Carboxyatractyloside effects on brown-fat mitochondria imply that the adenine nucleotide translocator isoforms ANT1 and ANT2 may be responsible for basal and fatty-acid-induced uncoupling respectively

Abstract

In brown-fat mitochondria, fatty acids induce thermogenic uncoupling through activation of UCP1 (uncoupling protein 1). However, even in brown-fat mitochondria from UCP1-/- mice, fatty-acid-induced uncoupling exists. In the present investigation, we used the inhibitor CAtr (carboxyatractyloside) to examine the involvement of the ANT (adenine nucleotide translocator) in the mediation of this UCP1-independent fatty-acid-induced uncoupling in brown-fat mitochondria. We found that the contribution of ANT to fatty-acid-induced uncoupling in UCP1-/- brown-fat mitochondria was minimal (whereas it was responsible for nearly half the fatty-acid-induced uncoupling in liver mitochondria). As compared with liver mitochondria, brown-fat mitochondria exhibit a relatively high (UCP1-independent) basal respiration ('proton leak'). Unexpectedly, a large fraction of this high basal respiration was sensitive to CAtr, whereas in liver mitochondria, basal respiration was CAtr-insensitive. Total ANT protein levels were similar in brown-fat mitochondria from wild-type mice and in liver mitochondria, but the level was increased in brown-fat mitochondria from UCP1-/- mice. However, in liver, only Ant2 mRNA was found, whereas in brown adipose tissue, Ant1 and Ant2 mRNA levels were equal. The data are therefore compatible with a tentative model in which the ANT2 isoform mediates fatty-acid-induced uncoupling, whereas the ANT1 isoform may mediate a significant part of the high basal proton leak in brown-fat mitochondria.

Figure 1. Fatty acid uncoupling efficiency in brown-fat mitochondria and liver from wild-type and UCP1^−/− mice

Representative traces showing the effects of oleate (solid line) and FCCP (broken line) on respiration of liver mitohondria from UCP1^−/− mice (a) and brown-fat mitochondria from UCP1^−/− (b) and wild-type (c) mice. Pyr, addition of 5 mM pyruvate; GDP, addition of 1 mM GDP. FCCP and oleate were successively added to the concentrations indicated. (d) Dose–response curve for FCCP compiled from experiments as in (a– c). Values are means±S.E.M. from five to seven independent mitochondrial preparations. Curves were drawn for simple Michaelis–Menten kinetics. (e and f) Oleate concentration–response curves for brown-fat (solid line) and liver (broken line) mitochondria from wild-type and UCP1^−/− mice. The experiments were performed principally as those illustrated in (a– c). The brown-fat data set in (d– f) includes a few also included in [14]; however, all brown-fat experiments presented here (but not all in [14]) had been performed in parallel with liver preparations. In (f), the change in oxygen consumption (Δ nmol of O₂·min⁻¹·mg of protein⁻¹) from the coupled state (in the presence of 1 mM GDP) is shown. The points are means±S.E.M. for seven to nine independent mitochondrial isolations for each group. The x-axis indicates the free oleate concentrations calculated as described in the Experimental section. Curves were drawn for simple Michaelis–Menten kinetics. For ΔV_max and K_m calculations, concentration–response curves were individually analysed in each mitochondrial preparation. These results are shown in Table 1.

Figure 2. Representative traces showing the effect of CAtr on basal oxygen consumption of liver (a) and brown-fat (b) mitochondria from UCP1^−/− and mitochondria from wild-type (c and d) mice

Pyr, addition of 5 mM pyruvate; GDP, 1 mM GDP; CAtr, 1.0 μM CAtr; and FCCP, 1.0 μM FCCP.

Figure 3. Representative traces showing the effect of CAtr on oleate-stimulated oxygen consumption of liver (a) and brown-fat (b) mitochondria from UCP1^−/− and mitochondria from wild-type (c) mice

Oleate indicates 70 μM oleate, otherwise additions were the same as in Figure 2.

Figure 4. Effect of CAtr on oleate-stimulated oxygen consumption in liver and brown-fat mitochondria

(a) Oxygen consumption after successive additions of oleate to liver mitochondria in the presence or absence of 1 mM GDP. (b– d) Oxygen consumption after different concentrations of oleate (broken line) or oleate+CAtr (solid line) were added to liver mitochondria (b) and brown-fat mitochondria from UCP1^−/− (c) and wild-type (d) mice. The experiments were performed principally as those illustrated in Figure 3, except that in each experiment a different concentration of oleate was added (from 0 to maximally 100 μM). The values represent the means±S.E.M. of five independent liver, eight wild-type and eight UCP1^−/− brown-fat mitochondrial preparations. (e) CAtr effect on oleate-stimulated oxygen consumption. CAtr-reduced oxygen consumption was estimated as basal (in the presence of GDP) respiration or oleate-stimulated respiration minus CAtr-insensitive oxygen consumption (as performed in Figure 3) in the same mitochondrial preparation.

Figure 5. ANT protein levels in brown-fat and liver mitochondria isolated from UCP1^−/− and wild-type mice

(a) Representative Western-blot analysis of mitochondria (15 μg of mitochondrial protein/lane) isolated from BAT from wild-type and UCP1^−/− mice and from liver and skeletal muscle (SkM) from wild-type mice. Upper panel: ANT. Lower panel: COX1. (b) Quantification of ANT protein in mitochondria. The level of ANT protein was normalized to COX1 protein level and this ratio was set to 100% in wild-type brown-fat mitochondria. The values are the means±S.E.M. for independent mitochondrial preparations each analysed in duplicate (n=7 for wild-type brown-fat mitochondria, n=4 for UCP1^−/− brown-fat mitochondria and n=2 for liver mitochondria). **Significantly different (P<0.01) from UCP1-containing mitochondria (BAT^+/+).

Figure 6. Real-time relative quantitative RT-PCR analysis of transcripts of ANT isoforms in BAT and liver

mRNA levels of the Ant1 and Ant2 isoforms were analysed in liver and in BAT of wild-type (BAT^+/+) and UCP1^−/− (BAT^−/−) mice relative to 18 S rRNA levels. The values are the means±S.E.M. for independent cDNA preparations analysed in duplicate (n=4 for BAT samples and n=2 for liver samples), *P<0.05 and ***P<0.001, significantly different from BAT of wild-type mice (BAT^+/+)