Membrane perturbing properties of toxin mycolactone from Mycobacterium ulcerans

Abstract

Mycolactone is the exotoxin produced by Mycobacterium ulcerans and is the virulence factor behind the neglected tropical disease Buruli ulcer. The toxin has a broad spectrum of biological effects within the host organism, stemming from its interaction with at least two molecular targets and the inhibition of protein uptake into the endoplasmic reticulum. Although it has been shown that the toxin can passively permeate into host cells, it is clearly lipophilic. Association with lipid carriers would have substantial implications for the toxin’s distribution within a host organism, delivery to cellular targets, diagnostic susceptibility, and mechanisms of pathogenicity. Yet the toxin’s interactions with, and distribution in, lipids are unknown. Herein we have used coarse-grained molecular dynamics simulations, guided by all-atom simulations, to study the interaction of mycolactone with pure and mixed lipid membranes. Using established techniques, we calculated the toxin’s preferential localization, membrane translocation, and impact on membrane physical and dynamical properties. The computed water-octanol partition coefficient indicates that mycolactone prefers to be in an organic phase rather than in an aqueous environment. Our results show that in a solvated membrane environment the exotoxin mainly localizes in the water-membrane interface, with a preference for the glycerol moiety of lipids, consistent with the reported studies that found it in lipid extracts of the cell. The calculated association constant to the model membrane is similar to the reported association constant for Wiskott-Aldrich syndrome protein. Mycolactone is shown to modify the physical properties of membranes, lowering the transition temperature, compressibility modulus, and critical line tension at which pores can be stabilized. It also shows a tendency to behave as a linactant, a molecule that localizes at the boundary between different fluid lipid domains in membranes and promotes inter-mixing of domains. This property has implications for the toxin’s cellular access, T-cell immunosuppression, and therapeutic potential.

A computational model to unravel the function of amyloid-βpeptide in contact with a phospholipid membrane

“…and selected trajectories (traj./REMD). Root-mean square errors are within brackets.Exogenous addition of Aβ peptide to the bilayer 323In the REMD Aβ/DMPC and Cu-Aβ/DMPC simulations the DMPC bilayer is in the 324 liquid crystal phase at all the probed temperatures, consistently with similar MD 325 simulations reported in the literature[76]. The temperature dependence of the area per 326 lipid in Aβ/DMPC REMD simulation is displayed inFig 4,together with the available 327 experimental results for DMPC[75], the result for CMD at T =311 K for DMPC, and 328 the average of 10 CMD trajectories at T = 303 K described below.…”

Mechanism of Mycolactone Toxin Membrane Permeation: Atomistic vs Coarse-Grained MARTINI Simulations

“…Both AT2R and WASP are membrane associated proteins that have been suggested to be regulated by ordered microdomains that localize and facilitate assembly of signaling complexes (14)(15)(16). Our previous computational work suggested that toxin decreases line tension, thereby disrupting domain formation (17). Recent experimental work confirmed this prediction.…”

“…The thickness of the simulation box (z direction) is about 10 nm. In order to compare our research with the previous work of Lopez et al (17), we adopted the same MD simulation parameters for mycolactone, which were developed with the General Amber Force Field (GAFF) (22,23) and restrained electrostatic potential (RESP) charges. To verify the legitimacy of these parameters, we compared them to those generated with the General Automated Atomic Model Parameterization protocol (24) and found them to be reasonably consistent.…”

“…We focused the comparison on the permeation of mycolactone B only. The MARTINI CG force field parameters were adapted from Lopez et al (17), where the MARTINI CG representation of mycolactone B was obtained from the insane protocol (46). After equilibration, the dimension of the MARTINI CG simulation box was about 5.5 nm × 5.5 nm × 6.6 nm.…”

“…We are additionally testing the use of our recently developed transition tempered metadynamics (TTMetaD) method to characterize membrane permeation of a relatively large molecule at the all atom level, and carefully comparing these results to those from coarse-grained MARTINI simulations. Following up on the work of Lopez et al (17), we have herein used both CG and all atom molecular dynamics (MD) simulations employing TTMetaD to characterize the association of the toxin with DPPC membranes. We calculate membrane permeation free energy profiles for both isomers A and B, which exist in a 60:40 ratio.…”