Download 5-Hydroxydecanoate and coenzyme A are inhibitors of native

Biochimica et Biophysica Acta 1800 (2010) 385–391
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a g e n
5-Hydroxydecanoate and coenzyme A are inhibitors of native sarcolemmal KATP
channels in inside-out patches
Xiantao Li a, Markus Rapedius b, Thomas Baukrowitz b, Gong Xin Liu c, D.K. Srivastava d,
Jürgen Daut a,⁎, Peter J. Hanley e,⁎
a
Institut für Normale und Pathologische Physiologie, Philipps-Universität Marburg, Deutschhausstrasse 2, 35037 Marburg, Germany
Institute of Physiology II, Friedrich Schiller University Jena, Teichgraben 8, 07743 Jena, Germany
Department of Medicine, Brown Medical School, Providence, RI 02908, USA
d
Department of Biochemistry and Molecular Biology, North Dakota State University, Fargo, ND 58105, USA
e
Institut für Physiologie II, Westfälische Wilhelms-Universität Münster, Robert-Koch-Strasse 27b, 48149 Münster, Germany
b
c
a r t i c l e
i n f o
Article history:
Received 29 July 2009
Received in revised form 11 November 2009
Accepted 12 November 2009
Available online 18 November 2009
Keywords:
KATP channel
Preconditioning
Single-channel current
Myocyte
a b s t r a c t
Background: 5-Hydroxydecanoate (5-HD) inhibits preconditioning, and it is assumed to be a selective
inhibitor of mitochondrial ATP-sensitive K+ (mitoKATP) channels. However, 5-HD is a substrate for
mitochondrial outer membrane acyl-CoA synthetase, which catalyzes the reaction: 5 HD + CoA + ATP → 5HD-CoA (5-hydroxydecanoyl-CoA) + AMP + pyrophosphate. We aimed to determine whether the reactants
or principal product of this reaction modulate sarcolemmal KATP (sarcKATP) channel activity.
Methods: Single sarcKATP channel currents were measured in inside-out patches excised from rat ventricular
myocytes. In addition, sarcKATP channel activity was recorded in whole-cell configuration or in giant insideout patches excised from oocytes expressing Kir6.2/SUR2A.
Results: 5-HD inhibited (IC50 ∼ 30 μM) KATP channel activity, albeit only in the presence of (non-inhibitory)
concentrations of ATP. Similarly, when the inhibitory effect of 0.2 mM ATP was reversed by 1 μM oleoyl-CoA,
subsequent application of 5-HD blocked channel activity, but no effect was seen in the absence of ATP.
Furthermore, we found that 1 μM coenzyme A (CoA) inhibited sarcKATP channels. Using giant inside-out
patches, which are weakly sensitive to “contaminating” CoA, we found that Kir6.2/SUR2A channels were
insensitive to 5-HD-CoA. In intact myocytes, 5-HD failed to reverse sarcKATP channel activation by either
metabolic inhibition or rilmakalim.
General significance: SarcKATP channels are inhibited by 5-HD (provided that ATP is present) and CoA but
insensitive to 5-HD-CoA. 5-HD is equally potent at “directly” inhibiting sarcKATP and mitoKATP channels.
However, in intact cells, 5-HD fails to inhibit sarcKATP channels, suggesting that mitochondria are the
preconditioning-relevant targets of 5-HD.
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
Ischemic and pharmacological preconditioning of the heart has been
shown to extend the time that cardiac muscle cells can survive during a
subsequent prolonged period of ischemia [1,2]. It has been suggested
that the various protocols used for preconditioning activate an
endogenous cytoprotective program in cardiac myocytes. In view of
the potential clinical importance, the mechanisms underlying the
cardioprotective action of preconditioning have been extensively
studied. It is generally believed that mitochondrial ATP-sensitive K+
(mitoKATP) channels play an important role, either as the initial trigger
or as the end effector (or both), of a complex signaling cascade leading to
⁎ Corresponding authors. J. Daut is to be contacted at tel.: +49 6421 2866494; fax:
+49 6421 2868960. P.J. Hanley, tel.: +49 83 251 55336; fax: +49 83 251 55331.
E-mail addresses: [email protected] (J. Daut), [email protected]
(P.J. Hanley).
0304-4165/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbagen.2009.11.012
cardioprotection. The evidence for the involvement of mitoKATP
channels was mainly derived from pharmacological interventions.
Preconditioning can be mimicked by application of diazoxide or various
other K+ channel openers, and it is often assumed that mitoKATP
channels are the main site of action of these drugs in cardiac muscle.
However, it cannot be excluded that other targets of these drugs, for
example, sarcolemmal KATP (sarcKATP) channels [3] or succinate
dehydrogenase [4,5], may play a major role in preconditioning.
