Non-vesicular lipid transport refers to the movement of monomeric lipid molecules from one bilayer to another, catalyzed by lipid transfer proteins (LTPs), and is implicated in construction of new bilayers de novo and maintenance of existing ones. Recent evidence suggests that this transport process is the major route through which subcellular lipid distributions are maintained, requiring directional lipid-flux against steep concentration-gradients. Such a transport scheme requires the input of cellular energy but, since LTPs do not directly hydrolyze ATP, it is uncertain from where this energy originates. Lipid asymmetry—that is, a non-uniform distribution of lipids between the two leaflets of a bilayer—has been proposed as one answer to this outstanding question, whereby the physicochemical consequences of this property bias non-vesicular lipid transport. Here, we investigate the physical underpinnings of this hypothesis through a thermodynamic model we have developed, emphasizing one aspect of lipid asymmetry: number asymmetry, referring to unequal numbers of lipids between the leaflets of a bilayer. First, we used our thermodynamic model to show that the mechanical-consequences of number asymmetry (i.e., lateral tensions) depend upon its generation-pathway. We then compared calculations from our model to data from vesicles harboring number asymmetry, finding good agreement between our model-predictions and experimental values. This comparison confirms that lateral tensions incurred by number asymmetry contribute significantly to lipid-thermodynamics. In agreement with this finding, we show that lateral tensions from osmotic-swelling and number asymmetry modulate the extraction/transfer of lipids by a full-length LTP and an isolated domain with lipid transport activity in vitro. Altogether, this contribution provides a comprehensive framework to rationalize biased lipid-flux through non-vesicular pathways as well as quantitative support for the hypothesized coupling between non-vesicular lipid transport and lipid asymmetry.
Reagle et al. (Sun,) studied this question.