🧬 biology

Prediction of the self-association of the nuclear inhibitor of DNA binding and differentiation proteins

This study utilizes computational biophysics and molecular dynamics simulations to demonstrate that ID1 and ID4 proteins form stable HLH-mediated dimers, revealing key interaction patterns that regulate their nuclear concentration and function, thereby offering potential targets for disrupting pathological interactions in cancer.

Original authors: Youssef Ahmed, Sohila Mo'men, Esraa Ali, Heba Hany, Sama Sayed, Nada A. Elbendary, Rehab Hassan, Abdo A. Elfiky

Published 2026-07-13

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Youssef Ahmed, Sohila Mo'men, Esraa Ali, Heba Hany, Sama Sayed, Nada A. Elbendary, Rehab Hassan, Abdo A. Elfiky

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine a group of four molecular "troublemakers" inside our cells, known as ID proteins (ID1, ID2, ID3, and ID4). Their job is to act like a chaotic crowd at a concert, constantly bumping into other proteins to stop them from doing their work. Specifically, they stop other proteins from reading DNA instructions, which is crucial for telling cells when to grow, stop growing, or change into something new. When these ID proteins go into overdrive, they can help cancer cells grow out of control.

For a long time, scientists knew these ID proteins liked to team up with other types of proteins to cause trouble. But this study asked a different question: Do these ID proteins ever decide to team up with themselves? Do ID1 and ID1 hold hands? Does ID1 grab ID4? And if they do, does that change how they behave?

To find out, the researchers didn't use test tubes or microscopes. Instead, they built a super-detailed digital movie of these proteins using powerful computers. They created 3D models of the four ID proteins, which are naturally floppy and messy (like uncooked spaghetti) rather than having a rigid shape. They then ran a 100-nanosecond simulation—which, in computer time, is like watching a movie of these proteins dancing, stretching, and twisting in a virtual cell environment.

The Chaotic Solo Dance
First, the team watched the proteins dance alone. The results were exactly what you'd expect from a floppy noodle: they were wild. The proteins drifted around, changing shape constantly. The computer measured how much they moved (called RMSD) and found they were all over the place, with ID4 moving the most, reaching a drift of up to 36.1 Ångströms. They were so flexible that they kept shrinking and expanding, trying to find a comfortable shape but never quite settling down. It was a lot of energy, but no stable structure.

The Power of the Handshake
Then, the researchers made the proteins pair up. They simulated ID1 holding hands with another ID1 (a homodimer) and ID1 holding hands with ID4 (a heterodimer). The difference was like night and day.

When these proteins paired up, the chaos stopped. The wild, floppy dancing turned into a synchronized, stable waltz.

The "drift" (RMSD) dropped dramatically. While the solo ID1 drifted by about 26–29 Ångströms, the ID1-ID1 pair only drifted by about 6.84 Ångströms.
The ID1-ID4 pair was also very stable, drifting by about 9.93 Ångströms.
Most importantly, the proteins stopped flailing their ends around. The "flexibility" of the pair was cut down significantly, with no part of the pair moving more than 10 Ångströms, whereas the solo proteins had parts flailing up to 25 Ångströms.

It's as if the proteins found a partner that gave them a sense of direction. By locking arms, they became a compact, stable unit instead of a loose, floppy mess.

Who Likes Who Best?
The team used a digital scale to weigh how tightly these pairs stuck together. They found that both pairs were stable, but one was a bit more "in love" than the other.

The ID1-ID1 pair stuck together well, mostly because their shapes fit together perfectly like puzzle pieces, creating a tight, hydrophobic (water-fearing) hug.
The ID1-ID4 pair, however, showed an even stronger bond in the simulation. Their total binding energy was calculated at -41.55 kcal/mol, which is lower (and therefore stronger) than the ID1-ID1 pair's -28.07 kcal/mol.

Why was ID1-ID4 stronger? The computer analysis showed that while ID1-ID1 relied on a tight physical fit, the ID1-ID4 pair was held together by a massive network of electrical attractions (electrostatic interactions). It was like they were magnetized to each other. However, this strong magnetic pull came with a cost: it took a lot of energy to pull the water molecules away from them to let them touch (a "desolvation penalty"). Despite this cost, the electrical attraction won out, making the ID1-ID4 pair the most energetically favorable match in the simulation.

What This Means
The study suggests that these ID proteins don't just randomly bump into each other; they have specific preferences. The simulation indicates that ID1 and ID4 have a particularly strong tendency to form a team. If this happens in real cells, it might mean that ID1 and ID4 are busy holding hands with each other, which could lower the amount of "free" ID1 and ID4 floating around to do their usual job of blocking other proteins.

The researchers are careful to note that this is a prediction based on computer simulations. They haven't seen this happen in a living cell yet, but the digital evidence is strong. The study rules out the idea that these proteins are just random, unstable blobs; instead, it suggests that when they pair up, they become stable, structured units. This discovery opens a new door for understanding how these proteins work in cancer, hinting that breaking up the ID1-ID4 team might be a way to stop them from causing trouble. But for now, this is a fascinating glimpse into the molecular dance floor, where the most stable couples might be the ones changing the rules of the game.