5-Hydroxydecanoate (5-HD) inhibits virtually all forms of preconditioning, and therefore elucidation of its mechanism of action could be
a key to understanding cytoprotective signal transduction [6]. Initially,
5-HD was reported by Notsu et al. in 1992 to block sarcKATP channel
currents recorded in guinea ventricular myocytes or inside-out patches
[7,8]. However, work in the late 1990s using isolated rat heart
mitochondria and K+ flux measurements [9] or ventricular myocytes
and combined whole-cell current and flavoprotein fluorescence
recordings [10] indicated that 5-HD selectively inhibited mitoKATP
386
X. Li et al. / Biochimica et Biophysica Acta 1800 (2010) 385–391
channels. Subsequently, 5-HD became established as a selective mitoKATP channel inhibitor, used to probe the role of mitoKATP versus sarcKATP
channels in preconditioning studies [6,11].
In 2002, independent groups reported that 5-HD is metabolized
[5,12]. After entering cells, 5-HD is thioester linked to CoA, generating
5-hydroxydecanoyl-CoA (5-HD-CoA). Subsequent work revealed that
5-HD-CoA is taken up by mitochondria, metabolized via the βoxidation pathway, and metabolites of the D-isomer may impair the
oxidation of endogenous fatty acids [13,14]. The metabolism of 5-HD
complicates the interpretation of experiments using this putative
blocker. For example, the flavoprotein fluorescence signals used as
readout for mitoKATP channel opening in earlier studies were
performed in glucose-free medium [10]. The supply of the metabolic
substrate 5-HD (medium-chain fatty acid) to glucose starved cells,
rather than mitoKATP channel inhibition, could explain how 5-HD
induces flavoprotein reduction, as recently demonstrated [15].
Moreover, “metabolic rescue” by 5-HD could explain the data of
Notsu et al. [7,8], who activated sarcKATP channel currents by
specifically blocking glucose metabolism.
Here, we readdressed the question of whether 5-HD inhibits
cardiac sarcKATP channels. The activation of 5-HD via acyl-CoA
synthetase (5-HD + CoA + ATP → 5-HD-CoA + AMP + PPi) on the
mitochondrial outer membrane introduces 5-HD-CoA as a potential
sarcKATP channel opener since acyl-CoA esters are now known to
antagonize the inhibition of sarcKATP channels by ATP [16]. We tested
the hypothesis that 5-HD blocks, and 5-HD-CoA opens, sarcKATP
channels, in analogy to oleate and oleoyl-CoA [16]. We found that 5HD inhibited single sarcKATP channels in inside-out patches in an ATPdependent fashion, such that no effect was observed in ATP-free
medium. Importantly, 5-HD inhibited single-channel activity when
oleoyl-CoA was used to prevent channel rundown. Surprisingly, we
found that CoA was a potent inhibitor of native sarcKATP channel
activity. Purified 5-HD-CoA had no effect on sarcKATP channel activity,
tested in Xenopus oocytes expressing ventricle-type KATP channels
(Kir6.2/SUR2A) [17-19] to obviate the confounding effects of
contaminating CoA. Furthermore, although 5-HD inhibits sarcKATP
channels directly, we show that 5-HD fails to reverse sarcKATP
channels activated by metabolic inhibition or a KATP channel opener
in intact myocytes.
myocytes were dispersed into a modified “KB medium” [20] for 1
h before resuspending the cells in Dulbecco's modified Eagle's
medium (Invitrogen, Germany), as previously reported [5]. Insideout patches were excised from myocytes using borosilicate glass
pipette with tip resistances of 4–6 MΩ. Single-channel currents were
recorded using an Axopatch 200B amplifier (Axon Instruments, Foster
City, CA). The pipette solution, as well as the bath (intracellular)
solution, contained (mM): 140 KCl, 10 HEPES and 5 EGTA (pH 7.3).
2.3. Expression of Kir6.2/SUR2A in oocytes
Murine Kir6.2 and rat SUR2A, the subunits of cardiac-type sarcKATP
channels [17-19], were kindly provided by Dr. F. M. Ashcroft. The
constructs were subcloned into the pBF expression vector [21] and
capped cRNAs were synthesized in vitro using SP6 polymerase
(Promega, Heidelberg, Germany). The constructs were stored at
−70 °C. Oocytes were surgically dissected from Xenopus laevis frogs,
which were anesthetized with 0.4% 3-aminobenzoic acid ethyl ester.
Around 50 nl of cRNA stock solution specific for Kir6.2 and SUR2A
subunits was injected into Dumont stage VI oocytes. The oocytes were
treated with 0.5 mg/ml collagenase type II (Sigma), defolliculated and
incubated at 19 °C for 1–3 days before use.
2.4. Giant patch recordings
Pipettes were made from borosilicate glass (tip resistances, 0.2–0.4
MΩ) and filled with (mM): 120 KCl, 10 HEPES and 1.8 CaCl2 (pH
adjusted to 7.2 with KOH). Giant patches were excised from oocytes
3–7 days after cRNA injection and recordings were made in the insideout configuration under voltage-clamp conditions. The standard bath
(intracellular) solution contained (mM): 120 KCl, 10 HEPES, 1 sodium
pyrophosphate and 2 K2EGTA (pH 7.2). Solutions were applied to the
cytosolic side of the giant patch by means of a multi-barrel pipette
system. Currents were evoked by repeatedly applying 20 mV (200
ms) and −90 mV (100 ms) pulses from a holding potential of −70 mV
and were recorded using an EPC9 amplifier (HEKA Electronics,
Lamprecht, Germany). All recordings of native or expressed KATP
channels were performed at room temperature (22 °C).
