This article, with insights from Babu Boga, Vice President of Medicinal Chemistry and Xiang Li, President of the Chemical Division, BioDuro, was originally published in the DDW Fall 2025, pages 30-31. Visit the website: DDW Fall 2025

The pharmaceutical industry constantly seeks innovative methods to identify novel therapeutic agents. Traditional high-throughput screening (HTS) often faces challenges like high false-positive rates and limited chemical diversity. Fragment-based drug discovery (FBDD) offers a compelling alternative. FBDD utilizes small, low-molecular-weight chemical fragments (typically<300 Da) that bind weakly to a target protein. Their smaller size leads to higher ‘ligand efficiency’ and enables them to access cryptic binding pockets, resulting in higher hit rates than HTS. These identified fragment hits serve as ideal starting points for rational elaboration into potent and selective lead compounds, often yielding novel chemical scaffolds.    

The success of any FBDD campaign hinges critically on the quality and design of its fragment library. Unlike the vast, diverse libraries used in HTS, FBDD libraries are typically smaller, ranging from hundreds to a few thousand compounds, and are meticulously curated. A common approach emphasizes a rational design strategy often guided by computational methods, such as fingerprint-based approaches, to ensure broad chemical space coverage and diversity. Fragments are selected to represent a broad spectrum of key chemical functionalities essential for molecular recognition, including various hydrogen bond donors and acceptors, hydrophobic centers, aromatic rings, and ionizable groups, ensuring the library can probe diverse interaction types within a binding site. The library is also designed to achieve broad coverage of chemical space in terms of molecular shape and physicochemical properties, ensuring that fragments with different geometries and interaction profiles can effectively sample and fit into various contours of the target's binding pockets. Crucially, fragments are designed with "growth vectors", which are specific, synthetically tractable sites or functional groups that can be readily elaborated or modified in subsequent optimization steps without disrupting the initial, weak binding interaction. This foresight significantly streamlines the fragment-to-lead optimization process. Beyond computational design principles, fragments are also rigorously filtered based on "Rule of 3" criteria (molecular weight<300 Da, cLogP <3, hydrogen bond donors <3, hydrogen bond acceptors <3, rotatable bonds <3). Adherence to these guidelines ensures good aqueous solubility, chemical stability, and synthetic accessibility, all of which are paramount for successful downstream development and eventual drug-likeness.    

Following library design, initial fragment hits are identified via highly sensitive biophysical screening. These methods are preferred over traditional biochemical assays for their direct, label-free detection of weak affinities.

There are a few key biophysical technologies playing critical roles. Surface Plasmon Resonance (SPR), a real-time, label-free optical technique that monitors changes in refractive index at a sensor surface as fragments bind to an immobilized target protein. SPR provides comprehensive kinetic data, allowing for precise determination of binding affinity (KD), association (kon), and dissociation (koff) rates, offering invaluable insights into the binding mechanism. MicroScale Thermophoresis (MST) measures the directed movement of molecules in a microscopic temperature gradient, which changes upon ligand binding. MST is highly sensitive, requires minimal sample consumption, and can be performed directly in solution, making it suitable for a wide range of targets. Isothermal Titration Calorimetry (ITC) is considered the gold standard for thermodynamic characterization, directly measuring the heat released or absorbed during a binding event. It provides a complete thermodynamic profile (KD, enthalpy (ΔH), and entropy (ΔS)) of the interaction, offering deep insights into the driving forces of binding. Nuclear Magnetic Resonance (NMR) Spectroscopy, encompassing both ligand-observed (e.g., STD NMR) and protein-observed techniques, is a powerful tool for identifying fragment binders—even in complex mixtures—and mapping their binding sites, while protein-observed NMR can provide detailed structural insights into conformational changes induced by fragment binding. Differential Scanning Fluorimetry (DSF) and Thermal Shift Assays (TSA) measure the thermal stability of a protein, which often increases upon ligand binding. These are rapid, high-throughput, and cost-effective for initial hit identification and validation.

Critical structural characterization follows fragment hit identification. Precise, atomic level understanding of each fragment's binding mode is paramount for rational optimization and lead generation, preventing empirical and inefficient processes.

X-ray Crystallography (XRC) remains the gold standard for elucidating atomic-level fragment protein interactions. Through co-crystallization, it provides an unambiguous three-dimensional map of the binding site, revealing specific interactions (e.g., hydrogen bonds, hydrophobic contacts, π-stacking) and, crucially, identifying unoccupied pockets or ‘hotspots’ for growth. While traditionally applied to larger macromolecular complexes, recent advancements in Cryo-Electron Microscopy (Cryo-EM) resolution are making it increasingly viable for structural determination of protein-ligand complexes, particularly for challenging targets that are difficult to crystallize or are membrane proteins. For targets amenable to solution-state studies, NMR can complement XRC by providing insights into dynamic interactions, conformational changes, and the presence of multiple binding poses that might not be captured in a static crystal structure.

With precise structural insights into fragment binding, the focus shifts to optimizing these initial hits into more selective, drug-like lead compounds. This iterative phase employs strategic approaches: Fragment growing systematically adds chemical moieties to the initial fragment, extending into adjacent, unoccupied pockets identified by structural analysis. This aims to improve affinity and selectivity through new interactions while maintaining the original fragment's core binding. Fragment linking covalently joins two or more distinct fragments binding to separate but adjacent sites, often resulting in a significant, synergistic affinity increase from multiple interaction points. Alternatively, when two fragments are found to bind to overlapping regions of the binding site, they can be merged into a single, more potent molecule. This new compound incorporates the key binding features and favorable interactions of both initial fragments into a single, optimized scaffold. This optimization process is highly iterative, involving cycles of design, synthesis, biological evaluation, and further structural characterization.

Computational chemistry plays an increasingly vital role throughout the FBDD workflow, especially during optimization, by efficiently exploring chemical space and guiding rational design. Molecular docking predicts binding poses and affinities of proposed fragment modifications or new compounds within the target's binding site, thus prioritizing synthesis and offering initial interaction insights. Molecular Dynamics (MD) simulations provide dynamic insights into the behavior of the protein ligand complex over time. MD simulations can reveal transient interactions, conformational flexibility of both the protein and ligand, and the role of water molecules in binding. Advanced computational techniques such as Free Energy Perturbation (FEP) can accurately predict the relative binding affinities of small chemical modifications. By simulating the transformation of one molecule into another within the binding site, FEP can significantly accelerate lead optimization cycles by providing quantitative predictions of affinity changes. Large virtual libraries (VL) of compounds, often generated by enumerating synthetically accessible derivatives of a fragment or by combining known fragments, can be rapidly screened computationally. This helps identify optimal modification sites and prioritize compounds for experimental synthesis and validation, dramatically reducing the number of compounds that need to be physically made and tested. Furthermore, de novo design algorithms can generate entirely new molecular structures that are designed to optimally fit and interact with the target's binding site, often starting from the binding pose of an initial fragment.

Fragment-Based Drug Discovery represents a paradigm shift in drug discovery, offering an efficient, rational pathway to novel chemical leads. This workflow, encompassing rational fragment design, sensitive biophysical screening, high resolution structural elucidation, and computationally-informed optimization, provides a robust and systematic framework.