The KATP channel opener diazoxide protects cardiac myocytes during metabolic inhibition without causing mitochondrial depolarization or flavoprotein oxidation

Abstract

1. The K(ATP) channel opener diazoxide has been proposed to protect cardiac muscle against ischaemia by opening mitochondrial K(ATP) channels to depolarize the mitochondrial membrane potential, DeltaPsi(m). We have used the fluorescent dye TMRE to measure DeltaPsi(m) in adult rat freshly isolated cardiac myocytes exposed to diazoxide and metabolic inhibition. 2. Diazoxide, at concentrations that are highly cardioprotective (100 or 200 microM), caused no detectable increase in TMRE fluorescence (n=27 cells). However, subsequent application of the protonophore FCCP, which should collapse DeltaPsi(m), led to large increases in TMRE fluorescence (>300%). 3. Metabolic inhibition (MI: 2 mM NaCN+1 mM iodoacetic acid (IAA) led to an immediate partial depolarization of DeltaPsi(m), followed after a few minutes delay by complete depolarization which was correlated with rigor contracture. Removal of metabolic inhibition led to abrupt mitochondrial repolarization followed in many cells by hypercontracture, indicated by cell rounding and loss of striated appearance. 4. Prior application of diazoxide (100 microM) reduced the number of cells that hypercontracted after metabolic inhibition from 63.7+/-4.7% to 24.2+/-1.8% (P< 0.0001). 5-hydroxydeanoate (100 microM) reduced the protection of diazoxide (46.8+/-2.7% cells hypercontracted, P< 0.0001 vs diazoxide alone). 5. Diazoxide caused no detectable change in flavoprotein autofluorescence (n=26 cells). 6. Our results suggest that mitochondrial depolarization and flavoprotein oxidation are not inevitable consequences of diazoxide application in intact cardiac myocytes, and that they are also not essential components of the mechanism by which it causes protection.

Figure 1

Diazoxide does not alter ΔΨ_m indicated by TMRE fluorescence in isolated rat cardiac myocytes. (a) Recording of fluorescence from a single cardiac myocyte loaded with TMRE. Fluorescence was excited at 475 nm for 25 ms at 0.2 Hz, and emitted light at >520 nm was imaged as described in the Methods. In this and subsequent figures, fluorescence was measured for a region of the image that encompassed the whole cell, and is expressed as relative fluorescence ΔF/F_o (see Methods). Diazoxide (200 μM) and FCCP (5 μM) were applied as indicated. (b) Mean (+s.e.mean) data from 27 cells in experiments like that shown in (a). (c) Images showing a field of four cells before (i), and during (ii) the application of diazoxide and during the response to FCCP (iii). Note the lack of change in fluorescence in response to diazoxide in contrast to the large increase that occurred with FCCP. The timing of the images is indicated on the recording of panel (a).

Figure 2

The effect of metabolic inhibition on ΔΨ_m and cell length. (a) Recordings of TMRE fluorescence from two cardiac myocytes in response to superfusion with 2 mM NaCN and 1 mM iodoacetic acid (CN⁻+IAA), applied as indicated. (b) Corresponding changes in cell length, taken from the cell in (a) that gave the smaller fluorescence signal. The shortening to rigor occurs after substantial depolarization of ΔΨ_m has occurred, while following removal of metabolic inhibition near complete repolarization precedes shortening into hypercontracture. (c) Mean (+s.e.mean) cell lengths for 17 cells during rigor and after hyercontracture, normalised to the initial length in normal Tyrode for each cell. *P<0.001. (d) Images showing a field of cells before metabolic inhibition (i), in rigor (ii) and in hypercontracture after removal of CN⁻+IAA (iii). The timing of the images is indicated on the recordings of panels (a) and (b).

Figure 3

Separating the effects of cyanide and iodoacetic acid on ΔΨ_m. (a) Recording of TMRE fluorescence in a cardiac myocyte superfused with CN⁻ (2 mM) and FCCP (5 μM) as indicated. CN⁻ caused a small and reversible depolarization of ΔΨ_m, compared to complete depolarization by FCCP. (b) TMRE fluorescence from a myocyte showing the lack of effect of IAA (1 mM) alone, followed by depolarization of ΔΨ_m by CN⁻+IAA. (c) Relative fluorescence in a number of cells exposed to CN⁻, IAA, CN⁻+IAA and FCCP. Bars show mean (+s.e.mean) values from 32, 22, 48, 18 cells respectively.

Figure 4

Diazoxide protection against the effects of metabolic inhibition. Mean (+s.e.mean) data showing the effect of diazoxide pretreatment (100 μM) and diazoxide in the presence of 5-hydroxydecanoate (100 μM) on the percentage of hypercontracted cells 5 min after removal of metabolic inhibition. Results are from 13, 15 and eight experiments (138, 150 and 62 cells) in control, diazoxide, and diazoxide+5-HD respectively. *P<0.0001 compared to control. †P<0.0001 compared to diazoxide alone.

Figure 5

Diazoxide and flavoprotein autofluorescence. (a) Recording of flavoprotein (FAD) autofluorescence from a cardiac myocyte. Excitation was at 450 nm for 50 ms at 0.2 Hz, emitted light at >520 nm was imaged. Diazoxide (100 μM), CN⁻+IAA and dinitrophenol (DNP, 200 μM) were added as indicated. (b) Mean (+s.e.mean) data from 26, 32 and 23 cells exposed to diazoxide as in (a) and/or to CN⁻+IAA or DNP.