2.5. Whole-cell recordings in ventricular myocytes
2. Materials and methods
The animal protocols were approved by the ethics committees of
the Universities of Marburg and Jena and conformed to the Guide for
the Care and Use of Laboratory Animals published by the US National
Institutes of Health (NIH Publication No. 85-23, revised 1996).
2.1. Preparation of 5-HD-CoA
5-HD-CoA was synthesized and purified as previously described
[13]. In brief, 5-HD-CoA was enzymatically synthesized from 5-HD
(RBI, Sigma) and CoA using acyl-CoA synthetase in solution containing
(mM): 1.2 Na2ATP, 100 Tris, 9.1 EDTA and 1.8 EDTA (pH adjusted to
7.5 with HCl). 5-HD-CoA was purified to N95% by preparative HPLC,
performed with a Waters HPLC system fitted with an Alltech C18column (0.1 cm × 25 cm, 10 μm, Econosil). A gradient increase of
methanol in 20 mM potassium phosphate solution (pH 7.0) was used
to elute reaction products. Methanol was subsequently removed by
rotary evaporation and 5-HD-CoA was stored at −70 °C until required.
Electrospray ionization and mass spectrometry were used to identify
5-HD-CoA [5].
2.2. Single-channel recordings from myocytes
Rats were anesthetized with 6–8% sevoflurane and the heart was
rapidly excised. The perfused heart was digested with collagenase and
Whole-cell recordings were made using an Axon Instruments
200B amplifier and pClamp8 software. To induce metabolic inhibition
cells were superfused with glucose-free solution containing (mM): 10
2-deoxy-D-glucose (inhibitor of hexokinase), 150 NaCl, 5.4 KCl, 1.2
MgCl2 and 10 Hepes/NaOH (pH 7.4). The pipette solution contained
(mM): 140 KCl, 5 NaCl, 5 EGTA, 2 MgCl2, 0.2 Na2ATP and 10 Hepes/
KOH (pH 7.2). Voltage ramps were applied between −80 and + 60 mV.
In experiments using the KATP channel opener rilmakalim, the bath
solution contained (mM): 140 KCl, 10 NaCl, 1.2 MgCl2 and 10 Hepes/
KOH (pH 7.4). Voltage-clamp recordings were made between −30
and + 60 mV using 10-mV steps.
3. Results
3.1. 5-HD inhibits sarcolemmal KATP channels
The inhibitory effect of 0.2 mM ATP on KATP channel activity could
be reversed by superfusing the inside-out patch with 1 μM oleoyl-CoA
(Fig. 1A), in accord with the previous observations [16]. However,
when the same patch was challenged with 100 μM 5-HD, channel
activity was reversibly inhibited, as shown in Fig. 1B, and consistently
observed in four excised patches. Interestingly, the potent inhibitory
effect of 5-HD was not observed when experiments were performed
in the complete absence of ATP, as shown in the example of Fig. 1C and
summarized in Fig. 1D.
X. Li et al. / Biochimica et Biophysica Acta 1800 (2010) 385–391
387
Fig. 1. Inhibition of native sarcKATP channels by 5-HD. (A) Voltage-clamp recording of single KATP channel activity in a membrane patch excised from a rat ventricular myocyte
(holding potential −40 mV). In the presence of 0.2 mM ATP, the channel was inhibited, whereas introduction of 1 μM oleoyl-CoA reversed the ATP-induced inhibition. (B) In the
same patch as in (A), 100 μM 5-HD reversibly inhibited sarcKATP channel activity in the presence of 0.2 mM ATP and 1 μM oleoyl-CoA. (C) In the absence of ATP (0 mM ATP), sarcKATP
channels were insensitive to 5 HD. (D) Summary of data for 0 mM ATP (n = 4 patches) and 0.2 mM ATP (n = 4 patches) showing that 5-HD-induced inhibition of sarcKATP channel
activity (indexed as open probability, NPo) is ATP-dependent (⁎P b 0.05).
5-HD has been reported to block native mitoKATP channels with an
IC50 value of 95 μM based on measurements of flavoprotein
fluorescence [10,22], an approach open to misinterpretations [5,15].
In more direct assays using isolated rat cardiac mitochondria or liver
mitoKATP channels reconstituted in liposomes, Jabůrek et al. [9] found
that 5-HD inhibited K+ fluxes with IC50 values in the range 45–85 μM.
To determine the potency of sarcKATP channel inhibition, we applied
various concentrations of 5-HD directly to patches superfused with a
non-inhibitory concentration of ATP (1 μM). A typical experiment is
shown in Fig. 2A. Application of 5-HD decreased the open probability
of single sarcKATP channels with an IC50 value of 28.7 μM in the
presence of 1 μM ATP (Fig. 2B). At higher ATP concentrations, the
apparent sensitivity of 5-HD-mediated block was increased (Fig. 2C).
