Structure of a PH domain from the C. elegans muscle protein UNC-89 suggests a novel function

BACKGROUND

Pleckstrin homology (PH) domains constitute a structurally conserved family present in many signaling and regulatory proteins. PH domains have been shown to bind to phospholipids, and many function in membrane targeting. They generally have a strong electrostatic polarization and interact with negatively charged phospholipids via the positive pole. On the basis of electrostatic modeling, however, we have previously identified a class of PH domains with a predominantly negative charge and predicted that these domains recognize other targets. Here, we report the first experimental structure of such a PH domain.

RESULTS

The structure of the PH domain from Caenorhabditis elegans muscle protein UNC-89 has been determined by heteronuclear NMR. The domain adopts the classic PH fold, but has an unusual closed conformation of the "inositol binding loops. This creates a small opening to a deep hydrophobic pocket lined with negative charges on one side, and provides a molecular explanation for the lack of association with inositol-1,4,5-triphosphate. As predicted, the PH domain of UNC-89 has a strongly negative overall electrostatic potential. Modeling the Dbl homology (DH)-linked PH domains from the C. elegans genome shows that a large proportion of these modules are negatively charged.

CONCLUSIONS

We present the first structure of a PH domain with a strong negative overall electrostatic potential. The presence of a deep pocket lined with negative charges suggests that the domain binds to ligands other than acidic phospholipids. The abundance of this class of PH domain in the C. elegans genome suggests a prominent role in mediating protein-protein interactions.

A new approach for applying residual dipolar couplings as restraints in structure elucidation

Residual dipolar couplings are useful global structural restraints. The dipolar couplings define the orientation of a vector with respect to the alignment tensor. Although the size of the alignment tensor can be derived from the distribution of the experimental dipolar couplings, its orientation with respect to the coordinate system of the molecule is unknown at the beginning of structure determination. This causes convergence problems in the simulated annealing process. We therefore propose a protocol that translates dipolar couplings into intervector projection angles, which are independent of the orientation of the alignment tensor with respect to the molecule. These restraints can be used during the whole simulated annealing protocol.

Classification of protein sequences by homology modeling and quantitative analysis of electrostatic similarity

Protein electrostatics plays a key role in ligand binding and protein-protein interactions. Therefore, similarities or dissimilarities in electrostatic potentials can be used as indicators of similarities or dissimilarities in protein function. We here describe a method to compare the electrostatic properties within protein families objectively and quantitatively. Three-dimensional structures are built from database sequences by comparative modeling. Molecular potentials are then computed for these with a continuum solvation model by finite difference solution of the Poisson-Boltzmann equation or analytically as a multipole expansion that permits rapid comparison of very large datasets. This approach is applied to 104 members of the Pleckstrin homology (PH) domain family. The deviation of the potentials of the homology models from those of the corresponding experimental structures is comparable to the variation of the potential in an ensemble of structures from nuclear magnetic resonance data or between snapshots from a molecular dynamics simulation. For this dataset, the results for analysis of the full electrostatic potential and the analysis using only monopole and dipole terms are very similar. The electrostatic properties of the PH domains are generally conserved despite the extreme sequence divergence in this family. Notable exceptions from this conservation are seen for PH domains linked to a Db1 homology (DH) domain and in proteins with internal PH domain repeats.

The PH superfold: a structural scaffold for multiple functions

Pleckstrin homology (PH) domains form a structurally conserved family that is associated with many regulatory pathways within the cell. Domains with a nearly identical fold are found in other families that share no sequence similarity, suggesting the existence of a stable PH superfold. The PH domains generally function as regulated membrane-binding modules that bind to inositol lipids and respond to upstream signals by targeting the host proteins to the correct cellular sites. The other domains with a similar fold, such as the phosphotyrosine binding domains, recognize protein ligands.

Functional diversity of PH domains: an exhaustive modelling study

BACKGROUND

Pleckstrin homology (PH) domains are found in many proteins involved in signal transduction or cytoskeletal organization. The general function for the domain is still unclear; phospholipid binding of some PH domains and a strong electrostatic polarization in the experimental structures suggest a role in localization on membranes. We have analyzed the electrostatic properties and the spatial amino acid distribution from homology models of the entire PH domain family.

RESULTS

Despite the sequence divergence, the quality of the models is sufficient for our study. Most PH domains have an electrostatic polarization similar to the experimental structures. but roughly half of the PH domains linked to a Dbl homology domain have very different electrostatic properties. We also found a striking electrostatic complementarity in some internal PH domain repeats. The analysis of the spatial distribution of amino acids identified residues in the phospholipid-binding site of the spectrin and dynamin PH domains as specific for these domains.

CONCLUSIONS

The mostly conserved electrostatic polarization supports a general function in binding to phospholipid membranes. But the presence of PH domains with opposite polarity suggests that ligands and functions have diverged during evolution. We also demonstrate homology modelling as a general sequence analysis tool that can yield significantly more information than conventional analysis.