A unique feature of chloroplast membrane of higher plants is the high content of MonoGalactosylDiacylGlycerol (MGDG) and DiGalactosylDiacylGlycerol (DGDG) which represents up to 80% of their lipid contents.The concerted action of two galactosyl transferases, ‘MGDG’ synthases and ‘DGDG’ synthases, is key in the synthesis of these galactolipids. One of these enzymes (MGD1) synthesizes the bulk of MGDG needed for the rapid and massive expansion of thylakoid membranes in response to light. The protein transfers the galactosyl residue from uridine 5’-diphosphate-α-D-galactose (UDP-Gal) to the sn-3 position of diacylglycerol (DAG) forming MGDG.
Despite the major importance of this process in plant growth and bioenergy conversion, as well as its large potential for industrial applications, there still remain many aspects of the protein action to be unraveled. Efficient functioning of MGD1 is associated with a complex molecular interplay between the protein, UDP-Gal donor molecule, DAG acceptor molecule and allosteric regulator lipids, on the surface of the outer leaflet of inner envelope membrane of chloroplast.
The resolution of the crystal structure of MGD1 provides an opportunity for an atomistic insight into the interactions and dynamics occurring between the protein and the constituents of the galactolipid membrane in the course of active complex formation. In the present study we employ the Coarse-Grained and All-Atom molecular dynamics simulations of the MGD1 - lipid membrane system which mimics the inner envelope membrane. It was revealed that protein assists the creation of lipid domains with an increased local concentration of DAG. In turn, the membrane surface influences the dynamics of the residues and modifies the correlated motions in the protein. A possible mechanism of allosteric signal transduction over the protein molecule is discussed.
The calculations were carried out using the equipment of the shared research facilities of HPC computing resources at Lomonosov Moscow State University (Moscow, Russia) and High-Performance Computing center of University Grenoble Alpes (Grenoble, France). This work was partly supported by the Russian Foundation for Basic Research and the Government of Tatarstan Republic, project № 18-44-160026. The authors acknowledge support by the ANR PIA Glyco@Alps (ANR-15-IDEX-02).