3.2. Coenzyme A is a potent inhibitor of native sarcKATP channels
In intact cells, 5-HD is thioester linked to CoA via mitochondrial outer
membrane acyl-CoA synthetase, producing 5-HD-CoA. We therefore
tested whether CoA or 5-HD-CoA alone could modulate sarcKATP
channel activity. Excised patches were initially challenged with 20 μM
CoA, which strongly blocked the channels (n = 3; not shown). We then
tested the effect of 1 μM CoA, which reversibly decreased the Po of single
sarcKATP channels from 0.84 ± 0.03 to 0.13± 0.05 (n = 4; P b 0.05), as
shown in the example in Fig. 3A. Thus, CoA is a potent inhibitor of native
sarcKATP channels in inside-out patches.
3.3. Sarcolemmal KATP channels are insensitive to 5-HD-CoA
5-HD-CoA was synthesized enzymatically and purified to N95% by
preparative HPLC (see Materials and methods). Using electrospray
ionization and mass spectrometry, we could detect two dominant
peaks corresponding to 5-HD-CoA ionized by H+ (m/z ratio ∼ 938)
and K+ (m/z ratio ∼ 976), respectively (Fig. 3B). Analytical HPLC
showed that 5-HD-CoA preparations additionally contained 3–4% CoA
(peak corresponding to an m/z ratio of 768 in the electrospray
ionization mass spectrometry analysis; Fig. 3B). We found that our
preparation of 5-HD-CoA (20 μM) inhibited native sarcKATP channels
(n = 3; not shown), but the presence of contaminating CoA (0.6–0.8
μM) could account for this effect. To circumvent the confounding
effects of contaminating CoA, we tested the effects of the acyl-CoA
ester 5-HD-CoA in oocytes expressing cardiac-type KATP channels,
which are weakly sensitive to CoA when assessed by the giant patch
technique [23].
388
X. Li et al. / Biochimica et Biophysica Acta 1800 (2010) 385–391
Fig. 3. Coenzyme A inhibits native sarcKATP channels. (A) Patch-clamp recording of a
single sarcKATP channel reversibly inhibited by application of 1 μM CoA. (B) Using
electrospray ionization mass spectrometry, 5-HD-CoA (synthesized and purified by
HPLC) could be identified, but contaminating CoA was also detected.
3.4. 5-HD does not reverse sarcKATP channel opening by rilmakalim or
metabolic inhibition
Fig. 2. Concentration–response relation of 5-HD-induced sarcKATP channel inhibition.
(A) In the presence of 1 μM ATP, which alone did not inhibit sarcKATP channel activity,
the introduction of 5-HD reduced channel open probability. (B) The inhibitory effect of
5 HD (in the presence of 1 μM ATP) was concentration-dependent with an IC50 value of
28.7 μM (each data point was obtained from at least 3 patches). (C) At higher ATP
concentrations, the apparent potency of 5-HD-mediated sarcKATP channel inhibition
was increased (⁎P b 0.05 compared to 1 μM ATP group; n = 4 patches).
Fig. 4 shows typical recordings, representative of 6–7 patches for
each experimental protocol, of the effects of 5-HD-CoA and decanoylCoA on Kir6.2/SUR2A channel activity in giant patches excised from
Xenopus oocytes. Application of 20 μM 5-HD-CoA to the cytosolic
side of the inside-out patch, either before (not shown) or after (Fig.
4A) treatment with inhibitory concentrations of ATP (50–5000 μM),
did not inhibit inward currents, recorded during −90-mV pulses from
a holding potential of −70 mV. Similar to 5-HD-CoA, 20 μM decanoylCoA also did not inhibit KATP (Kir6.2/SUR2A) channels (not shown;
n = 6). In contrast to oleoyl-CoA, neither 5-HD-CoA nor decanoyl-CoA
could reverse channels after rundown, as clearly illustrated in Fig. 4B.
5-HD is commonly used to block preconditioning in intact cells or
tissues. Therefore, we investigated whether 5-HD inhibits sarcKATP
channel currents in intact cells evoked by either a KATP channel opener
(rilmakalim) or inhibition of glucose metabolism. The application of
the KATP channel opener rilmakalim (1 μM) induced robust outward
current (measured 1.5–2 min after application) at positive potentials,
consistent with KATP channel activation (Fig. 5A). Subsequent
introduction of 100 μM 5-HD, applied for 5 min, did not reverse the
current (n = 3). 5-HD (applied for at least 5 min) also failed to reverse
sarcKATP channels opened by blocking glucose metabolism (n = 4), as
shown in Fig. 5B. To elicit sarcKATP channel opening via metabolic
inhibition, it was necessary to apply glucose-free medium plus 2deoxy-D-glucose (DOG) for 10–15 min.
