GLYCOMIMETIC LIGANDS TARGETING BACTERIAL LECTINS: RATIONAL DESIGN, SYNTHESIS AND BIOPHYSICAL EVALUATION

The research presented in the thesis is part of a broader effort to develop new strategies for combating bacterial infections, which are becoming increasingly difficult to treat due to rising antibacterial resistance. In this context, anti-adhesion therapy (AAT) is emerging as a complementary approach to traditional antibiotics. AAT aims to inhibit the adhesion of pathogens to host cells, which is one of the initial and essential steps in host colonization and infection. Many bacteria use lectins, which are carbohydrate-binding proteins, to recognize glycoconjugates on the surface of host cells and facilitate adhesion. These bacterial proteins represent an attractive target for designing antagonists that disrupt glycoconjugate-lectin interactions with application in AAT. The focus of this research is the ‘superlectin’ BC2L-C from the opportunistic Gram-negative pathogen Burkholderia cenocepacia. B. cenocepacia is a globally spread, multidrug-resistant bacterium that causes severe respiratory infections, primarily in immunocompromised individuals or cystic fibrosis patients. Its lectin BC2L-C, proposed to play a key role in bacterial adhesion and biofilm formation, exhibits a dual carbohydrate specificity through its hexameric structure. This structure includes a mannose- and heptose-specific, calcium-dependent C-terminal dimeric domain, and a fucose-specific N-terminal trimeric domain. In previous work, a collection of bifunctional glycomimetics specifically targeting the N-terminal domain of BC2L-C (referred to as BC2LCNt) was discovered using a combination of fragment- and structure-based design methods. The synthesized first-generation ligands consisted of a L-fucose core linked to fragments capable of binding to a secondary site on the lectin (referred to as site X), near its natural fucose-binding region. Despite the promising results obtained with these ligands, their affinity for the N-terminal domain of the lectin remains in the low/sub-millimolar range. Therefore, to obtain novel glycomimetics with improved affinity for BC2LCNt, two approaches were employed in this work: i) optimization of first-generation ligands through bioisostery and extension strategies; and ii) development of covalent glycomimetic ligands capable of forming covalent bonds with specific nucleophilic residues of the lectin. Regarding the first approach, new compounds were designed and synthesized to optimize a first generation hit ligand. Modifications were made to improve potential interactions with the protein and enhance physico-chemical properties, such as the solubility in aqueous buffers. The work resulted in the discovery of a β-fucosylamide glycomimetic ligand as the best monovalent synthetic ligand of BC2LCNt so far, with a 20-fold affinity improvement to the parent monosaccharide. In the second approach, the covalent strategy was employed to enhance affinity with the target lectin. Through computer-aided design, more than one hundred compounds were designed, each featuring a fucose core linked through a spacer to an appropriate electrophilic group capable of reacting with a Cys72 or Lys108 nucleophilic residue within the lectin binding site. The compounds were screened in silico by means of docking methods, enabling the selection of the most promising covalent ligands. Following their synthesis and biophysical evaluation, notable results were achieved, particularly with compounds featuring a salicylaldehyde warhead designed to react with the Lys108 side chain. These compounds demonstrated further improved affinity for the target, exhibiting IC50 values in competitive binding assays within the same micromolar range as natural oligosaccharides targeting BC2LCNt. Additionally, the interaction of the putative covalent ligands with the target lectin was evaluated through mass spectrometry experiments, confirming their ability to form covalent adducts. Notable results were also obtained from the set of compounds designed to target the Cys72 side chain, although they exhibited lower binding affinity compared to the lysine-targeting compounds. Among the cysteine-targeting ligands, those featuring a proline spacer showed particularly promising outcomes. Notably, for one of these ligands, the first crystal structure of a ligand covalently bound to a bacterial lectin was resolved. These findings provide a foundation for further refinement and optimization of ligands targeting the cysteine residue in BC2LCNt. Overall, the insights gained from this thesis could be beneficial and broadly applicable in the design, synthesis, and evaluation of glycomimetics targeting bacterial lectins, particularly in the context of AAT.