry (35). Along necessary step in the first phase of preconditioning in this these lines, prevention of the opening of these channels in species (1); in wild-type mice, pharmacological analysis ischemia was shown to protect against ischemia-induced showed that the delayed phase of ischemic precondition- ventricular fibrillation (35). It may be considered a proof ing is also mediated via sarcKATP channel opening (10). of concept that the selective sarcKATP channel blocker, There is, however, overwhelming pharmacological evi- HMR 1883, afforded effective protection against sudden dence also for a mitochondrial pathway of ischemic pre- cardiac death in rats, pigs, and dogs subjected to cardiac conditioning (9,10,41). This is mainly based on the ischemia after coronary artery ligation (7). differential effects of the “mitoKATP channel–selective” Using the selective block of sarcKATP channels as an blocker 5-hydroxydecanoate versus the sarcKATP channel– antiarrhythmic strategy in cardiac ischemia, however, selective blocker HMR 1883 or on the effects of the raises two points of concern. First, the opening of “mitoKATP channel–selective” opener diazoxide (9,41). The sarcKATP channels in cardiac ischemia leads to heteroge- selectivity of 5-hydroxydecanoate and diazoxide for the neity of the plateau phase of myocardial action potentials mitoKATP channel is, however, not established beyond and induces the deviation (elevation or depression) of the doubt (41). At reasonable concentrations, both substances ST-segment of the electrocardiogram observed in cardiac target also proteins other than KATP channels, and the very ischemia (38). Accordingly, in animals with experimental existence of the mitoKATP channel has been called into cardiac ischemia, pretreatment with sulfonylureas or HMR question (42). On the other hand, an opener with improved 1883 (7,35) blunted the deviation. In a recent retrospective selectivity for the mitoKATP channel has been character- study, the electrocardiogram (ECG) charts from 88 pa- ized recently (43). Whatever the pathway, mitochondria tients with type 2 diabetes and myocardial infarction were play an essential part in ischemic preconditioning

DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 S161 CARDIAC KATP CHANNELS AND INSULINOTROPES

(9,10,41). Unfortunately, little is known about the ability of over the cardiovascular KATP channels. Hence, some of the clinically used insulin secretagogues to interfere with these drugs may inadvertently reduce coronary blood flow ischemic preconditioning with the exception of gliben- at rest (31) and in hypoxia (33). In addition, they may clamide and glimepiride. In the Langendorff-perfused rat interfere with the ability of the heart to adapt to stress heart, glibenclamide (10 ␮mol/l) abolished ischemic pre- (34), with ischemic preconditioning (44–48), and with the conditioning completely, whereas glimepiride (10 ␮mol/l) diagnostically important electrocardiographic ST eleva- was ineffective (44). tion in acute myocardial infarction (39). Ischemic preconditioning also occurs in the human Two large-scale prospective and randomized clinical heart. The most direct evidence stems from patients studies have addressed the question of whether chronic undergoing coronary artery bypass grafting. If prolonged treatment of type 2 diabetic patients with sulfonylureas global ischemia (by aortic clamping) or cardioplegic arrest affects cardiovascular outcome. The University Group were preceded by a preconditioning stimulus and reperfu- Diabetes Program (UGDP) study in one arm tested tolbu- sion, the ischemic damage as judged by troponin C release tamide and provided evidence for an increased risk of or decrease in ATP content in cardiac muscle was allevi- cardiovascular mortality of ϳ1% per year as compared ated (9). Experiments in human ventricular myocytes and with the other arms (51). This result is difficult to reconcile atrial trabeculae have provided evidence for both the early with the high selectivity of tolbutamide for the pancreatic and delayed phases of ischemic preconditioning in the KATP channel, and arguments were raised that the study human heart and for the involvement of the mitochondrial may be flawed (52). However, the effect of tolbutamide on pathway in these phenomena (9). As an example, recovery the mitoKATP channel is unknown, and an interference of contractile force of atrial trabeculae obtained from with ischemic preconditioning could explain the increase nondiabetic patients or from diabetic patients taking insu- in cardiovascular mortality. The U.K. Prospective Diabetes lin was augmented by ischemic preconditioning; however, Study (UKPDS) has shown that there was no difference in ischemic preconditioning was ineffective when applied to cardiovascular outcome between the groups assigned to trabeculae from type 2 diabetic patients taking gliben- intensive treatment with glibenclamide, chlorpropamide, clamide (2ϫ5 mg/day; n ϭ 6) or glipizide (10 mg/day; n ϭ or insulin (53). Hence, chlorpropamide (a short-chain 1). This suggests that chronic treatment with these sulfo- nylureas abolishes ischemic preconditioning (45). Ample sulfonylurea) and glibenclamide seem to be acquitted. but indirect evidence for ischemic preconditioning in However, concern about glibenclamide and sulfonylureas patients with coronary artery disease is provided by in general continues to be raised. As an example, one phenomena such as the “warm-up” or “first effort” angina small-scale study with type 2 diabetic patients suggested (i.e., the conditioning of the heart by a first exercise- that sulfonylureas increase the risk of in-hospital mortality induced angina attack for subsequent exercise) or by the after coronary angioplasty for acute myocardial infarction fact that a first angina attack limits the consequences of a (54). Another study suggested that patients taking sulfo- following infarction if the time window between the two nylureas have an increased incidence of coronary events events is 24–72 h (9). A large body of literature treats the after a prior myocardial infarction compared with patients effects of repeated balloon inflation in patients undergoing treated with insulin, metformin, or diet alone (55). Based coronary angioplasty. If the duration of the conditioning on studies of this kind, it is often recommended to inflation exceeded 60–90 s, indicators of myocardial isch- discontinue sulfonylurea treatment in clinical situations emia such as chest pain, ST-segment elevation, or lactate such as acute coronary syndromes, coronary angioplasty, production were attenuated during subsequent inflations and coronary artery bypass grafting (29).). In contrast to (9). Pretreatment with glibenclamide was repeatedly these reports, a recent analysis of the UKPDS showed that shown to abolish or reduce this effect, whereas patients taking a sulfonylurea at the time of a myocardial glimepiride at equivalent hypoglycemic doses did not infarction had the same case fatality (51%) as those not (46–48). These studies must, however, be interpreted with taking a sulfonylurea (53%) (56). Hence, sulfonylureas do caution since ST-segment elevation was used as the only definitely not increase case fatality in this group of pa- (46,47) or a major (48) indicator of the ischemic burden. tients at high cardiovascular risk. As mentioned above, pretreatment with sulfonylureas may From a survey of the clinical literature, three points blur the ST-segment deviation in humans (39), and, at least emerge: First, in therapy with insulin secretagogues, much in rabbits, ST-segment elevation is a parameter dissociated emphasis is placed on a potential interference with isch- from ischemic preconditioning (49). These controversies emic preconditioning; however, the effect of most clini- show that ischemic preconditioning is difficult to quantify cally used insulin secretagogues on this phenomenon is in the clinical setting; in addition, aging reduces the unknown. Second, the interference of insulin secreta- effectiveness of preconditioning in animal models and, gogues with coronary blood flow is considered more rarely possibly, in patients (9). The effect of most clinically used (notable exceptions are refs. 31–33,50,55). The dramatic insulin secretagogues on ischemic preconditioning in the phenotype of the mice lacking the vascular KATP channel human heart remains unknown. (25,30) may stimulate more work on the channel in the human coronary circulation. Third, in many small-scale studies, all insulin secretagogues were put in the same CLINICAL PERSPECTIVES basket, regardless of their KATP channel subtype selectiv- Type 2 diabetic patients have an increased risk of cardio- ity; often the names of the prescribed drugs could not be vascular complications (50). They are generally treated retrieved. It has been hypothesized that subtype selectivity with insulin secretagogues, and, as outlined above, these is of importance for the cardiovascular outcome of dia- drugs differ greatly in their selectivity for the pancreatic betic patients (50); however, randomized prospective stud-

