What is PROTAC?

Proteolysis Targeting Chimeras (PROTACs) are a transformative innovation at the intersection of molecular biology, drug discovery, and clinical medicine. PROTACs offer a radically novel approach to regulating and controlling the quantity of intracellular proteins for therapeutic objectives, and they constitute a developing discipline known as targeted protein degradation (TPD). By utilizing the ubiquitin-proteasome system, the cell’s natural protein disposal mechanism, PROTACs physically destroy disease-causing proteins, in contrast to traditional small-molecule inhibitors that just block the target protein’s active site. This overview offers a thorough investigation of PROTAC technology, including its background, design, mode of action, applications, clinical advancements, difficulties, new developments, and prospects for the future.

Historical Foundations and Evolution:

The shortcomings of traditional medication treatments served as the impetus for the idea of targeted protein breakdown. Many disease-related proteins, including transcription factors, scaffolding proteins, and some receptor classes, are “undruggable” by conventional means because they lack accessible active sites for small-molecule inhibition. Scientists realized in the late 20th and early 21st centuries that a practical method of selectively and programmably controlling the fate of such proteins would be to take over the cell’s own quality control system.

The initial PROTAC molecules were peptide-based and somewhat big, which created major challenges for immunogenicity, pharmacokinetics, and cellular absorption. The discovery of entirely small-molecule PROTACs in 2008 marked a significant advancement. These were employed as ligands for well-characterized E3 ubiquitin ligases, including von Hippel–Lindau (VHL), cereblon (CRBN, accessible by thalidomide derivatives), and MDM2 (nutlin-3a). More drug-like compounds with enhanced in vivo efficacy and safety profiles were produced during the following ten years as the chemistry and repertory of E3 ligase ligands improved.

  

Principles of Targeted Protein Degradation

The intracellular ubiquitin-proteasome system (UPS), a regulatory process essential for preserving protein homeostasis, lies at the heart of the PROTAC strategy. An enzymatic cascade including the E1 (activating), E2 (conjugating), and E3 (ligating) enzymes catalyses the UPS’s mechanism of covalently attaching the tiny protein ubiquitin to lysine residues on substrates. The massive multi-subunit protease complex known as the 26S proteasome then recognizes and breaks down polyubiquitinated proteins.

PROTAC Structure and Mechanism

A PROTAC is a modular, heterobifunctional molecule consisting of three main components:

• A ligand that binds the protein of interest (POI), also called the “warhead”
• A ligand for an E3 ubiquitin ligase
• A chemical linker joining these two moieties

When a PROTAC enters a cell, it bridges the POI with the E3 ligase, forming a ternary complex. This physical proximity catalyzes the transfer of ubiquitin to the POI, effectively “tagging” it for proteasomal degradation. Notably, PROTACs function catalytically—with one molecule able to degrade multiple copies of the target protein, offering much greater potency and often lower dosing requirements than inhibitors.

Information on PROTACs, Warheads, E3 Ligands and Linkers from ProtacDB (Ref1)

PROTACS	Warheads	E3 Ligands	Linkers
6,111	570	107	2753

 

PROTACS in clinical development from various therapeutic areas

PROTAC Modeling using MolSoft:

• Start with structural data from the target protein and E3 ligase (CRBN or VHL), each bound to their respective ligands.
• Use these binary complexes as validated templates for accurate ternary modeling.
• Introduce the designed PROTAC, featuring a target warhead and an E3-ligase ligand connected by a flexible linker.
• Position the PROTAC by aligning its ends to the respective binding sites in the target and E3 complexes.
• Optimize the linker using internal coordinate Monte Carlo (MC) simulations to sample possible conformations (Ref2).
• Allow local side chain flexibility to enable induced-fit adjustments and avoid steric clashes.
• Generate an ensemble of low-energy ternary conformations.
• Rescore conformers with a docking-specific scoring function to identify the most stable assemblies (Ref3).

Example of PROTAC Modeling:

• Load the target protein of interest (POI) and E3 ligase structures (e.g., bromodomain with PDB ID 3MXF and cereblon with PDB ID 4CI1) into ICM software.
• Remove non-essential components from PDB files, such as crystallographic water, buffer ions, and unrelated ligands, leaving only receptor proteins and bound ligands.
• Use CTRL + left-click to select unwanted molecules in ICM and delete them via right-click menu.
• Convert cleaned PDB files into ICM objects, necessary for docking and modeling tasks within ICM-Pro.
• Prepare the PROTAC molecule for docking; sources can be sketched in ICM, imported from MOL files, or extracted from experimental complexes.
• In this example, the PROTAC structure is taken from PDB entry 6BN7, which includes bromodomain, cereblon, and the PROTAC molecule itself.

3D-Grpahical window after preparation:

PROTAC Docking Setup:

• Open the PROTAC Model Builder in ICM via Docking > Protein-Protein Docking > PROTAC Model Builder.
• Select the prepared receptor and ligase ICM objects from dropdown menus (e.g., target bromodomain as a_3mxf, cereblon as a_4ci1).
• Specify the PROTAC molecule location by entering the name of the table containing the PROTAC(s) and, if multiple present, provide the molecule index.
• Set simulation parameters, including the effort value to control Monte Carlo simulation thoroughness; for detailed sampling, use effort = 3; for parallel runs (e.g., 5), effort = 1 is sufficient.
• Optionally, calculate RMSD of the docked PROTAC against a known crystal structure like PDB 6BN7 (enter PDB ID, chain, and molecule name).
• Click OK to start the simulation.
• Track progress via the progress bar and Windows > Background Jobs menu.

