Limits...
Lipid-Loving ANTs: Molecular Simulations of CardiolipinInteractions and the Organization of the Adenine Nucleotide Translocase in Model Mitochondrial Membranes

View Article: PubMed Central - PubMed

ABSTRACT

The exchange of ADPand ATP across the inner mitochondrial membraneis a fundamental cellular process. This exchange is facilitated bythe adenine nucleotide translocase, the structure and function ofwhich are critically dependent on the signature phospholipid of mitochondria,cardiolipin (CL). Here we employ multiscale molecular dynamics simulationsto investigate CL interactions within a membrane environment. Usingsimulations at both coarse-grained and atomistic resolutions, we identifythree CL binding sites on the translocase, in agreement with thoseseen in crystal structures and inferred from nuclear magnetic resonancemeasurements. Characterization of the free energy landscape for laterallipid interaction via potential of mean force calculations demonstratesthe strength of interaction compared to those of binding sites onother mitochondrial membrane proteins, as well as their selectivityfor CL over other phospholipids. Extending the analysis to other membersof the family, yeast Aac2p and mouse uncoupling protein 2, suggestsa degree of conservation. Simulation of large patches of a model mitochondrialmembrane containing multiple copies of the translocase shows thatCL interactions persist in the presence of protein–proteininteractions and suggests CL may mediate interactions between translocases.This study provides a key example of how computational microscopymay be used to shed light on regulatory lipid–protein interactions.

No MeSH data available.


Related in: MedlinePlus

CL bindingsites on bovine ANT1. (A) In the left panel, the initialsimulated system consisted of a single molecule of bovine ANT1 embeddedin a mixed PC (lime) and CL (magenta) bilayer, solvated with standardMARTINI water particles (transparent surface) and neutralized with∼0.15 M NaCl (blue spheres). The right panel is the final snapshotof a 6 μs simulation showing three bound CL molecules (magenta).(B) Time-averaged 2D density maps of CL in the membrane plane forthe headgroup (left) and acyl tail moieties (right). (C) View ontothe base of the translocase from the matrix showing the positioningof the three CL (magenta) binding sites around the protein withinthe X-ray structure (left) and CG simulations (right). The CL bindingsites within the CG simulations are shown by the time-averaged probabilitydensity surface for the CL molecules, contoured to show the threebinding sites. The density surface was calculated from 5 × 6μs simulation repeats, each starting from a different randomlipid distribution. See Figure S4 for persimulation density maps.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5120876&req=5

fig2: CL bindingsites on bovine ANT1. (A) In the left panel, the initialsimulated system consisted of a single molecule of bovine ANT1 embeddedin a mixed PC (lime) and CL (magenta) bilayer, solvated with standardMARTINI water particles (transparent surface) and neutralized with∼0.15 M NaCl (blue spheres). The right panel is the final snapshotof a 6 μs simulation showing three bound CL molecules (magenta).(B) Time-averaged 2D density maps of CL in the membrane plane forthe headgroup (left) and acyl tail moieties (right). (C) View ontothe base of the translocase from the matrix showing the positioningof the three CL (magenta) binding sites around the protein withinthe X-ray structure (left) and CG simulations (right). The CL bindingsites within the CG simulations are shown by the time-averaged probabilitydensity surface for the CL molecules, contoured to show the threebinding sites. The density surface was calculated from 5 × 6μs simulation repeats, each starting from a different randomlipid distribution. See Figure S4 for persimulation density maps.

Mentions: CL molecules were seen to encounter and bind to threespecific sites on the protein [sites I–III (Figure 2 and Movie S1)]. Each site was composed of a groove on the protein surfaceformed by the N-terminal end of the even-numbered transmembrane helices(H2, H4, and H6) of each repeat, and the short amphipathic matrixhelices (MH1–MH3). The location of the sites thus has a symmetryrelated to the internal three-fold structural repeat. The locationof these sites is in excellent agreement with those identified inX-ray crystal structures.8,9 This indicates thatthe sites occupied in the crystal structure persist within a membraneenvironment. Interactions were predominantly mediated by the headgroupmoieties of CL, as seen in the time-averaged 2D density maps for CL(Figure 2B). In contrast,the acyl tails exhibited more dynamic interactions, which resultedin less well-defined density, though a small degree of ordering wasseen at site III. This is consistent with a number of crystal structuresin which only the headgroup moieties and adjacent atoms of the acylchains were resolved.29 The dominance ofheadgroup interactions seen here is also consistent with functionalassays performed in yeast showing Aac2p assembles normally regardlessof the acyl chain composition of CL.50