S162 DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 U. QUAST AND ASSOCIATES ies addressing this point are lacking. At this point, it glibenclamide binding site of the ␤-cell K-ATP channel. FEBS Lett 499: remains unknown whether the selectivity ratios of the 154–160, 2001 15. Hansen AM, Christensen IT, Hansen JB, Carr RD, Ashcroft FM, Wahl P: insulin secretagogues determined for the recombinant Differential interactions of nateglinide and repaglinide on the human KATP channels reflect the selectivity for the native human beta-cell sulphonylurea receptor 1. Diabetes 51:2789–2795, 2002 channels in physiological and pathophysiological condi- 16. Dabrowski M, Wahl P, Holmes WE, Ashcroft FM: Effect of repaglinide on cloned beta cell, cardiac and smooth muscle types of ATP-sensitive tions. In the cardiac myocyte, the sarcKATP channel is physically associated with adenylate (57) and creatine potassium channels. Diabetologia 44:747–756, 2001 kinase (58), thereby facilitating delivery of metabolic 17. Chachin M, Yamada M, Fujita A, Matsuoka T, Matsushita K, Kurachi Y: Nateglinide, a D-phenylalanine derivative lacking either a sulfonylurea or signals from the mitochondria to the plasma membrane ␤ benzamido moiety, specifically inhibits pancreatic -cell-type KATP chan- (57,58). In the diabetic state, vascular KATP channel func- nels. J Pharmacol Exp Ther 304:1025–1032, 2003 tion is depressed (33), and the sympathetic surge that 18. Reimann F, Proks P, Ashcroft FM: Effects of mitiglinide (S 21403) on patients generally experience at the time of a myocardial Kir6.2/SUR1, Kir6.2/SUR2A and Kir6.2/SUR2B types of ATP-sensitive po- infarction may dramatically change the state of the cardio- tassium channel. Br J Pharmacol 132:1542–1548, 2001 19. Hu S, Wang S, Dunning BE: Tissue selectivity of antidiabetic agent vascular K channels (e.g., by phosphorylation). Clearly, ␤ ATP nateglinide: study on cardiovascular and -cell KATP channels. J Pharma- these points are not reflected in the in vitro experiments col Exp Ther 291:1372–1379, 1999 using recombinant channels. Hence, in view of the widely 20. Russ U, Hambrock A, Artunc F, Lo¨ ffler-Walz C, Horio Y, Kurachi Y, Quast differing in vitro selectivity ratios of repaglinide and nateg- U: Coexpression with the inward rectifier Kϩ channel Kir6.1 increases the linide, a clinical study comparing these two glinides in affinity of the vascular sulfonylurea receptor SUR2B for glibenclamide. Mol Pharmacol 56:955–961, 1999 type 2 diabetic patients with coronary artery disease 21. Hambrock A, Lo¨ ffler-Walz C, Russ U, Lange U, Quast U: Characterization of would be of great interest for both therapeutic and merely a mutant sulfonylurea receptor SUR2B with high affinity for sulfonylureas scientific reasons. and openers: differences in the coupling to Kir6.x subtypes. Mol Pharmacol 60:190–199, 2001 ϩ ϩ ACKNOWLEDGMENTS 22. Quast U: Do the K channel openers relax smooth muscle by opening K channels. Trends Pharmacol Sci 14:332–337, 1993 U.Q. is supported by the Deutsche Forschungsgemein- 23. Mederos y Schnitzler M, Derst C, Daut J, Preisig-Mu¨ ller R: ATP-sensitive schaft grant Qu 100/3-1. We are grateful to Drs. Y. Kurachi potassium channels in capillaries isolated from guinea-pig heart. J Physiol and Y. Horio (Osaka) for the generous gift of the murine 525:307–317, 2000 clones of SUR2A, SUR2B, and Kir6.2; to Dr. C. Derst 24. Lu¨ ckhoff A, Busse R: Calcium influx into endothelial cells and formation of endothelium-derived relaxing factor is controlled by the membrane poten- (Freiburg) for the rat clone of SUR1; and to several tial. Pflu¨gers Arch 416:305–311, 1990 pharmaceutical companies for the generous gift of drugs. 25. Miki T, Suzuki M, Shibasaki T, Uemura H, Sato T, Yamaguchi K, Koseki H, The technical assistance of P. Ku¨ hner is gratefully Iwanaga T, Nakaya H, Seino S: Mouse model of Prinzmetal angina by acknowledged. disruption of the inward rectifier Kir6.1. Nat Med 8:466–472, 2002 26. Mathew V, Lerman A: Altered effects of potassium channel modulation in the coronary circulation in experimental hypercholesterolemia. Athero- REFERENCES sclerosis 154:329–335, 2001 1. Seino S, Miki T: Physiological and pathophysiological roles of ATP- 27. Posa I, Kocsis E, Nieszner T, Pogatsa G, Koltai MZ: Haemodynamic and sensitive Kϩ channels. Prog Biophys Mol Biol 81:133–176, 2003 metabolic effects of low daily dose sulphonylureas in diabetic dog hearts. 2. Proks P, Reimann F, Green N, Gribble F, Ashcroft F: Sulfonylurea Arzneim-Forsch/Drug Res 52:552–559, 2002 ϩ stimulation of insulin secretion. Diabetes 51 (Suppl. 3):S368–S376, 2002 28. Melchert PJ, Duncker DJ, Traverse JH, Bache RJ: Role of KATP channels 3. Gribble FM, Reimann F: Sulphonylurea action revisited: the post-cloning and adenosine in regulation of coronary blood flow in the hypertrophied era. Diabetologia 46:875–891, 2003 left ventricle. Am J Physiol 277:H617–H625, 1999 4. Hasselblatt A: Sulfonylureas: pharmacokinetics in animal experiments. In 29. Yamamoto M, Egashira K, Arimura K, Tada H, Shimokawa H, Takeshita A: Handbook of Experimental Pharmacology 119: Oral Antidiabetics. Kuhl- Coronary vascular Kϩ ATP channels contribute to the maintenance of mann J, Puls W, Eds. Berlin/Heidelberg, Springer Verlag, 1996, p. 161–183 myocardial perfusion in dogs with pacing-induced heart failure. Jpn Circ