Technical Summary: Prediction of the Self-Association of Nuclear Inhibitor of DNA Binding and Differentiation Proteins

Problem Statement
The Inhibitor of DNA binding and differentiation (ID) proteins (ID1–ID4) are helix-loop-helix (HLH) transcriptional regulators critical for cell cycle progression and differentiation. While their role as dominant-negative regulators via heterodimerization with E-proteins is well-established, their biophysical behavior regarding self-association (homodimerization) remains poorly understood. In the context of malignancy, where ID proteins are often overexpressed, the balance between monomers, homodimers, and heterodimers may dictate protein stability, nuclear localization, and function. Specifically, the structural determinants governing the self-association of ID proteins and the thermodynamic stability of these complexes in physiological environments require elucidation to identify potential therapeutic targets for disrupting pathological interactions.

Methodology
This study employed a multi-stage computational biophysics and bioinformatics pipeline to model and analyze ID protein self-association:

Structural Modeling: Due to the lack of experimental structures for these intrinsically disordered proteins, 3D models for ID1–ID4 were generated using AlphaFold3, I-TASSER, and Phyre2. Models were rigorously validated using MolProbity (Ramachandran plots, rotamer analysis, clash scores) and refined via the GalaxyRefine server. AlphaFold3-derived models were selected as the primary candidates due to superior stereochemical quality.
Monomer Molecular Dynamics (MD): Refined monomer structures underwent 100 ns all-atom MD simulations in explicit solvent (TIP3P) using the CHARMM36m force field and GROMACS. Trajectories were analyzed for Root-Mean-Square Deviation (RMSD), Root-Mean-Square Fluctuation (RMSF), Radius of Gyration (Rg), Solvent Accessible Surface Area (SASA), and intramolecular hydrogen bonding to characterize conformational dynamics.
Clustering: To identify representative conformations, trajectory clustering was performed using the Fraction of Common Contacts (FCC) metric rather than traditional RMSD, to better capture structural similarity in large, flexible systems.
Protein-Protein Docking: Representative conformations from the most populated clusters were docked using HADDOCK to investigate both homodimeric (ID1:ID1, ID2:ID2, etc.) and heterodimeric (e.g., ID1:ID4) interactions. The HLH domains were defined as active sites. Complexes were ranked by HADDOCK scores and predicted binding affinity (PRODIGY).
Complex Stability and Binding Free Energy: The top-scoring complexes (ID1–ID1 and ID1–ID4) underwent additional 100 ns MD simulations to verify stability. Binding free energies ( $\Delta G_{binding}$ ) were calculated using the Molecular Mechanics Poisson-Boltzmann Surface Area (MM/PBSA) method, decomposing energy contributions into van der Waals, electrostatic, polar solvation, and non-polar solvation terms.

Key Results

Structural Modeling: AlphaFold3 consistently outperformed I-TASSER and Phyre2 in initial model quality, particularly regarding rotamer percentages and Ramachandran favored regions. Post-refinement with GalaxyRefine yielded high-quality structures (e.g., ID3 achieved 100% Ramachandran-favored residues) suitable for downstream analysis.
Monomer Dynamics: MD simulations confirmed the intrinsically disordered nature of ID1–ID4. All monomers exhibited high RMSD values (26–30 Å) and significant flexibility (RMSF), particularly at the N- and C-termini. However, all proteins underwent spontaneous hydrophobic compaction (decreasing Rg and SASA) and formed transient intramolecular hydrogen bonds, suggesting a transition from extended to partially structured ensembles in solution.
Dimerization Stability: Docking and subsequent MD simulations revealed that dimerization significantly stabilizes the ID proteins.
- RMSD Reduction: The RMSD of the ID1–ID1 homodimer (6.84 Å) and ID1–ID4 heterodimer (9.93 Å) was approximately three times lower than their respective monomers, indicating that dimerization quenches conformational entropy.
- Flexibility: Both dimers showed drastically reduced RMSF (mean ~3.8–4.2 Å) compared to monomers, with no residues exceeding 10 Å, confirming that the HLH interface imposes mutual conformational restraint.
- Compactness: Unlike monomers, the dimers maintained stable, compact quaternary structures (constant Rg) throughout the 100 ns simulations.
Binding Energetics (MM/PBSA):
- ID1–ID1 Homodimer: Stabilized primarily by van der Waals interactions ( $\Delta E_{VDW} = -43.70$ kcal/mol), indicating efficient hydrophobic packing. Total $\Delta G_{binding}$ was $-28.07$ kcal/mol.
- ID1–ID4 Heterodimer: Exhibited a stronger overall binding affinity ( $\Delta G_{binding} = -41.55$ kcal/mol). This complex was driven predominantly by electrostatic interactions ( $\Delta E_{EL} = -433.58$ kcal/mol), despite a substantial desolvation penalty ( $\Delta E_{GB} = +418.97$ kcal/mol).
- Comparison: The ID1–ID4 complex was found to be energetically more favorable than the ID1–ID1 homodimer, suggesting a preference for heterotypic association under the simulated conditions.

Significance and Claims
The authors claim that this study successfully maps the structural foundations of ID protein self-association, demonstrating that these proteins can form stable homodimers and heterodimers via their HLH domains. The primary finding is that the ID1–ID4 heterodimer exhibits a stronger binding affinity and greater thermodynamic stability compared to the ID1–ID1 homodimer.

The paper posits that the formation of the ID1–ID4 complex may serve as a mechanism to lower the concentration of free ID1 and ID4 monomers in the nucleus, thereby modulating their biological activity. Given the role of ID proteins in cancer progression and their potential as therapeutic targets, the identification of these specific interaction patterns and binding hotspots provides a molecular basis for understanding how ID protein activity is regulated. The authors conclude that these computational insights lay the groundwork for future experimental investigations into the biological relevance of the ID1–ID4 complex and the development of inhibitors targeting these pathological interactions.

More like this