UC Berkeley

Cell homeostasis requires spatiotemporal regulation of heterogeneous membrane compositions. One key way that proper lipid distributions are maintained is through non-vesicular transport of individual lipids between membranes. Despite its biological importance, non-vesicular transport remains poorly understood at the molecular level. While lipid transfer proteins may be largely responsible for the selective transport of individual lipids in vivo, lipid transfer can also occur passively. Here, we expand our biophysical knowledge of non-vesicular lipid transport mechanisms by investigating the dynamics of passive lipid transport and then using those insights to examine how lipid transfer proteins catalyze lipid transport.

A detailed understanding of passive lipid transport has remained elusive due in part to inconsistencies between experimental findings and previous molecular simulations. We resolve these discrepancies by discovering the reaction coordinate for passive lipid transport, which enables a complete biophysical characterization of the rate-limiting step of lipid transport. Through analysis of unbiased all-atom and coarse-grained molecular dynamics simulations, we find that the reaction coordinate measures the formation and breakage of hydrophobic contacts between the membrane and transferring lipid. Consistent with experiments, free energy profiles along the reaction coordinate exhibit both a rate-limiting activation barrier for lipid desorption from a membrane and a significant barrier for lipid insertion, which was entirely missed in previous computational studies. Using our newly identified reaction coordinate, we formulate an expression for the rate of passive lipid transport to enable a quantitative comparison with experiments. Most importantly, we find that the breakage of hydrophobic lipid–membrane contacts is rate limiting for passive lipid transport.

Knowledge of the reaction coordinate allows us to systematically investigate how the activation free energy of passive lipid transport depends on membrane physicochemical properties. Through all-atom molecular dynamics simulations of 11 chemically distinct glycerophospholipids, we determine how lipid acyl chain length, unsaturation, and headgroup influence the free energy barriers for lipid desorption from and insertion into liquid-crystalline and gel phase membranes. Consistent with previous experimental measurements, we find that lipids with longer, saturated acyl chains have increased activation free energies compared to lipids with shorter, unsaturated chains. Lipids with different headgroups exhibit a range of activation free energies; however, no clear trend based solely on chemical structure can be identified, mirroring difficulties in the interpretation of previous experimental results. Compared to liquid-crystalline phase membranes, gel phase membranes exhibit substantially increased free energy barriers. Overall, we find that the activation free energy depends on a lipid’s local hydrophobic environment in a membrane and that the free energy barrier for lipid insertion depends on a membrane’s interfacial hydrophobicity. Both of these properties can be altered through local changes in membrane composition and phase, suggesting that variations in cell membrane hydrophobicity may be exploited to direct non-vesicular lipid traffic.

Our discovery that the rate of passive lipid exchange is limited by the disruption of a lipid’s local hydrophobic environment suggests that lipid transfer proteins may catalyze lipid transport by lowering the free energy barrier for hydrophobic lipid–membrane contact breakage. To test this hypothesis, we investigate how ceramide-1-phosphate transfer protein (CPTP) catalyzes the transport of ceramide-1-phosphate (C1P), a bioactive sphingolipid, between membranes. To resolve how CPTP extracts and inserts C1P into a membrane, we utilize a multiscale simulation approach that builds upon our findings about passive lipid exchange. We find that both the apo and C1P-bound forms of CPTP bind a membrane poised to extract and insert C1P, confirming predictions based on crystal structures. Membrane binding promotes conformational changes that widen the entrance to CPTP’s hydrophobic cavity, further facilitating the exchange of C1P between CPTP and a membrane. Due to its stronger electrostatic attraction to the membrane, the apo form binds deeper into the membrane, substantially disrupting a lipid’s local hydrophobic environment in the membrane below. As a result, CPTP lowers the free energy barrier for the breakage of hydrophobic C1P–membrane contacts. After extracting C1P, CPTP likely unbinds a membrane through an electrostatic switching mechanism similar to that used by other lipid transfer proteins. Thus, we provide novel insights into the molecular mechanisms used by lipid transfer proteins to efficiently traffic lipids between membranes.

Many lipid transfer proteins, including CPTP, function in local thermodynamic equilibrium. Yet, others require an expenditure of energy to drive lipid transport, thus functioning out of equilibrium. Efforts to characterize such lipid transfer proteins are currently limited by deficiencies in nonequilibrium simulation methods. To address some of the challenges in using simulations to calculate generalized free energies, or large deviation functions, of nonequilibrium systems, we develop a new transition path sampling method. Specifically, we devise a novel set of path sampling moves based on Brownian bridges, which are stochastic trajectories constrained to start and end at specified configurations. We use our method to efficiently calculate large deviation functions of asymmetric simple exclusion processes, a paradigmatic nonequilibrium transport model that could foreseeably be used to model lipid transfer proteins with tubular structures.

The studies presented in this thesis together significantly advance our biophysical knowledge of lipid transport. Furthermore, the approaches used and methods developed provide a means to investigate the function of numerous lipid transfer proteins. By understanding how individual lipid transfer proteins function at the molecular level, we can begin to understand how they collectively function to spatiotemporally regulate cell membrane compositions.