AI-enforced ultra-large virtual screening is currently revolutionizing how we identify binders for challenging targets. In a landmark study, researchers targeted the CD28 immune receptor. This receptor is critical for T-cell activation but remains notoriously difficult to modulate with small molecules. Specifically, traditional screening methods often fail due to the receptor's shallow, solvent-exposed protein-protein interfaces. Consequently, the researchers utilized the PyRMD2Dock platform to overcome existing computational bottlenecks.

First, they docked a diverse subset of 2.4 million molecules. Next, they used these data to train 672 classification models. Subsequently, these models screened 46 million compounds from the Enamine REAL Diversity Space. This AI virtual screening CD28 approach identified 232 highly prioritized ligands. Ultimately, the experimental validation of 150 candidates yielded several potent binders.

Harnessing AI Virtual Screening for CD28 Binders

Specifically, lead compounds 100 and 104 demonstrated submicromolar affinities of 343.8 nM and 407.1 nM. These molecules successfully disrupted the CD28-CD80 interaction. Moreover, the compounds reduced cytokine secretion in primary human tumor-PBMC models. In addition, this study validates PyRMD2Dock as a scalable protocol for mining massive chemical libraries. Therefore, this breakthrough offers a new path for discovering modulators for "undruggable" immune receptors.

FAQ

What is the role of CD28 in the immune response?

CD28 is a critical costimulatory receptor on T-cells. It binds to CD80 or CD86 on antigen-presenting cells to provide the "second signal" required for full T-cell activation and survival.

Why is CD28 a difficult target for small-molecule drugs?

Targeting CD28 is challenging because its binding interface is shallow and solvent-exposed. Unlike typical enzymes with deep pockets, these protein-protein interfaces lack the structural features usually required for high-affinity small-molecule binding.

How does AI improve the drug screening process for immune receptors?

AI improves screening by rapidly analyzing millions of compounds that would be computationally impossible to dock using traditional methods. It learns from a smaller subset of data to predict high-affinity binders across massive chemical libraries.

Disclaimer: This content is for informational and educational purposes only. It does not constitute medical advice or a professional relationship. Refer to the latest local and national guidelines for clinical practice.

References

Upadhyay S et al. AI-Enforced Ultra-Large Virtual Screening Discovers Potent CD28 Binders. J Chem Inf Model. 2026 Jun 11. doi: 10.1021/acs.jcim.6c00859. PMID: 42273851.

Amendola G, Cosconati S. PyRMD: AI-powered Virtual Screening. J Chem Inf Model. 2021;61(10):4905-4914. doi: 10.1021/acs.jcim.1c00653.

Roggia M et al. Streamlining Large Chemical Library Docking with Artificial Intelligence: the PyRMD2Dock Approach. J Chem Inf Model. 2024. doi: 10.1021/acs.jcim.4c00072.