Fig. 4. The FAD-gatekeeper loop stabilises the ACAD dimeric conformation but does not determine ECSIT binding.

A Superposition of VLCAD crystal structure (7S7G) and the ACAD9_S191A cryo-EM structure. B Representation of ACAD9 structure based on the evolutionary conserved residues estimated by the ConSurf prediction software²⁰. Aligned sequences of the ACAD9 and VLCAD FAD-gatekeeper loop. Mutated residues are shown in bold. C VLCAD_WT is a very stable dimer (blue) and does not form a complex when reconstituted with ECSIT_CTER (orange). D ACAD9_{A184S/SR194TS} primarily forms a dimer (pink) although it also forms higher order species upon reconstitution with ECSIT_CTER (blue), indicating a decrease in the stability of the complex, while retaining the ability to bind to ECSIT. Experiments were repeated thrice with similar results. E DLS measurements of ACAD9 mutants designed to mimic the VLCAD loop. The ACAD9_A184S mutant exhibits similar behaviour to ACAD9_WT. The double mutant ACAD9_SR194TS is destabilised, although the effect was less profound in complex with ECSIT_CTER; some complex is formed but at higher MW than ACAD9_WT-ECSIT_CTER. In contrast, the double mutant ACAD9_{A184S/SR194TS} shows no change in MW before and after reconstitution with ECSIT_CTER, however, we can attribute this to a reduction in protein stability while retaining the ability to bind ECSIT_CTER. Blue and red dashed lines indicate the expected MW for ACAD9_WT homodimer and ACAD9_WT-ECSIT_CTER complex, respectively. Data are presented as mean values ± SD of at least n = 3 biological replicas. F Acyl-CoA dehydrogenase (ACAD) activity of ACAD9_WT, ACAD9 mutants, and VLCAD determined by an ETF fluorescence reduction assay. After addition of the ACAD specific substrate palmitoyl-CoA (C16:0), there is a clear loss of ETF fluorescence in ACAD9_WT, ACAD9_A184S and VLCAD in comparison to the other ACAD9 mutants, over 300 s of reaction measurement. Data are presented as box-whisker plots with median quartiles and from minimum to maximum values, at least n = 3 biological replicas. Statistical significance was calculated with an RM-one way ANOVA with Dunnett’s multiple comparisons test relative to ACAD9. p values are indicated (non-significant p = 0.8508, p = 0.3179; ****p ≤ 0.0001). Source data are provided as a Source Data file.