5. Brayden JE: Functional roles of KATP channels in vascular smooth muscle. J 64:701–707, 2000 Clin Exp Pharmacol Physiol 29:312–316, 2002 30. Chutkow WA, Pu J, Wheeler MT, Wada T, Makielski JC, Burant CF, 6. Aguilar-Bryan L, Bryan J: Molecular biology of adenosine triphosphate- McNally EM: Episodic coronary artery vasospasm and hypertension de-

sensitive potassium channels. Endocrine Rev 20:101–135, 1999 velop in the absence of Sur2 KATP channels. J Clin Invest 110:203–208, 7. Go¨ gelein H, Englert HC, Kotzan A, Hack R, Lehr K-H, Seiz W, Becker RHA, 2002 Sultan E, Scho¨ lkens BA, Busch AE: HMR 1098: an inhibitor of cardiac 31. Farouque HM, Worthley SG, Meredith IT, Skyrme-Jones RA, Zhang MJ: ATP-sensitive potassium channels. Cardiovasc Drug Rev 18:157–174, 2000 Effect of ATP-sensitive potassium channel inhibition on resting coronary 8. Nielsen FE, Bodvarsdottir TB, Worsaae A, MacKay P, Stidsen CE, Boonen vascular responses in humans. Circ Res 90:231–236, 2002 HC, Pridal L, Arkhammar PO, Wahl P, Ynddal L, Junager F, Dragsted N, 32. Reffelmann T, Klues HG, Hanrath P, Schwarz ER: Post-stenotic coronary Tagmose TM, Mogensen JP, Koch A, Treppendahl SP, Hansen JB: 6-Chloro- blood flow at rest is not altered by therapeutic doses of the oral 3-alkylamino-4H-thieno[3,2-e]-1,2,4-thiadiazine 1,1-dioxide derivatives po- antidiabetic drug glibenclamide in patients with coronary artery disease. tently and selectively activate ATP sensitive potassium channels of Heart 87:54–60, 2002 pancreatic beta-cells. J Med Chem 45:4171–4187, 2002 33. Miura H, Wachtel RE, Loberiza FR, Saito T, Miura M, Nicolosi AC, 9. Yellon DM, Downey JM: Preconditioning the myocardium: from cellular Gutterman DD: Diabetes mellitus impairs vasodilation to hypoxia in physiology to clinical cardiology. Physiol Rev 83:1113–1151, 2003 human coronary arterioles: reduced activity of ATP-sensitive potassium