Lipid-Loving ANTs: Molecular Simulations of CardiolipinInteractions and the Organization of the Adenine Nucleotide Translocase in Model Mitochondrial Membranes
CL bindingsites on bovine ANT1. (A) In the left panel, the initialsimulated system consisted of a single molecule of bovine ANT1 embeddedin a mixed PC (lime) and CL (magenta) bilayer, solvated with standardMARTINI water particles (transparent surface) and neutralized with∼0.15 M NaCl (blue spheres). The right panel is the final snapshotof a 6 μs simulation showing three bound CL molecules (magenta).(B) Time-averaged 2D density maps of CL in the membrane plane forthe headgroup (left) and acyl tail moieties (right). (C) View ontothe base of the translocase from the matrix showing the positioningof the three CL (magenta) binding sites around the protein withinthe X-ray structure (left) and CG simulations (right). The CL bindingsites within the CG simulations are shown by the time-averaged probabilitydensity surface for the CL molecules, contoured to show the threebinding sites. The density surface was calculated from 5 × 6μs simulation repeats, each starting from a different randomlipid distribution. See Figure S4 for persimulation density maps.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5120876&req=5

fig2: CL bindingsites on bovine ANT1. (A) In the left panel, the initialsimulated system consisted of a single molecule of bovine ANT1 embeddedin a mixed PC (lime) and CL (magenta) bilayer, solvated with standardMARTINI water particles (transparent surface) and neutralized with∼0.15 M NaCl (blue spheres). The right panel is the final snapshotof a 6 μs simulation showing three bound CL molecules (magenta).(B) Time-averaged 2D density maps of CL in the membrane plane forthe headgroup (left) and acyl tail moieties (right). (C) View ontothe base of the translocase from the matrix showing the positioningof the three CL (magenta) binding sites around the protein withinthe X-ray structure (left) and CG simulations (right). The CL bindingsites within the CG simulations are shown by the time-averaged probabilitydensity surface for the CL molecules, contoured to show the threebinding sites. The density surface was calculated from 5 × 6μs simulation repeats, each starting from a different randomlipid distribution. See Figure S4 for persimulation density maps.
Mentions: CL molecules were seen to encounter and bind to threespecific sites on the protein [sites I–III (Figure 2 and Movie S1)]. Each site was composed of a groove on the protein surfaceformed by the N-terminal end of the even-numbered transmembrane helices(H2, H4, and H6) of each repeat, and the short amphipathic matrixhelices (MH1–MH3). The location of the sites thus has a symmetryrelated to the internal three-fold structural repeat. The locationof these sites is in excellent agreement with those identified inX-ray crystal structures.8,9 This indicates thatthe sites occupied in the crystal structure persist within a membraneenvironment. Interactions were predominantly mediated by the headgroupmoieties of CL, as seen in the time-averaged 2D density maps for CL(Figure 2B). In contrast,the acyl tails exhibited more dynamic interactions, which resultedin less well-defined density, though a small degree of ordering wasseen at site III. This is consistent with a number of crystal structuresin which only the headgroup moieties and adjacent atoms of the acylchains were resolved.29 The dominance ofheadgroup interactions seen here is also consistent with functionalassays performed in yeast showing Aac2p assembles normally regardlessof the acyl chain composition of CL.50

View Article: PubMed Central - PubMed

ABSTRACT

The exchange of ADPand ATP across the inner mitochondrial membraneis a fundamental cellular process. This exchange is facilitated bythe adenine nucleotide translocase, the structure and function ofwhich are critically dependent on the signature phospholipid of mitochondria,cardiolipin (CL). Here we employ multiscale molecular dynamics simulationsto investigate CL interactions within a membrane environment. Usingsimulations at both coarse-grained and atomistic resolutions, we identifythree CL binding sites on the translocase, in agreement with thoseseen in crystal structures and inferred from nuclear magnetic resonancemeasurements. Characterization of the free energy landscape for laterallipid interaction via potential of mean force calculations demonstratesthe strength of interaction compared to those of binding sites onother mitochondrial membrane proteins, as well as their selectivityfor CL over other phospholipids. Extending the analysis to other membersof the family, yeast Aac2p and mouse uncoupling protein 2, suggestsa degree of conservation. Simulation of large patches of a model mitochondrialmembrane containing multiple copies of the translocase shows thatCL interactions persist in the presence of protein–proteininteractions and suggests CL may mediate interactions between translocases.This study provides a key example of how computational microscopymay be used to shed light on regulatory lipid–protein interactions.

No MeSH data available.


Related in: MedlinePlus