Letter-to-the-editor on “Acetyl-CoA synthetase (ACSS2) does not generate butyryl- and crotonyl-CoA”

Dear editors of Molecular Metabolism,

We read with great interest the article by Zeaiter et al. concerning “Acetyl-CoA synthetase (ACSS2) does not generate butyryl- and crotonyl-CoA” in Molecular Metabolism. The authors conducted LC-MS/MS for coenzyme A (CoAs) measurement with short-chain fatty acids (SCFAs) and purified or recombinant acetyl-CoA synthetase 2 (ACSS2) enzymes in vitro to test the ability of ACSS2 to generate SCFA-CoAs from corresponding 3- and 4-carbon SCFAs, such as propionate, butyrate and crotonate. ACSS2 was unable to generate SCFA-CoAs from butyrate and crotonate and had very low activity with propionate. Additionally, structural modeling also indicated that the ACCS2 active site was poorly compatible with crotonyl-AMP. This topic is interesting, and I believe the methods in this article are clear. However, in view of the CoAs testing results, I would like to point out some concerns that may make the conclusions more convincing.

Firstly, the use of commercially available purified acetyl-CoA synthetase enzymes for the catalysis of SCFAs in vitro reaction systems need to be discussed. Enzyme activity may be affected in vitro and could lead to abnormal function of ACSS2, which differs from physiological conditions. Thus, interfering with ACSS2 expression (overexpressed or silenced) in cells to detect CoAs may be a better approach, as described in other articles.

Secondly, in my opinion, one of the greatest limitations of this study is that it may not be appropriate to optimize the reaction conditions with acetate as the substrate. It is worth noting that the unit of MS intensity is “10⁴ AU”, as shown in Table 1; therefore, we speculate that the MS intensity of crotonyl-CoA and butyryl-CoA may be much less than “10⁴ AU”, rather than completely “zero”. We believe that the amounts of crotonyl-CoA and butyryl-CoA are small but measurable, while using the highest amount of acetyl-CoA as a control would lead to smaller values for crotonyl-CoA and butyryl-CoA. Therefore, ACSS2 is much more capable of using acetate than butyrate and crotonate, but this finding does not indicate that ACSS2 cannot catalyze butyrate or crotonate. Additionally, for the catalytic effect of ACSS2, 0.5 mM acetate is sufficient but may not be sufficient for crotonate or butyrate. Previous study reported that intracellular crotonyl-CoA concentration is about 600- to 1,000-fold lower than that of acetyl-CoA, and the doses of crotonate used for crotonyl-CoA detection were usually 2.5 mM, 5 mM, or 10 mM. Sabari et al. also reported that knockdown of ACSS2 reduced the amount of crotonyl-CoA produced in the presence of 10 mM crotonate in HeLa S3 cells. Thus, we believe that increasing the content of crotonate may promote the generation of crotonyl-CoA by ACSS2.

In summary, we obtained a new method to investigate the ability of ACSS2 to generate other CoAs, such as structural modeling. However, for these CoAs detection, we suppose that regulating ACSS2 expression in living cells may be more credible. Using the highest amount of acetyl-CoA as a contrast for other CoAs may not be appropriate. Additionally, increasing the concentration of butyrate and crotonate to measure the CoAs content might be more convincing.