Targeting undruggable carbohydrate recognition sites through focused fragment library design

Carbohydrate-protein interactions (CPIs) play pivotal roles in numerous biological processes, including cell-cell communication and host-pathogen interactions. The specificity of these interactions arises from the intricate three-dimensional structures of carbohydrates and the complementary binding sites on proteins. Consequently, CPIs have emerged as promising targets for therapeutic intervention, particularly in areas like infectious diseases and cancer. However, developing effective drugs that target CPIs presents significant challenges. Carbohydrates are typically hydrophilic and lack the features that are traditionally associated with “drug-likeness,” such as lipophilicity and structural rigidity. This poses obstacles to the design and development of small-molecule inhibitors that can effectively compete with the native carbohydrate ligands for binding to their protein targets. This has limited the ability to use traditional drug discovery methods to target these interactions. The present research focused on addressing this critical challenge by exploring alternative strategies to target CPIs. Specifically, the researchers focused on identifying and characterizing novel scaffolds or molecular frameworks that can effectively inhibit Ca²⁺-dependent lectin-carbohydrate interactions. The overall goal was to establish new strategies for the discovery and development of effective drug-like inhibitors that can modulate CPIs, opening up new avenues for therapeutic interventions in a variety of diseases.

The literature review extensively covered existing approaches to targeting carbohydrate-protein interactions, highlighting the challenges posed by the hydrophilic nature of carbohydrates and the lack of druggable features. Studies demonstrating the therapeutic potential of modulating CPIs in various diseases, particularly infectious diseases and cancer, were discussed. The authors reviewed previous work on lectins and their roles in pathogenesis. Existing strategies, including the use of carbohydrate-based glycomimetics, were critiqued, noting limitations like poor pharmacokinetic properties, which limit their clinical viability. This emphasized the need for non-carbohydrate drug-like inhibitors, setting the stage for the researchers' approach of utilizing metal-binding pharmacophores (MBPs).

The study employed a multi-faceted approach combining computational and experimental techniques. Four fragment libraries were used in the study, including a focused library of metal-binding pharmacophores (MBPs). These libraries were screened against four clinically relevant lectins: DC-SIGN, Langerin, LecA, and LecB, as well as a control lectin, BambL. Initial screening was performed using computational methods followed by experimental validation using various biophysical techniques. Nuclear Magnetic Resonance (NMR) spectroscopy, including 19F and 1H-15N HSQC/TROSY NMR, played a crucial role in identifying and characterizing the interactions between fragments and lectins. Specifically, 19F NMR was used for screening the fragment libraries, and 1H-15N HSQC/TROSY NMR was employed to study the interactions between identified hits and 15N-labeled lectins. In addition to NMR, other techniques were used, such as surface plasmon resonance (SPR), fluorescence polarization (FP), and protein-observed 19F (PrOF) NMR. These techniques provided complementary information regarding binding affinity, selectivity, and the nature of the interactions between the identified fragments and their target lectins. Structure-activity relationship (SAR) studies were conducted using commercially available analogs of the lead fragments to investigate how the modifications impacted their affinity and selectivity for the target lectins. Docking studies were performed to provide structural insights into the binding modes of identified compounds. The effectiveness of the identified inhibitors in a cellular context was evaluated using a cell-based assay (cellFy). The study used crystallography to obtain high-resolution structures of the lectin-inhibitor complexes, which provided crucial details on the binding interactions.

The study identified two promising scaffolds, hydroxamates and malonates, as novel metal-binding pharmacophores (MBPs) for targeting Ca²⁺-dependent lectins. Hydroxamates were shown to effectively target the orthosteric site of LecA, a lectin from Pseudomonas aeruginosa, in a Ca²⁺-dependent manner. Structure-activity relationship studies using numerous hydroxamate analogs revealed that the hydroxamic acid functional group plays a key role in binding, and the specific presentation of this group dictates its binding affinity and specificity. The most potent hydroxamate derivative (35) exhibited a dissociation constant (KD) in the millimolar range, indicating relatively weak affinity compared to what is typically observed with metalloenzyme inhibitors. Malonates, on the other hand, showed broader activity against LecA, LecB, and DC-SIGN, which demonstrated that they can target lectins with varying numbers of Ca²⁺ ions in or near the orthosteric site. Structure-activity relationship studies of malonate analogs revealed that the core malonic acid group is essential for Ca²⁺-dependent binding. The study demonstrated that the scaffold (61) had a stronger tendency to bind LecA, while LecB and DC-SIGN showed a preference for scaffolds 58 and 62, respectively. Importantly, malonate 58 demonstrated low cytotoxicity and specific binding to DC-SIGN-expressing cells in vitro. The crystal structure of LecA in complex with hydroxamate 35 revealed the galactose-mimicking properties of the inhibitor, showing that the hydroxamic acid group coordinates Ca²⁺ and forms hydrogen bonds mimicking interactions of galactose with LecA. The findings demonstrated that the selectivity of both hydroxamates and malonates is tunable, providing valuable insights for designing lectin inhibitors with tailored specificity. Overall, the study provided robust evidence that metal-coordinating fragments, particularly hydroxamates and malonates, are potent lead compounds for developing effective drug-like inhibitors for Ca²⁺-dependent lectins.

The findings of this study address the critical need for novel strategies to target carbohydrate-protein interactions, particularly those that are challenging to target with conventional drug discovery approaches. The identification of metal-binding pharmacophores (MBPs), such as hydroxamates and malonates, as effective inhibitors of Ca²⁺-dependent lectins represents a significant advance in the field. The observed selectivity and tunability of these MBPs suggest that they can serve as excellent starting points for developing more potent and selective inhibitors. The demonstration of cellular activity of malonate 58 further validates the therapeutic potential of this approach. The findings underscore the potential of fragment-based drug design (FBDD) for developing novel therapeutics targeting CPIs. This work has implications for addressing the urgent need for new treatments against various infectious diseases. The research also highlights the importance of considering the metal ion coordination in the design of novel inhibitors targeting carbohydrate binding sites on proteins, a strategy that warrants further investigation.

This research successfully identified hydroxamates and malonates as promising scaffolds for the development of drug-like inhibitors targeting Ca²⁺-dependent lectins. The study established the feasibility of using a fragment-based drug design approach to tackle the challenges of targeting undruggable carbohydrate recognition sites. The identified compounds exhibited specificity for different lectins and demonstrated the potential for tuning selectivity. Future research could focus on further optimizing these scaffolds through fragment evolution to enhance potency and selectivity, and on exploring their therapeutic efficacy in relevant disease models.

While the study demonstrated the potential of hydroxamates and malonates as lead compounds, the affinities of these compounds are relatively weak compared to those typically observed with traditional drug molecules. Further optimization is required to enhance their potency. The in vitro cellular assays provided preliminary evidence of cellular activity, but in vivo studies are needed to confirm their efficacy in a living organism. The SAR studies were conducted using commercially available analogs, and the range of chemical modifications explored may not be exhaustive. More comprehensive SAR studies could reveal additional opportunities for improving compound properties.