Protein-lipid interactions: correlation of a predictive algorithm for lipid-binding sites with three-dimensional structural data

Renal Unit, Leukocyte Biology & Inflammation Program, Structural Biology Program and the Massachusetts General Hospital/Harvard Medical School, 149 13th Street, Charlestown, MA 02129, USA.
Theoretical Biology and Medical Modelling (Impact Factor: 1.27). 02/2006; 3:17. DOI: 10.1186/1742-4682-3-17
Source: PubMed

ABSTRACT Over the past decade our laboratory has focused on understanding how soluble cytoskeleton-associated proteins interact with membranes and other lipid aggregates. Many protein domains mediating specific cell membrane interactions appear by fluorescence microscopy and other precision techniques to be partially inserted into the lipid bilayer. It is unclear whether these protein-lipid-interactions are dependent on shared protein motifs or unique regional physiochemistry, or are due to more global characteristics of the protein.
We have developed a novel computational program that predicts a protein's lipid-binding site(s) from primary sequence data. Hydrophobic labeling, Fourier transform infrared spectroscopy (FTIR), film balance, T-jump, CD spectroscopy and calorimetry experiments confirm that the interfaces predicted for several key cytoskeletal proteins (alpha-actinin, Arp2, CapZ, talin and vinculin) partially insert into lipid aggregates. The validity of these predictions is supported by an analysis of the available three-dimensional structural data. The lipid interfaces predicted by our algorithm generally contain energetically favorable secondary structures (e.g., an amphipathic alpha-helix flanked by a flexible hinge or loop region), are solvent-exposed in the intact protein, and possess favorable local or global electrostatic properties.
At present, there are few reliable methods to determine the region of a protein that mediates biologically important interactions with lipids or lipid aggregates. Our matrix-based algorithm predicts lipid interaction sites that are consistent with the available biochemical and structural data. To determine whether these sites are indeed correctly identified, and whether use of the algorithm can be safely extended to other classes of proteins, will require further mapping of these sites, including genetic manipulation and/or targeted crystallography.

  • [Show abstract] [Hide abstract]
    ABSTRACT: Phosphoinositide lipids (PIPns) control numerous critical biological pathways, typically through the regulation of protein function driven by non-covalent protein-lipid binding interactions. Despite the importance of these systems, the unraveling of the full scope of protein-PIPn interactions has represented a significant challenge due to the massive complexity associated with these events, including the large number of diverse proteins that bind to these lipids, variations in the mechanisms by which proteins bind to lipids, and the presence of multiple distinct PIPn isomers. As a result of this complexity, global methods in which numerous proteins that bind PIPns can be identified and characterized simultaneously from complex samples, which have been enabled by key technological advancements, have become popular as an efficient means for tackling this challenge. This review article provides an overview of advancements in large-scale methods for profiling protein-PIPn binding, including experimental methods, such as affinity enrichment, microarray analysis and activity-based protein profiling, as well as computational methods, and combined computational/experimental efforts.
    Chemistry and Physics of Lipids 11/2013; DOI:10.1016/j.chemphyslip.2013.10.014 · 2.59 Impact Factor
  • Source
  • [Show abstract] [Hide abstract]
    ABSTRACT: Flavin mono-nucleotide (FMN) is a cofactor which is involved in many biological reactions. The insights on protein-FMN interactions aid the protein functional annotation and also facilitate in drug design. In this study, we have established a new method, making use of an encoding scheme of the three-dimensional probability density maps that describe the distributions of 40 non-covalent interacting atom types around protein surfaces, to predict FMN-binding sites on protein surfaces. One machine learning model was trained for each of the 30 protein atom types to predict tentative FMN-binding sites on protein structures. The method's capability was evaluated by five-fold cross validation on a dataset containing 81 non-redundant FMN-binding protein structures and further tested on independent datasets of 30 and 15 non-redundant protein structures respectively. These predictions achieved an accuracy of 0.94, 0.94 and 0.96 with Matthews correlation coefficient (MCC) of 0.53, 0.53 and 0.65 respectively for the three protein structure sets. The prediction capability is superior to the existing method. This is the first structure-based approach that does not rely on evolutionary information for predicting FMN-interacting residues. The webserver for the prediction is available at
    Journal of Theoretical Biology 11/2013; 192. DOI:10.1016/j.jtbi.2013.10.020 · 2.30 Impact Factor


Available from