Acyl-CoA dehydrogenase substrate promiscuity: Challenges and opportunities for development of substrate reduction therapy in disorders of valine and isoleucine metabolism

Abstract

Toxicity of accumulating substrates is a significant problem in several disorders of valine and isoleucine degradation notably short-chain enoyl-CoA hydratase (ECHS1 or crotonase) deficiency, 3-hydroxyisobutyryl-CoA hydrolase (HIBCH) deficiency, propionic acidemia (PA), and methylmalonic aciduria (MMA). Isobutyryl-CoA dehydrogenase (ACAD8) and short/branched-chain acyl-CoA dehydrogenase (SBCAD, ACADSB) function in the valine and isoleucine degradation pathways, respectively. Deficiencies of these acyl-CoA dehydrogenase (ACAD) enzymes are considered biochemical abnormalities with limited or no clinical consequences. We investigated whether substrate reduction therapy through inhibition of ACAD8 and SBCAD can limit the accumulation of toxic metabolic intermediates in disorders of valine and isoleucine metabolism. Using analysis of acylcarnitine isomers, we show that 2-methylenecyclopropaneacetic acid (MCPA) inhibited SBCAD, isovaleryl-CoA dehydrogenase, short-chain acyl-CoA dehydrogenase and medium-chain acyl-CoA dehydrogenase, but not ACAD8. MCPA treatment of wild-type and PA HEK-293 cells caused a pronounced decrease in C3-carnitine. Furthermore, deletion of ACADSB in HEK-293 cells led to an equally strong decrease in C3-carnitine when compared to wild-type cells. Deletion of ECHS1 in HEK-293 cells caused a defect in lipoylation of the E2 component of the pyruvate dehydrogenase complex, which was not rescued by ACAD8 deletion. MCPA was able to rescue lipoylation in ECHS1 KO cells, but only in cells with prior ACAD8 deletion. SBCAD was not the sole ACAD responsible for this compensation, which indicates substantial promiscuity of ACADs in HEK-293 cells for the isobutyryl-CoA substrate. Substrate promiscuity appeared less prominent for 2-methylbutyryl-CoA at least in HEK-293 cells. We suggest that pharmacological inhibition of SBCAD to treat PA should be investigated further.

Keywords: branched-chain amino acid; catabolism; genome editing; lipoylation; substrate reduction.

Figure 1.. MCPA inhibits SBCAD, but not ACAD8, and lowers C3-carnitine.

Quantification of C4- and C5-carnitine isomers, and C8-carnitine in the extracellular medium of HEK-293 cell lines in the presence of 0–100 μM MCPA. IC₅₀ values for inhibition of SBCAD, IVD, MCAD were calculated using the full range of MCPA concentrations (0–100 μM). The IC₅₀ value for inhibition of SCAD was calculated using MCPA concentrations up to 10 μM, at which the highest butyryl-carnitine concentration was reached. The estimated IC₅₀ value did not change by much if we used concentration up to 5 (IC₅₀ = 1.9 ± 0.9 μM) or 25 μM (IC₅₀ = 1.1 ± 0.2 μM). C3-carnitine data were fitted to an EC₅₀ curve demonstrating the dose dependent reduction of PCC substrate. The insets represent the corresponding IC₅₀ or EC₅₀. Error bars indicate SD.

Figure 2.. Deletion of ACADSB in HEK-293 cells increases C5-carnitine and limits the accumulation of C3-carnitine reflecting PCC substrate limitation.

(A) Production of C5-carnitine and C3-carnitine in the extracellular medium of selected wild-type and ACADSB KO HEK-293 cell lines. Error bars indicate SD. (B) Production of C5- and C3-carnitine in the extracellular medium of selected wild-type and PCCB KO HEK-293 cell lines in the presence of increasing concentrations of the ACAD inhibitor MCPA.

Figure 3.. Rescue of deficient lipoylation by MCPA in ECHS1/ACAD8 KO cell lines.

(A) Lipoylation of E2p and E2o in ECHS1/ACAD8 KO cell lines. Western blots of cell lysates were probed with an antibody against lipoic acid, and E2p (DLAT) and E2o (DLST) were identified based on their molecular weight. Citrate synthase was used as loading control and the position of the molecular weight marker proteins is indicated. (B) ECHS1/ACAD8 KO, ECHS1 KO, ACAD8 KO and HEK-293 cell lines were treated with 25μM MCPA or vehicle for 72 hours as indicated. Four different ECHS1/ACAD8 KO cell lines (clones #1-4) were analyzed and samples distributed over 2 blots.

Figure 4.. ACADSB KO is not sufficient to rescue deficient lipoylation in ECHS1/ACAD8 KO cell lines.

(A) Lipoylation of E2p in ECHS1/ACAD8/ACADSB KO cell lines. Western blots of cell lysates were probed with an antibody against lipoic acid and E2p (DLAT) was identified based on its molecular weight. E2o (DLST) was not reliably detected in these blots. Citrate synthase was used as loading control and the position of the molecular weight marker proteins is indicated. Two cell lines without incomplete ACADSB KO are marked with an asterisk. (B) ECHS1/ACAD8/ACADSB KO and selected other cell lines were treated with 25μM MCPA or vehicle for 72 hours as indicated.

Figure 5.. ACADM KO is not sufficient to rescue deficient lipoylation in ECHS1/ACAD8/ACADSB KO cell lines.

(A) Lipoylation of E2p in ECHS1/ACAD8/ACADSB/ACADM KO cell lines. Western blots of cell lysates were probed with an antibody against lipoic acid and E2p (DLAT) was identified based on its molecular weight. E2o (DLST) was not reliably detected in these blots. Citrate synthase was used as loading control and the position of the molecular weight marker proteins is indicated. Two cell lines without incomplete ACADSB KO are marked with an asterisk. (B) ECHS1/ACAD8/ACADSB/ACADM KO and selected other cell lines were treated with 25μM MCPA or vehicle for 72 hours as indicated. A faint signal for E2o was noted in these immunoblots.