4. Discussion
4.1. 5-HD-induced inhibition of sarcKATP versus mitoKATP channels
The original demonstration that 5-HD inhibits ATP-dependent K+
uptake into intact rat heart and liver mitochondria by Jabůrek et al. [9]
was presented by the authors as a “crucial piece of evidence” for the
hypothesis that mitoKATP channels are integrally involved in
cardioprotection. Since that time, the picture has become more
complex. It has been clearly shown that 5-HD is rapidly activated to 5-
X. Li et al. / Biochimica et Biophysica Acta 1800 (2010) 385–391
389
Fig. 4. 5-HD-CoA has no effect on cardiac-type KATP channels in giant patches excised from oocyte. (A) Inward KATP channel (Kir6.2/SUR2A) currents were evoked by repeated pulses
to −90 mV from a −70 mV holding potential in giant inside-out patches excised from Xenopus oocytes (inward currents are depicted as upward deflections). The current was
reversibly inhibited by incremental ATP concentrations as indicated, whereas application of 20 μM 5-HD-CoA had no effect. Note that there is gradual rundown of current during the
course of the experiment. (B) Following rundown, neither 5-HD-CoA nor decanoyl-CoA evoked KATP channel current, whereas oleoyl-CoA (positive control) fully restored KATP
channel activity.
HD-CoA by acyl-CoA synthetase on the mitochondrial outer membrane [5,12], and it is subsequently metabolized by rat heart and liver
mitochondria, albeit the D-isomer is less efficiently metabolized than
the L-isomer [14]. We now show that 5-HD can inhibit cardiac-type
sarcKATP channels when it is directly applied to the cytosolic face of
membrane patches excised from rat ventricular myocytes. Moreover,
we found that 5-HD inhibited sarcKATP channels with an IC50 value of
∼30 μM, which is similar to the IC50 range (45–85 μM) reported for
mitoKATP channels [9], although the latter values were determined in
the presence of K+ channel openers or GTP. Similar to Jabůrek et al.
[9], we also found that 5-HD had no inhibitory effect in the absence of
ATP. Hence, mitochondria express pharmacologically overlapping
sarcKATP-like channels or, alternatively, the isolated mitochondria
used by Jabůrek et al. [9] were contaminated with sarcolemmal
membrane. In favor of the former, Zhang et al. [24] reconstituted
bovine heart mitoKATP channels in lipid bilayers and measured singlechannel activity, which was inhibited by ATP and 5-HD, but not the
sarcKATP channel antagonist HMR-1098. One of the major challenges
in the field is to identify the molecular components of the channel
recorded by Zhang et al. [24].
In contrast to our observations, Light et al. [25] reported that 5-HD
(10–500 μM) had no effect on the open probability of KATP channels
excised from rat ventricular myocytes in the presence of 10 μM ATP.
The reason for this discrepancy is not clear. We could observe a clear
5 HD-induced decrease in open probability in patches containing a
single KATP channel under conditions in which channels were
activated by a combination of 0.2 mM ATP and 1 μM oleoyl-CoA.
This observation is important since oleoyl-CoA prevents the potentially confounding effects of channel rundown [16], a phenomenon
whereby the open probability of KATP channels steadily decreases
following patch excision.
4.2. Mechanism of sarcKATP channel inhibition by 5-HD
The negatively charged lipid oleoyl-CoA, as well as phosphatidylinositol 4,5-bisphosphate, decrease the sensitivity of sarcKATP
channels to the inhibitory ligand ATP [26-28]. The ability of 5-HD to
override the potent stimulatory effect of oleoyl-CoA is surprising and
it suggests that this medium-chain fatty acid either impedes the
interaction of oleoyl-CoA with the channel protein or allosterically
modulates sensitivity to inhibitory ATP. Another clue to the
mechanism of action of 5-HD at sarcKATP channels, as well as mitoKATP
channels [9], is the absolute requirement for the second ligand ATP.
This is not completely surprising since the efficacy of most KATP
channel openers and inhibitors is highly dependent on the binding of
nucleotide ligands such as ATP, ADP and UDP [29-31]. In particular,
the binding of KATP channel openers to the SUR subunit requires Mg2+
and ATP, the latter of which binds to a high affinity site (b5 μM) [31].
Diazoxide is an unusual case since it was originally thought to be a
selective mitoKATP channel opener in the heart, but subsequently it
was shown that diazoxide activates sarcKATP channels in the presence
of the second ligand ADP [3,32]. The binding of ATP to a high affinity
nucleotide binding domain of the SUR2A subunit probably induces a
conformational state, which allows 5-HD to directly interact with and
inhibit the channel. In the absence of ATP, another open state
configuration exists, which is inaccessible to 5-HD.
