Gisele A. Tavares's research while affiliated with Yale University and other places

The three-dimensional structure of the major horse allergen Equ c 1 has been determined at 2.3 Å resolution by x-ray crystallography. Equ c 1 displays the typical fold of lipocalins, a β-barrel flanked by a C-terminal α-helix. The space between the two β-sheets of the barrel defines an internal cavity that could serve, as in other lipocalins, for the binding and transport of small hydrophobic ligands. Equ c 1 crystallizes in a novel dimeric form, which is distinct from that observed in other lipocalin dimers and corresponds to the functional form of the allergen. Binding studies of point mutants of the allergen with specific monoclonal antibodies raised in mouse and IgE serum from horse allergic patients allowed to identify putative B cell antigenic determinants. In addition, total inhibition of IgE serum recognition by a single specific monoclonal antibody revealed the restricted nature of the IgE binding target on the molecular surface of Equ c 1.

The intracellular parasite Trypanosoma cruzi, the etiological agent of Chagas disease, sheds a developmentally regulated surface trans-sialidase, which is involved in key aspects of parasite-host cell interactions. Although it shares a common active site architecture with bacterial neuraminidases, the T.cruzi enzyme behaves as a highly efficient sialyltransferase. Here we report the crystal structure of the closely related Trypanosoma rangeli sialidase and its complex with inhibitor. The enzyme folds into two distinct domains: a catalytic beta-propeller fold tightly associated with a lectin-like domain. Comparison with the modeled structure of T.cruzi trans-sialidase and mutagenesis experiments allowed the identification of amino acid substitutions within the active site cleft that modulate sialyltransferase activity and suggest the presence of a distinct binding site for the acceptor carbohydrate. The structures of the Trypanosoma enzymes illustrate how a glycosidase scaffold can achieve efficient glycosyltransferase activity and provide a framework for structure-based drug design.

The secreted protein Equ c 1 is the major component responsible for the induction of specific IgE antibodies in patients sensitized to horse allergens. Equ c 1 belongs to the lipocalin superfamily of hydrophobic ligand-binding proteins, which also includes other known allergens. Equilibrium sedimentation and gel-filtration studies demonstrate that both the glycosylated form of Equ c 1 purified from horse salivary glands and the non-glycosylated recombinant form expressed in bacteria exist predominantly as dimers in solution. As observed for other dimeric lipocalins, acidic pH and low protein concentration favour dimer dissociation. The recombinant form of Equ c 1 has been crystallized using ammonium sulfate as a precipitant. The crystals belong to the tetragonal space group P41212 with cell parameters a = b = 84.0, c = 56.1 A, and contain a single molecule in the asymmetric unit. A complete data set from native crystals was collected at the synchrotron source in Hamburg to 2.9 A resolution using a frozen crystal, and structure determination is in progress.

The quaternary organization of the cellulosome, a multi-enzymatic extracellular complex produced by cellulolytic bacteria, depends on specific interactions between dockerin domains, double EF-hand subunits carried by the catalytic components, and cohesin domains, individual receptor subunits linearly arranged within a non-catalytic scaffolding polypeptide. Cohesin-dockerin complexes with distinct specificities are also thought to mediate the attachment of cellulosomes to the cell membrane. We report here the crystal structure of a single cohesin domain from the scaffolding protein of Clostridium thermocellum. The cohesin domain folds into a nine-stranded beta-sandwich with an overall "jelly roll" topology, similar to that observed in bacterial cellulose-binding domains. Surface-exposed patches of conserved residues promote extensive intermolecular contacts in the crystal, and suggest a possible binding target for the EF-hand pair of the cognate dockerin domain. Comparative studies of cohesin domains indicate that, in spite of low sequence similarities and different functional roles, all cohesin domains share a common nine-stranded beta-barrel fold stabilized by a conserved hydrophobic core. The formation of stable cohesin-dockerin complexes requires the presence of Ca2+. However, the structure of the cohesin domain reported here reveals no obvious Ca2+-binding site, and previous experiments have failed to detect high affinity binding of Ca2+ to the unliganded dockerin domain of endoglucanase CelD. Based on structural and biochemical evidence, we propose a model of the cohesin-dockerin complex in which the dockerin domain requires complexation with its cohesin partner for protein stability and high-affinity Ca2+ binding.

The partial amino acid sequence of the tetrameric isolectin B4 from Vicia villosa seeds has been determined by peptide analysis, and its three-dimensional structure solved by molecular replacement techniques and refined at 2.9 A resolution to a crystallographic R-factor of 21%. Each subunit displays the thirteen-stranded beta-barrel topology characteristic of legume lectins. The amino acid residues involved in metal- and sugar-binding are similar to those of other GalNAc-specific lectins, indicating that residues outside the carbohydrate-binding pocket modulate the affinity for the Tn glycopeptide. Isolectin B4 displays an unusual quaternary structure, probably due to protein glycosylation.

The trans-sialidase from T. cruzi (TcTS), the protozoan agent of Chagas disease in the American continent, is member of a family of glycoproteins expressed on the parasite's surface and shed into the medium [1,2]. Trypanosomes are unable to synthesize sialic acid and use this enzyme to scavenge the monosaccharide from host glycoconjugates to sialylate mucin-like acceptor molecules present in the parasite plasma membrane. Sialic acid in mucins has been implicated in key aspects of parasite/host-cell interactions such as cell adhesion and invasion, and resistance to non-specific complement attack. Given the involvement of trans-sialidase in T. cruzi infection of humans and the absence of a similar enzyme in eukaryotic cells, TcTS constitutes a good target for the development of compounds useful in controlling the infection. The TcTS protein family is encoded by about 140 genes in the Trypanosoma genome [3], many of which code for inactive protein products. The protein consists of a 70 kDa globular core conveying the enzymatic activity followed by a variable number of immunogenic repeats. A closely related American trypanosome parasite, T. rangeli expresses a surface sialidase (TrSA) which is, as TcTS, the product of a multi-gene family encoding both active and inactive protein products [4]. Although TrSA is very similar in amino acid sequence to the globular core of TcTS (~70% identity for 640 amino acids), it lacks a repetitive C-terminal domain and is completely devoid of trans-sialidase activity. Using diffraction data collected at the BW7B beamline at Hamburg, we have determined the structure of TrSA at 2.2 Å resolution [5]. The crystals are orthorhombic, space group P2 1 2 1 2 1 , with cell dimensions a = 76.09 Å, b = 93.29 Å, c = 105.35 Å and one molecule in the asymmetric unit. The structure was determined by a combination of molecular replacement and heavy atom methods using a single mercurial derivative. The enzyme folds into two distinct domains: a catalytic β-propeller fold tightly associated with a lectin-like domain. Comparison with the modeled structure of T. cruzi trans-sialidase and mutagenesis experiments show that amino acid substitutions within the active site cleft modulate sialyltransferase activity by creating a distinct binding site for the acceptor carbohydrate. The structures of the trypanosoma enzymes illustrate how a glycosidase scaffold can achieve efficient glycosyltransferase activity and provide a framework for structure-based drug design.