NORMAL GENOMIC

• Seigneuret M
• Complete predicted three-dimensional structure of the facilitatortransmembrane protein and hepatitis C virus receptor CD81: conserved andvariable structural domains in the tetraspanin superfamily.
• Biophys J. 2006; 90: 212-27
• Display abstract

• Tetraspanins are a superfamily of transmembrane proteins implicated incellular development, motility, and activation through their interactionswith a large range of proteins and with specific membrane microdomains.The complete three-dimensional structure of the tetraspanin CD81 has beenpredicted by molecular modeling and from the crystallographic structure ofthe EC2 large extracellular domain. Periodicity of sequence conservation,homology modeling, secondary structure prediction, and protein dockingwere used. The transmembrane domain appears organized as a four-strandedleft-handed coiled coil directly connecting to two helices of the EC2. Asmaller extracellular loop EC1 contains a small largely hydrophobicbeta-strand that packs in a conserved hydrophobic groove of the EC2. Thepalmitoylable intracellular N-terminal segment forms an amphipathicmembrane-parallel helix. Structural variability occurs mainly in anhypervariable subdomain of the EC2 and in intracellular regions.Therefore, the variable interaction selectivity of tetraspanins originatesboth from sequence variability within structurally conserved domains andfrom the occurrence of small structurally variable domains. In CD81 andother tetraspanins, the numerous membrane-exposed aromatic residues areasymmetrically clustered and protrude on one side of the transmembranedomain. This may represent a functional specialization of these two sidesfor interactions with cholesterol, proteins, or membrane microdomains.

• Perry J
• The Epc-N domain: a predicted protein-protein interaction domain found inselect chromatin associated proteins.
• BMC Genomics. 2006; 7: 6-6
• Display abstract

• BACKGROUND: An underlying tenet of the epigenetic code hypothesis is theexistence of protein domains that can recognize various chromatinstructures. To date, two major candidates have emerged: (i) thebromodomain, which can recognize certain acetylation marks and (ii) thechromodomain, which can recognize certain methylation marks. RESULTS: TheEpc-N (Enhancer of Polycomb-N-terminus) domain is formally defined herein.This domain is conserved across eukaryotes and is predicted to form aright-handed orthogonal four-helix bundle with extended strands at bothtermini. The types of amino acid residues that define the Epc-N domainsuggest a role in mediating protein-protein interactions, possiblyspecifically in the context of chromatin binding, and the types ofproteins in which it is found (known components of histoneacetyltransferase complexes) strongly suggest a role in epigeneticstructure formation and/or recognition. There appear to be two major Epc-Nprotein families that can be divided into four unique protein subfamilies.Two of these subfamilies (I and II) may be related to one another in thatsubfamily I can be viewed as a plant-specific expansion of subfamily II.The other two subfamilies (III and IV) appear to be related to one anotherby duplication events in a primordial fungal-metazoan-mycetozoan ancestor.Subfamilies III and IV are further defined by the presence of anevolutionarily conserved five-center-zinc-binding motif in the loopconnecting the second and third helices of the four-helix bundle. Thismotif appears to consist of a PHD followed by a mononuclear Zn knuckle,followed by a PHD-like derivative, and will thus be referred to as thePZPM. All non-Epc-N proteins studied thus far that contain the PZPM havebeen implicated in histone methylation and/or gene silencing. In addition,an unusual phyletic distribution of Epc-N-containing proteins is observed.CONCLUSION: The data suggest that the Epc-N domain is a protein-proteininteraction module found in chromatin associated proteins. It is possiblethat the Epc-N domain serves as a direct link between histone acetylationand methylation statuses. The unusual phyletic distribution ofEpc-N-containing proteins may provide a conduit for future insight intohow different organisms form, perceive and respond to epigeneticinformation.

• Zhai P et al.
• The interaction of the human GGA1 GAT domain with rabaptin-5 is mediatedby residues on its three-helix bundle.
• Biochemistry. 2003; 42: 13901-8
• Display abstract

• GGA proteins regulate clathrin-coated vesicle trafficking by interactingwith multiple proteins during vesicle assembly. As part of this process,the GAT domain of GGA is known to interact with both ARF and Rabaptin-5.Particularly, the GAT domains of GGA1 and -2, but not of GGA3,specifically bind with a coiled-coil region of Rabaptin-5. Rabaptin-5interacts with Rab5 and is an essential component of the fusion machineryfor targeting endocytic vesicles to early endosomes. The recentlydetermined crystal structure of the GGA1 GAT domain has provided insightsinto its interactions with partner proteins. Here, we describe mutagenesisstudies on the GAT-Rabaptin-5 interaction. The results demonstrate that ahydrophobic surface patch on the C-terminal three-helix bundle motif ofthe GAT domain is directly involved in Rabaptin-5 binding. A GGA3-likemutation, N284S, in this Rabaptin-5 binding patch of GGA1 led to a reducedlevel of Rabaptin-5 binding. Furthermore, a reversed mutation, S293N, inGGA3 partially establishes Rabaptin-5 binding ability in its GAT domain.These results provide a structural explanation for the binding affinitydifference among GGA proteins. The current results also suggest that thebinding of GAT to Rabaptin-5 is independent of its interaction with ARF.

• Davidson AL
• Structural biology. Not just another ABC transporter.
• Science. 2002; 296: 1038-40