4.3. Lack of effect of 5-HD on sarcKATP channel currents in intact myocytes
In a study using whole-cell recordings, 5-HD has been shown to
inhibit KATP channels composed of Kir6.1/SUR1 or Kir6.2/SUR1, but
not Kir6.2/SUR2A [22]. Kir6.1/SUR1 is the neuronal and pancreatic βcell KATP channel [31], whereas Kir6.2/SUR1 has recently been shown
to be the predominant surface KATP channel in atrial cells [19]. The
reason why 5-HD directly inhibits sarcKATP channels in the inside-out,
but not the whole-cell, configuration is not clear. The finding, though,
is consistent with the previous report that 5-HD does not block
pinacidil-induced KATP-induced currents [10]. The unusual feature in
the study of Sato et al. [10] was that repeated application of pinacidil
produced a much larger current, possibly a consequence of the
substrate-free medium (metabolic inhibition) used to facilitate
flavoprotein fluorescence responses. In the same experiments, 5-HD
reversed pinacidil-induced flavoprotein oxidation, emphasizing the
point that mitochondrial actions of 5-HD manifest when it is applied
extracellularly to intact cells. We speculate that 5-HD fails to inhibit
sarcKATP channels in intact cells due to rapid thioester linking to CoA
390
X. Li et al. / Biochimica et Biophysica Acta 1800 (2010) 385–391
reaction, long-chain fatty acids and ATP are inhibitors of KATP
channels, whereas the principal reaction products, long-chain acylCoA esters, are potent channel openers [16,23,26-28,35]. We found
that CoA alone is a potent inhibitor of sarcKATP channel activity, which
reinforces and extends the link between long-chain fatty acid
metabolism and membrane excitability. On a structural basis, it is
not unexpected that CoA interacts with sarcKATP channels since it
contains an ADP moiety, albeit with the addition of a 3′-phosphate
group on the ribose ring, arising from the linking of adenosine 3′,5′bisphosphate to phosphorylated pantothenic acid (vitamin B5).
Similar to the regulatory role of the ATP/ADP ratio, our data suggest
that the cytosolic CoA to long-chain acyl-CoA ratio is another
important determinant of sarcKATP channel activity. The total
concentration of “cytosolic” CoA in cardiac tissue is estimated to be
about 14 μM, and its distribution between free and long-chain acyl
derivatives varies widely [35,36]. During ischemia, the CoA to acylCoA ratio decreases [35,36], which would favor the activation of
sarcKATP channels.
In analogy to long-chain fatty acids [16,37], we speculated that the
acyl-CoA ester 5-HD-CoA may stimulate sarcKATP channels and
thereby mask the inhibitory action of the free fatty acid. To test
whether 5-HD-CoA stimulated or inhibited cardiac-type KATP channels, we switched to an oocyte expression system since channels
formed by heterologous expression of Kir6.2 and SUR2A in this system
are largely insensitive to the concentrations of free CoA contaminating
our 5-HD-CoA preparations [23]. In giant excised patches, neither 5HD-CoA nor decanoyl-CoA reversed channel rundown, whereas
oleoyl-CoA dramatically recovered channel function. Furthermore,
introduction of 5-HD-CoA did not inhibit channel activity in giant
excised patches. Hence, unlike long-chain acyl-CoA esters, the C10
(medium chain) esters decanoyl-CoA and 5-HD-CoA are probably too
short to allow the head groups to modulate sarcKATP channel gating
[26].
4.5. Conclusions
Fig. 5. 5-HD does not reverse sarcKATP channel currents activated in intact ventricular
myocytes. (A) Whole-cell recordings using symmetrical K+ (pipette to bath ratio ∼ 1:1).
Application of 100 μM 5-HD (red) fails to reverse sarcKATP channel currents activated by
the KATP channel opener rilmakalim (1 μM). (B) Whole-cell recording using
asymmetrical K+ (pipette to bath ratio ∼ 25:1). Application of 100 μM 5-HD (red)
fails to reverse sarcKATP channel currents activated by metabolic inhibition (glucosefree medium and 2-deoxy-D-glucose).
The findings of our study are illustrated schematically in Fig. 6. We
found that 5 HD inhibits sarcKATP (IC50 ∼ 30 μM) when it is directly
applied to the cytosolic face, even in the presence of oleoyl-CoA (which
via acyl-CoA synthetase on the mitochondrial outer membrane. The
“scavenging” of the endogenous sarcKATP channel inhibitor CoA may
further promote channel activation.
We anticipated that 5-HD would act as metabolic fuel for the βoxidation pathway and reverse sarcKATP channel activation induced
by specific inhibition of glucose metabolism. This was not the case.
However, we have previously shown that 5-HD, compared to
decanoate, is not a good substrate for isolated heart mitochondria,
and the D-isomer may in fact inhibit the β-oxidation pathway [14].
Hence, 5-HD is probably not sufficiently metabolized to support
energy demand in the absence of glycolysis. Another possibility is that
sarcKATP channels activated by metabolic inhibition are generally
more resistant to pharmacological inhibitors. For example, Rainbow et
al. [33] reported that the inhibitory action of the selective sarcKATP
channel blocker HMR-1098 is severely impaired during metabolic
inhibition. In addition, the KATP channel blocker glibenclamide is less
effective when channels are activated by metabolic inhibition,
possibly due to increased cytosolic ADP levels [33,34].
4.4. Inhibition of native KATP channels by coenzyme A
Acyl-CoA synthetase is emerging as a pivotal enzyme controlling
the activity of sarcKATP channels in intact cells. On the left of the
Fig. 6. 5-HD and CoA inhibit sarcKATP channels. The putative mitoKATP-selective blocker
5 HD is activated by acyl-CoA synthetase (ACS) on the mitochondrial outer membrane.
All three reactants of the ACS-catalyzed reaction (5-HD, CoA and ATP) inhibit sarcKATP
channels. Unlike long-chain acyl-CoA esters, the substituted medium-chain acyl-CoA
ester 5-HD-CoA has no stimulatory effect on sarcKATP channel activity. 5-HA-CoA is
metabolized in the mitochondrial matrix via the β-oxidation pathway.
