GPI-anchored proteins tethered to lipid bilayers: modelling a complex interplay of carbohydrates, proteins and lipids

S2.3 Synthetic glycoproteins
Location (hall): 
Start/end time: 
Monday, July 1, 2019 - 15:00 to 15:15

Pallavi Banerjee1, Reinhard Lipowsky1,2, Mark Santer1

1Max Planck Institute Of Colloids And Interfaces, Potsdam, Germany, 2University of Potsdam, Potsdam, Germany

The glycosylphosphatidylinositol(GPI) anchor is a complex glycolipid abundant on the surface of eukaryotic cells. It serves as an anchor for a variety of proteins, which are connected to a lipid tail through a conserved glycan backbone. GPIs have been thought to contribute to a number of processes involving the cell surface, such as sorting, trafficking or signal transduction, e.g., through the ultimate release of the protein into the extracellular matrix through cleavage by phospholipase C or D. The glycan and also the fatty acid part can be subject to a number of variations, but how to define a structure-function relationship is still an open question. This is mainly because no clear picture exists of how protein and glycan part arrange with respect to the lipid layer[1][2]. Direct experimental evidence is rather scarce, and it appears that atomistic computational modeling through molecular dynamics (MD) simulation would be a convenient method to make progress. However, a GPI anchored protein implies three mutually interacting molecular species in close proximity. Here we show how to construct a modular molecular model of GPIs and GPI anchored proteins that can readily be extended to a broad variety of systems, addressing the micro heterogeneity of GPIs. We do so by creating a hybrid link to which different components with their respective optimized force fields can be attached, such as additional carbohydrates (GLYCAM06), Lipids (Lipid14) and the protein (ffSB14), all sharing the general protocol of the Amber family.

Our results demonstrate that GPI prefers flopping down on the membrane, thereby, strongly interacting with the lipid heads, over standing tall like a lollipop. When attaching green fluorescent protein (GFP) to the GPI, it was seen to lie in close proximity to the bilayer, interacting both with the lipid head and glycan part of the GPI. On extending this model to Toxoplasma gondii GPI, it was seen that the side branch galactosamine was always exposed to the solvent, with barely any interaction with the the bilayer. This finding is in correspondence with experimental studies[3] that show that the galactose residue of the parasite's GPI is the recognition moeity for the binding of Galectin-3 so as to be targeted by macrophages. Our GPI model can be used to study many parasitic and clinically relevant GPIs like trypanosomes and malaria parasites which would pave the way for developing new strategies in vaccine therapy. 

Favored conformations of GPI core (left), and GPI-anchored GFP (right) when inserted in DMPC bilayers.

  1. Lehto, M. T.; Sharom, F. J. Proximity of the Protein Moiety of a GPI-Anchored Protein to the Membrane Surface:  A FRET Study. Biochemistry, 2002, 41, 8368-8376.
  2. Paulick, M. G.; Forstner, M. B.; Groves, J. T.; Bertozzi, C. R. Synthetic Analogues of Glycosylphosphatidylinositol-Anchored Proteins and Their Behavior in Supported Lipid Bilayers. J. Am. Chem. Soc., 2007, 129 , 11543 -11550
  3. Debierre-Grockiego, F.; Niehus, S.; Coddeville, B.; Elass, E.; Poirier, F.; Weingart, R.; Schmidt, R.R.; Mazurier, J.; Guérardel, Y.; Schwarz, R.T. The binding of Toxoplasma gondii glycosylphosphatidylinositols to galectin-3 is required for their recognition by macrophages. J. Biol. Chem., 2010, jbc-M110.