Below is the process after the completion of the simulation, a stack of multiple conformations will be embedded into the workspace and PROTAC results table consists of multiple scoring parameters are tabulated in detail below

Scoring Parameters

Parameter	Detailed definition
I	rank order
Ener	total energy is the main score for ranking
RMSD	RMSD to reference structure (if provided)
comm	user defined comments as string
ey	force-field energy
sf	surface energy term
vw	nonbonded interatomic pairwise interactions van der Waals energy.
el	Electrostatic energy
deltaSurf	deltaSurf is the solvent-accessible surface area buried between the two proteins (the degrader small molecule is not included in this calculation)
EnerSc	Energy of side chains

 

Results Analysis:

• The “ener” column shows the internal energy of each PROTAC conformer, used to filter out unrealistic, high-energy structures.
• Selecting the lowest internal energy conformer does not always mean biological relevance, especially for flexible molecules like PROTACs with multiple equilibrium conformations.
• Clustering identifies conformer geometries that are both energetically favourable and entropically accessible, indicating stable, biologically meaningful binding modes.
• Dense clusters of low-energy conformers suggest reproducible and thermodynamically reasonable conformations.
• Isolated low-energy conformers without nearby structures may indicate strained or rare geometries, less likely biologically relevant.
• Top 10-15 complexes are selected for molecular dynamics simulations to assess the stability of the complexes as a final filter

Benchmarking:

Three well-known PROTAC modeling tools—PRosettaC, MOE, and ICM (MolSoft’s engine)—are methodically compared against a larger collection of ternary PROTAC complexes in this study by Rovers et al (Ref4). They evaluate RMSD deviations from experimentally established structures, ranking accuracy, and the ability to reproduce near-crystal poses. The MolSoft ICM (ICM PROTAC modeling) approach builds PROTAC ternary complexes by optimizing side chains and flexibly sampling linker torsions, followed by scoring using a docking tool that is aware of PROTAC.

The Cα-RMSD between the reference and projected protein-ligand locations is a crucial statistic; a lower RMSD denotes a more accurate posture. ICM frequently produces low RMSD forecasts and good ranking in many systems, while the benchmarking showed that no single tool dominates all scenarios. Limitations include sample bottlenecks, inaccurate force fields, and scoring algorithms that are insensitive to entropic effects, according to the authors. To increase confidence in predicted models, they advise combining RMSD filtering, consensus scoring, and clustering. All things considered, this benchmark offers helpful advice on how to choose and improve computational tools (such as MolSoft’s ICM) for accurate PROTAC ternary complex modeling.

Success story:

By building bivalent PROTACs utilizing copper-catalyzed azide–alkyne cycloaddition (CuAAC “click”) chemistry to connect a known TDP1-binding imidazopyridine scaffold to an E3 ligase recruiter, Zhao et al (Ref5). hope to transform TDP1 inhibitors into degraders.
The authors created a series of chimeric PROTACs with linkers comprising 1, 2, and 3-triazoles by starting with the crystal structures of lead inhibitors bound to TDP1.
The viability of the bivalent method was demonstrated by the fact that many variants maintained TDP1 binding and inhibitory activity even after bulk was added via linkers and E3-recruiting moieties. Enabling the selective breakdown of TDP1 is intended to improve the anticancer therapeutic efficacy of TOP1 inhibitors.

The ternary complex geometries were modeled using MolSoft’s ICM / PROTAC modeling tools. For instance, a lead imidazopyridine-derived TDP1 ligand was docked to a thalidomide recruiter via a linker to the CRBN E3 ligase, confirming that the linker could plausibly bridge the two binding sites. This simulation aids in directing linker optimization and bolsters the structural justification of the PROTAC designs. Overall, the work effectively illustrates a synthetic click-based pathway to TDP1-directed PROTACs and emphasizes how structure-guided modeling, such as that done with MolSoft, can be used to create bifunctional degraders for difficult targets like TDP1.

References:

• https://cadd.zju.edu.cn/protacdb/
• R Abagyan, M Totrov “Biased probability Monte Carlo conformational searches and electrostatic calculations for peptides and proteins” Journal of Molecular Biology. 1994 Jan 21;235(3):983-1002
• R Abagyan et al “ICM—A new method for protein modeling and design: Applications to docking and structure prediction from the distorted native conformation” Journal of Computational Chemistry. 1994 Vol 15 (5) 488-506
• Rovers et al “Benchmarking Methods for PROTAC Ternary Complex Structure Prediction” Journal of chemical information and modeling. 2024 Aug 12;64(15):6162-6173
• Xue Zhi Zhao et al “Application of a bivalent “click” approach to target tyrosyl-DNA phosphodiesterase 1 (TDP1)” RSC Medicinal Chemistry., 2025, 16, 1969