X. Li et al. / Biochimica et Biophysica Acta 1800 (2010) 385–391
prevents channel rundown). Activated 5-HD, 5-HD-CoA, had no effect
on Kir6.2/SUR2A channel activity. As in the case of mitoKATP channels,
inhibition of sarcKATP channels by 5-HD requires the presence of ATP.
However, 5-HD fails to reverse sarcKATP channels activated in whole
cells by the KATP channel opener rilmakalim or inhibition of glycolysis.
Thus, our results suggest that mitochondrial targets of 5-HD or its
metabolites account for the negative effect of 5-HD on preconditioning. In addition, we identified CoA as a potent inhibitor of native
sarcKATP channels, suggesting a new paradigm in which the CoA to
long-chain acyl-CoA ratio controls sarcKATP channel gating.
References
[1] D.M. Yellon, J.M. Downey, Preconditioning the myocardium: from cellular
physiology to clinical cardiology, Physiol. Rev. 83 (2003) 1113–1151.
[2] P.J. Hanley, J. Daut, KATP channels and preconditioning: a re-examination of the
role of mitochondrial KATP channels and an overview of alternative mechanisms,
J. Mol. Cell. Cardiol. 39 (2005) 17–50.
[3] N. D'hahan, C. Moreau, A.-L. Prost, H. Jacquet, A.E. Alekseev, A. Terzic, M. Vivaudou,
Pharmacological plasticity of cardiac ATP-sensitive potassium channels toward
diazoxide revealed by ADP, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 12162–12167.
[4] G. Schäfer, C. Wegener, R. Portenhauser, D. Bojanovski, Diazoxide, an inhibitor of
succinate oxidation, Biochem. Pharmacol. 18 (1969) 2678–2681.
[5] P.J. Hanley, M. Mickel, M. Löffler, U. Brandt, J. Daut, KATP channel-independent
targets of diazoxide and 5-hydroxydecanoate in the heart, J. Physiol. 542 (2002)
735–741.
[6] H. Ardehali, B. O'Rourke, Mitochondrial KATP channels in cell survival and death,
J. Mol. Cell. Cardiol. 39 (2005) 7–16.
[7] T. Notsu, K. Ohhashi, I. Tanaka, et al., 5-Hydroxydecanoate inhibits ATP-sensitive
K+ channel currents in guinea-pig single ventricular myocytes, Eur. J. Pharmacol.
220 (1991) 35–41.
[8] T. Notsu, I. Tanaka, M. Takano, A. Noma, Blockade of the ATP-sensitive K+ channel
by 5-hydroxydecanoate in guinea pig ventricular myocytes, J. Pharmacol. Exp.
Ther. 260 (1992) 702–708.
[9] M. Jabůrek, V. Yarov-Yarovoy, P. Paucek, K.D. Garlid, State-dependent inhibition of
the mitochondrial KATP channel by glyburide and 5-hydroxydecanoate, J. Biol.
Chem. 273 (1998) 13578–13582.
[10] T. Sato, B. O'Rourke, E. Marbán, Modulation of mitochondrial ATP-dependent K+
channels by protein kinase C, Circ. Res. 83 (1998) 110–114.
[11] B. O'Rourke, Mitochondrial ion channels, Annu. Rev. Physiol. 69 (2007) 19–49.
[12] K.H. Lim, S.A. Javadov, M. Das, S.J. Clarke, M.S. Suleiman, A.P. Halestrap, The effects
of ischaemic preconditioning, diazoxide and 5-hydroxydecanoate on rat heart
mitochondrial volume and respiration, J. Physiol. 545 (2002) 961–974.
[13] P.J. Hanley, K.V. Gopalan, R.A. Lareau, D.K. Srivastava, M. von Meltzer, J. Daut, βOxidation of 5-hydroxydecanoate, a putative blocker of mitochondrial ATPsensitive potassium channels, J. Physiol. 547 (2003) 387–393.
[14] P.J. Hanley, S. Dröse, U. Brandt, et al., 5-Hydroxydecanoate is metabolised in
mitochondria and creates a rate-limiting bottleneck for β-oxidation of fatty acids,
J. Physiol. 562 (2005) 307–318.
[15] S. Dröse, U. Brandt, P.J. Hanley, K+-independent actions of diazoxide question the
role of inner membrane KATP channels in mitochondrial cytoprotective signaling,
J. Biol. Chem. 281 (2006) 23733–23739.
391
[16] G.X. Liu, P.J. Hanley, J. Ray, J. Daut, Long-chain acyl-coenzyme A esters and fatty
acids directly link metabolism to KATP channels in the heart, Circ. Res. 88 (2001)
918–924.
