Acyl-CoA thioesterase 9 (ACOT9) in mouse may provide a novel link between fatty acid and amino acid metabolism in mitochondria

Abstract

Acyl-CoA thioesterase (ACOT) activities are found in prokaryotes and in several compartments of eukaryotes where they hydrolyze a wide range of acyl-CoA substrates and thereby regulate intracellular acyl-CoA/CoA/fatty acid levels. ACOT9 is a mitochondrial ACOT with homologous genes found from bacteria to humans and in this study we have carried out an in-depth kinetic characterization of ACOT9 to determine its possible physiological function. ACOT9 showed unusual kinetic properties with activity peaks for short-, medium-, and saturated long-chain acyl-CoAs with highest V max with propionyl-CoA and (iso) butyryl-CoA while K cat/K m was highest with saturated long-chain acyl-CoAs. Further characterization of the short-chain acyl-CoA activity revealed that ACOT9 also hydrolyzes a number of short-chain acyl-CoAs and short-chain methyl-branched CoA esters that suggest a role for ACOT9 in regulation also of amino acid metabolism. In spite of markedly different K ms, ACOT9 can hydrolyze both short- and long-chain acyl-CoAs simultaneously, indicating that ACOT9 may provide a novel regulatory link between fatty acid and amino acid metabolism in mitochondria. Based on similar acyl-CoA chain-length specificities of recombinant ACOT9 and ACOT activity in mouse brown adipose tissue and kidney mitochondria, we conclude that ACOT9 is the major mitochondrial ACOT hydrolyzing saturated C2-C20-CoA in these tissues. Finally, ACOT9 activity is strongly regulated by NADH and CoA, suggesting that mitochondrial metabolic state regulates the function of ACOT9.

Fig. 1

Mouse Acot9 mRNA and protein is widely expressed. a mRNA expression of Acot9. Total RNA was isolated and pooled into two pools per tissue, with 2–3 animals in each pool, or one tissue pool each for gallbladder, adrenal and the intestinal epithelium segments, with samples from 4–6 animals. Samples were analyzed in triplicate and the relative expression was calculated using the 2^−ΔΔCt method, using 18 S as a reference gene. The data are shown as the mean expression in each tissue (±range for the tissues analyzed as two RNA pools) relative to the expression in liver set to 1. BAT brown adipose tissue, WAT white adipose tissue. The intestinal epithelium was divided into four equal segments, S1 (proximal intestine) to S4 (distal intestine). b Western-blot analysis of ACOT9 tissue expression. Forty μg of each tissue pool was separated by 10 % SDS/PAGE, blotted and probed with an ACOT9 peptide antibody. The first lane contains the molecular mass standard with the 50- and 70-kDa bands indicated and the band corresponding to ACOT9 indicated by an arrow. BAT brown adipose tissue, WAT white adipose tissue, Int epi prox and dist correspond to the S1 and S4 segments shown in a. Skel. mu (Gast) gastrocnemius skeletal muscle, Skel. mu (Abd) abdominis skeletal muscle

Fig. 2

ACOT9 is a more efficient thioesterase in the presence of both short-chain and long-chain acyl-CoAs. a Spectrophotometric measurement of ACOT9 activity with two substrates. ACOT9 activity was measured with various concentrations of isobutyryl-CoA (grey circles with the Michaelis–Menten curve fit in light grey), C₁₂-CoA (black circles with the Michaelis-Mentens curve fit in black) and the activity with different concentrations of isobutyryl-CoA at a fixed concentration of 7 μM C₁₂-CoA (grey squares, with two Michaelis–Mentens curve fits, one for the lower concentrations of isobutyryl-CoA and one for the higher). b HPLC analysis of ACOT9 activity with two substrates. A mixture of 5 μM of isobutyryl-CoA and 5 μM C₁₂-CoA were incubated together with 3-hydroxy-3-methylglutaryl-CoA with ACOT9, and the reaction was terminated by acidification at different time points (0–40 s) and analyzed by HPLC. The figure shows five merged HPLC chromatograms representing different time points. The peak at 17 min corresponds to isobutyryl-CoA and the peak at 27 min corresponds to C₁₂-CoA. The large peak at 18 min is DTNB that was added to the reaction to prevent possible inhibition by released CoA during the reaction