10. Gross GJ, Peart JN: KATP channels and myocardial preconditioning: an channels. Circ Res 92:151–158, 2003 update. Am J Physiol 285:H921–H930, 2003 34. Zingman LV, Hodgson DM, Bast PH, Kane GC, Perez-Terzic C, Gumina RJ, 11. Schwanstecher M, Schwanstecher C, Dickel C, Chudziak F, Moshiri A, Pucar D, Bienengraeber M, Dzeja PP, Miki T, Seino S, Alekseev AE, Terzic Panten U: Location of the sulphonylurea receptor at the cytoplasmic face A: Kir6.2 is required for adaptation to stress. Proc Natl Acad SciUSA of the beta-cell membrane. Br J Pharmacol 113:903–911, 1994 99:13278–13283, 2002 12. Brown GR, Foubister AJ: Receptor binding sites of hypoglycemic sulfo- 35. Go¨ gelein H: Inhibition of cardiac ATP-dependent potassium channels by nylureas and related [(acylamino)alkyl]benzoic acids. J Med Chem 27:79– sulfonylurea drugs. Curr Op Investigational Drugs 2:72–80, 2001 81, 1984 36. Liu GX, Hanley PJ, Ray J, Daut J: Long-chain acyl-coenzyme A esters and

13. Ashfield R, Gribble FM, Ashcroft SJH, Ashcroft FM: Identification of the fatty acids directly link metabolism to KATP channels in the heart. Circ Res high-affinity tolbutamide site on the SUR1 subunit of the KATP channel. 88:918–924, 2001 Diabetes 48:1341–1347, 1999 37. Bienengraeber M, Olson TM, Selivanov VA, Kathmann EC, O’Cochlain F, 14. Mikhailov MV, Mikhailova EA, Ashcroft SJH: Molecular structure of the Gao F, Karger AB, Ballew JD, Hodgson D, Zingman LV, Pang Y-P, Alekseev

DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004 S163 CARDIAC KATP CHANNELS AND INSULINOTROPES

AE, Terzic A: ABCC9 mutations identified in human dilated cardiomyopa- 48. Lee T-M, Chou T-F: Impairment of myocardial protection in type 2 diabetic