[17] A.P. Babenko, G. Gonzalez, L. Aguilar-Bryan, J. Bryan, Reconstituted human cardiac
KATP channels. Functional identity with the native channels from the sarcolemma
of human ventricular cells, Circ. Res. 83 (1998) 1132–1143.
[18] M. Suzuki, R.A. Li, T. Miki, et al., Functional roles of cardiac and vascular ATP-sensitive
potassium channels clarified by Kir6.2-knockout mice, Circ. Res. 88 (2001) 570–577.
[19] T.P. Flagg, H.T. Kurata, R. Masia, et al., Differential structure of atrial and ventricular
KATP. Atrial KATP channels require SUR1, Circ. Res. 103 (2008) 1458–1465.
[20] G. Isenberg, U. Klockner, Calcium tolerant ventricular myocytes prepared by
preincubation in a “KB medium”, Pflügers Arch. 395 (1982) 6–18.
[21] B. Fakler, U. Brändle, E. Glowatzki, S. Weidemann, H.P. Zenner, J.P. Ruppersberg,
Strong voltage-dependent inward-rectification of inward rectifier K+ channels is
caused by intracellular spermine, Cell 80 (1995) 149–154.
[22] Y. Liu, G. Ren, B. O'Rourke, E. Marbán, J. Seharaseyon, Pharmacological comparison
of native mitochondrial KATP channels with molecularly defined surface KATP
channels, Mol. Pharmacol. 59 (2001) 225–230.
[23] D. Schulze, M. Rapedius, T. Krauter, T. Baukrowitz, Long-chain acyl-CoA esters and
phosphatidylinositol phosphates modulate ATP inhibition of KATP channels by the
same mechanism, J. Physiol. 552 (2003) 357–367.
[24] D.X. Zhang, Y.-F. Chen, W.B. Campbell, A.-P. Zou, G.J. Gross, P.-L. Li, Characteristics
and superoxide-induced activation of reconstituted myocardial mitochondrial
ATP-sensitive potassium channels, Circ. Res. 89 (2001) 1177–1183.
[25] P.E. Light, H.D. Kanji, J.E.M. Fox, R.J. French, Distinct myoprotective roles of cardiac
sarcolemmal and mitochondrial KATP channels during metabolic inhibition and
recovery, FASEB J. 15 (2001) 2586–2594.
[26] M. Rapedius, M. Soom, E. Shumilina, et al., Long chain CoA esters as competitive
antagonists of phosphatidylinositol 4,5-bisphosphate activation in Kir channels,
J. Biol. Chem. 280 (2005) 30760–30767.
[27] D. Enkvetchakul, C.G. Nichols, Gating mechanism of KATP channels: function fits
form, J. Gen. Physiol. 122 (2003) 471–480.
[28] S.J. Tucker, T. Baukrowitz, How highly charged anionic lipids bind and regulate ion
channels, J. Gen. Physiol. 131 (2008) 431–438.
[29] P.A. Brady, A.E. Alekseev, A. Terzic, Operative condition-dependent response of cardiac
ATP-sensitive K+ channels toward sulfonylureas, Circ. Res. 82 (1998) 272–278.
[30] A. Hambrock, C. Löffler-Walz, U. Quast, Glibenclamide binding to sulphonylurea
receptor subtypes: dependence on adenine nucleotides, Br. J. Pharmacol. 136
(2002) 995–1004.
[31] C. Moreau, A.-L. Prost, R. Dérand, M. Vivaudou, SUR ABC proteins targeted by KATP
channel openers, J. Mol. Cell. Cardiol 38 (2005) 951–963.
[32] T. Matsuoka, K. Matsushita, Y. Katayama, et al., C-terminal tails of sulfonylurea
receptors control ADP-induced activation and diazoxide modulation of ATPsensitive K+ channels, Circ. Res. 87 (2000) 873–880.
[33] R.D. Rainbow, R.I. Norman, D. Hudman, N.W. Davies, N.B. Standen, Reduced
effectiveness of HMR 1098 in blocking cardiac sarcolemmal KATP channels during
metabolic stress, J. Mol. Cell. Cardiol. 39 (2005) 637–646.
[34] N. Venkatesh, S.T. Lamp, J.N. Weiss, Sulfonylureas, ATP-sensitive K+ channels, and
cellular K+ loss during hypoxia, ischemia, and metabolic inhibition in mammalian
ventricle, Circ. Res. 69 (1991) 623–637.
[35] J.F. Oram, J.I. Wenger, J.R. Neely, Regulation of long chain fatty acid activation in
heart muscle, J. Biol. Chem. 250 (1975) 73–78.
[36] J.A. Idell-Wenger, L.W. Grotyohann, J.R. Neely, Coenzyme A and carnitine
distribution in normal and ischemic hearts, J. Biol. Chem 253 (1978) 4310–4318.
[37] E. Shumilina, N. Klöcker, G. Korniychuk, M. Rapedius, F. Lang, T. Baukrowitz,
Cytoplasmic accumulation of long-chain coenzyme A esters activates KATP and
inhibits Kir2.1 channels, J. Physiol. 575 (2006) 433–442.