Fig. 3

Acyl-CoA chain-length specificity of recombinant ACOT9 and thioesterase activity in isolated mitochondria. a Thioesterase activity of ACOT9 was measured spectrophotometrically at 412 nm with various concentrations of saturated acyl-CoAs and the V _max values were calculated and plotted in the upper panel (mean ± SEM, n = 2–5). Mitochondria were isolated from BAT, kidney, and liver, sonicated and centrifuged, and thioesterase activity was measured in the supernatants as described under the "Materials and methods" section. The black circles show the activity with 50 μM of acyl-CoA esters ranging from C₂ to C₁₂-CoA, 25 μM C₁₄-CoA and 10 μM C₁₆-C₁₈-CoA (data are mean ± SEM of two independent experiments). Note the very similar activity pattern for recombinant ACOT9 and BAT mitochondria. The grey squares show the remaining thioesterase activity after 2 min pre-incubation with 20 μM C₁₄-CoA thioether with mitochondrial extracts. b C₁₄-CoA thioether is a potent inhibitor of ACOT9 activity. Recombinant ACOT9 was pre-incubated with various concentrations of C₁₄-CoA thioether (0–10 μM) and the activity was measured at fixed concentrations of C₃-CoA (50 μM) (open circles) and C₁₂-CoA (25 μM) (grey squares) using the spectrophotometric assay at 412 nm (mean ± SEM with two different enzyme preparations)

Fig. 4

ACOT9 activity is regulated by NADH and CoA. a Recombinant ACOT9 was incubated at fixed concentrations of C₃-CoA (50 μM) or C₁₄-CoA (10 μM) and varying concentrations of NADH (0–1,000 μM). The activity was measured using the standard spectrophotometric assay at 412 nm and the graph shows one representative experiment out of three for C₃-CoA (upper panel) and one out of two experiments for C₁₄-CoA (lower panel). The data are shown as the % remaining activity in the presence of NADH (mean ± SEM of 1–4 measurements). b Thioesterase activity was measured with soluble extracts of mitochondria isolated from BAT, kidney, and liver with 50 μM C₃-CoA or 25 μM C₁₄-CoA in the presence (light grey bars) or absence (dark grey bars) of 400 μM NADH. c Recombinant ACOT9 was incubated at fixed concentrations of C₃-CoA or (50 μM) and C₁₄-CoA (10 μM) in the presence of varying concentrations of CoA (0–200 μM) and activity was measured using the UV spectrophotometric assay. One representative experiment out of three is shown for C₃-CoA and one out of two for C₁₄-CoA. The values represent mean ± SEM of 2–5 measurements

Fig. 5

ACOT9 potentially regulates fatty acid and amino acid metabolism in mitochondria. Hypothetical scheme of fatty acid and amino acid metabolism pathways in which ACOT9 may be involved. Products and intermediates as CoA esters that were tested and found to be substrates for ACOT9 are colored in grey, where a darker shade indicates higher activity. Color: Darkest grey = V _max > 15 μmol/min/mg, Darker grey = V _max 10–15 μmol/min/mg, Light grey = V _max < 10 μmol/min/mg. Boxes with graded shade intensity of grey symbolize groups of substrates that include both good and poor substrates for ACOT9. BCAT branched-chain aminotransferase, BCKDH branched-chain ketoacid dehydrogense, αKIV α-ketoisovaleric acid, αKMV α-keto-β-methylvaleric acid, αKIC α-ketoisocaproic acid, DMN-CoA dimethylnonanoyl-CoA. The activity with 3-methylcrotonyl-CoA was too low to allow kinetic analysis. 3-Hydroxyisobutyryl-CoA was not tested as a substrate for ACOT9 but a specific mitochondrial thioesterase (HIBCH) that hydrolyzes 3-hydroxyisobutyryl-CoA and also 3-hydroxypropionyl-CoA has been identified in mammals [50]