The use of inhibitors clearly
The use of inhibitors clearly has its caveats and the field should come to a consensus on etomoxir concentration to maximally block LC-FAO while minimalizing off-target effects. The current papers convincingly show that 200 μM is too high, but it should be noted that 3 μM does not fully suppress LC-FAO in intact macrophages and only suffices to maximally block LC-FAO in permeabilized cells. It should therefore be tested which etomoxir concentrations lead to LC-FAO blockade without causing off-target effects as observed at 200 μM. However, genetic models need to be interpreted with caution. Part of the discrepancies in dependency of LC-FAO between earlier and recent studies could be explained by the possibility that 17 alpha hydroxylase can adapt to long-term genetic deletion of Cpt1a by activating compensatory metabolic mechanisms. Using other fuels to support processes normally supported by LC-FAO, CPT1a can become dispensable in those cells, while cells subjected to acute knockdown or pharmacological inhibition might not be able to adapt in the same way. Moreover, the residual 30% LC-FAO observed in CPT1a-deficient macrophages and Tmem cells (at least in T cells possibly caused by the presence of CPT1b and CPT1c isoforms) may suffice to support IL-4-induced macrophage responses and Tmem development.
Thus, the genetic tools used in these studies come with certain limitations too, and thus is it too early to rewrite the “Guide to immunometabolism for immunologists” (O’Neill et al., 2016). Clearly more detailed comparative studies will be needed to reconcile some of the contradictory findings in this field to unequivocally establish the role of LC-FAO in macrophage and T cell biology. Meanwhile, the new studies highlight that the role of LC-FAO in shaping Tmem, Treg, and M(IL-4) cells is not as straightforward as previously envisioned and underpin the need for attention when interpreting studies with inhibitors.
Acknowledgments We thank Bart Everts and Sanne G.S. Verberk for their careful reading and suggestions. J.V.d.B. received a VENI grant from ZonMW (91615052), and a Netherlands Heart Foundation Junior Postdoctoral grant (2013T003) and Senior Fellowship (2017T048).
Introduction Betaine lipids are non-phosphorous glycerolipids that are structurally similar to the phospholipid PC. Both phospho- and betaine lipids have positively charged trimethylammonium group and similar phase transition temperatures (Sato and Murata, 1991). Diacylglyceryl-N,N,N-trimethylhomoserines (DGTS) are the most widespread class of betaine lipids. DGTS are abundant in several groups of bacteria (Benning et al., 1995, Geiger et al., 1999), green algae (Eichenberger, 1982, Sato and Furuya, 1985, Vaskovsky et al., 1996, Künzler and Eichenberger, 1997), primitive vascular plants such as mosses (Sato and Furuya, 1985, Künzler and Eichenberger, 1997), lycophytes and ferns (Künzler and Eichenberger, 1997, Rozentsvet et al., 2000, Rozentsvet, 2004), as well as lichens (Künzler and Eichenberger, 1997) and fungi (Künzler and Eichenberger, 1997, Dembitsky, 1996, Vaskovsky et al., 1998, Kotlova and Popov, 2005). Biosynthesis of DGTS has been studied in detail only in bacterial, algal and yeast cells. Two enzymes named BtaA and BtaB are required for DGTS production in bacteria (Klug and Benning, 2001, Riekhof et al., 2005). BtaA transfers a four-carbon backbone from S-adenosylmethionine to the diglyceride moiety, forming the intermediate diacylglycerylhomoserine. BtaB catalyses a three-step N-methylation of the amino group on the intermediate to form the final DGTS product. The green algae Chlamydomonas reinhardtii has a single polypeptide, BTA1, containing BtaA- and BtaB-like domains that carry out all steps in DGTS biosynthesis consecutively (Riekhof et al., 2005). Analyses of whole-genome sequences have revealed that CrBTA1 orthologs are abundant among eukaryotic organisms, and their distribution correlates with the distribution of DGTS. However, eukaryotic DGTS synthases, aside from BTA1 of C. reinhardtii, have been functionally studied only in the ascomycete yeast Kluyveromyces lactis (Riekhof et al., 2014). The amino acid sequence of KlBta1 was shown to be similar to CrBTA1 and contains conserved residues that have been implicated in the binding of AdoMet.