thy disrupt catalytic KATP channel gating. Nat Genet 36:382–387, 2004 patients. J Clin Endocrinol Metab 88:531–537, 2003 38. Li RA, Leppo M, Miki T, Seino S, Marban E: Molecular basis of electrocar- 49. Birincioglu M, Yang XM, Critz SD, Cohen MV, Downey JM: S-T segment diographic ST-segment elevation. Circ Res 87:837–839, 2000 voltage during sequential coronary occlusions is an unreliable marker of 39. Huizar JF, Gonzalez LA, Alderman J, Smith HS: Sulfonylureas attenuate preconditioning. Am J Physiol 277:H2435–H2441, 1999 electrocardiographic ST-segment elevation during an acute myocardial 50. Wascher TC, Boes U: Ischemia in type 2 diabetes: tissue selectivity of infarction in diabetics. J Am Coll Cardiol 42:1017–1021, 2003 sulfonylureas and clinical implications. Metabolism 52:3–5, 2003 40. Brady PA, Jovanovic A: The sulfonylurea controversy: much ado about 51. Meinert CL, Knatterud GL, Prout TE, Klimt CR: A study of the effects of nothing or cause for concern? J Am Coll Cardiol 42:1022–1025, 2003 hypoglycemic agents on vascular complications in patients with adult- onset diabetes. II. Mortality results. Diabetes 19 (Suppl. 2):789–830, 1970 41. O’Rourke B: Myocardial KATP channels in preconditioning. Circ Res 87:845–855, 2000 52. Kilo C, Miller JP, Williamson JR: The crux of the UGDP. Spurious results 42. Das M, Parker JE, Halestrap AP: Matrix volume measurements challenge and biologically inappropriate data analysis. Diabetologia 18:179–185, the existence of diazoxide/glibencamide-sensitive K channels in rat 1980 ATP 53. UK Prospective Diabetes Study (UKPDS) Group: Intensive blood-glucose mitochondria. J Physiol 547:893–902, 2003 control with sulphonylureas or insulin compared with conventional treat- 43. Grover GJ, D’Alonzo AJ, Darbenzio RB, Parham CS, Hess TA, Bathala MS: ment and risk of complications in patients with type 2 diabetes (UKPDS In vivo characterization of the mitochondrial selective K opener (3R)- ATP 33). Lancet 352:837–853, 1998 trans-4-((4-chlorophenyl)-N-(1H-imidazol-2-ylmethyl)dimethyl-2H-1- 54. Garratt KN, Brady PA, Hassinger NL, Grill DE, Terzic A, Holmes D-RJ: benzopyran-6-carbonitril monohydrochloride (BMS-191095): cardioprotec- Sulfonylurea drugs increase early mortality in patients with diabetes tive, hemodynamic, and electrophysiological effects. J Pharmacol Exp mellitus after direct angioplasty for acute myocardial infarction. J Am Coll Ther 303:132–140, 2002 Cardiol 33:119–124, 1999 44. Mocanu MM, Maddock HL, Baxter GF, Lawrence CL, Standen NB, Yellon 55. Aronow WS, Ahn C: Incidence of new coronary events in older persons DM: Glimepiride, a novel sulfonylurea, does not abolish myocardial with diabetes mellitus and prior myocardial infarction treated with sulfo- protection afforded by either ischemic preconditioning or diazoxide. nylureas, insulin, metformin, and diet alone. Am J Cardiol 88:556–557, Circulation 103:3111–3116, 2001 2001 45. Cleveland J-CJ, Meldrum DR, Cain BS, Banerjee A, Harken AH: Oral 56. Stevens RJ, Coleman RL, Adler AI, Stratton IM, Matthews DR, Holman RR: sulfonylurea hypoglycemic agents prevent ischemic preconditioning in Risk factors for myocardial infarction case fatality and stroke case fatality human myocardium: two paradoxes revisited. Circulation 96:29–32, 1997 in type 2 diabetes: UKPDS 66. Diabetes Care 27:201–207, 2004 46. Tomai F, Crea F, Gaspardone A, Versaci F, De Paulis R, Penta-de-Peppo A, 57. Carrasco AJ, Dzeja PP, Alekseev AE, Pucar D, Zingman LV, Abraham MR, Chiariello L, Gioffre PA: Ischemic preconditioning during coronary angio- Hodgson D, Bienengraeber M, Puceat M, Janssen E, Wieringa B, Terzic A: ϩ plasty is prevented by glibenclamide, a selective ATP-sensitive K channel Adenylate kinase phosphotransfer communicates cellular energetic signals blocker. Circulation 90:700–705, 1994 to ATP-sensitive potassium channels. Proc Natl Acad SciUSA98:7623– 47. Klepzig H, Kober G, Matter C, Luus H, Schneider H, Boedeker KH, Kiowski 7628, 2001 W, Amann FW, Gruber D, Harris S, Burger W: Sulfonylureas and ischaemic 58. Crawford RM, Ranki HJ, Botting CH, Budas GR, Jovanovic A: Creatine preconditioning; a double-blind, placebo-controlled evaluation of kinase is physically associated with the cardiac ATP-sensitive Kϩ channel glimepiride and glibenclamide. Eur Heart J 20:439–446, 1999 in vivo. FASEB J 16:102–104, 2002

S164 DIABETES, VOL. 53, SUPPLEMENT 3, DECEMBER 2004