Deep learning discovers minimalist conopeptide [ΔQP,S8R]SS1 as potent α7 nAChR antagonist with nanomolar potency.
Background
The α7 nicotinic acetylcholine receptor (nAChR) is a clinically significant target implicated in various neurological and inflammatory disorders. However, developing highly selective and potent peptide-based modulators for this receptor has been challenging due to limited structural insights into conopeptide interactions and the inherent difficulty in screening natural product libraries. Current pharmacological tools for nAChRs, particularly specific subtypes like α7, are often scarce, hindering detailed functional characterization and therapeutic development. This gap necessitates innovative approaches to accelerate the discovery and optimization of novel peptide therapeutics targeting this crucial receptor.
Study Design
Researchers developed an integrated pipeline combining deep learning, structural biology, computational modeling, and electrophysiology to discover and optimize α7 nAChR-targeting conopeptides. They utilized an ESM-2 protein language framework to create a deep learning model, enabling efficient screening of 689 disulfide-poor conopeptides. This initial screening identified SS1, a novel antagonist. Subsequent structure-activity relationship (SAR) studies systematically optimized SS1 to [ΔQP,S8R]SS1. Functional characterization was performed via electrophysiology, while structural insights were gained through cryo-EM and computational modeling of the α7 nAChR bound to [S8R]SS1.
Results
The deep learning-driven pipeline successfully identified SS1, a novel α7 nAChR antagonist. Systematic optimization through SAR studies yielded [ΔQP,S8R]SS1, a minimalist peptide demonstrating significantly enhanced properties. This optimized conopeptide achieved nanomolar potency with an IC50 = 49.2 nmol/L against the α7 nAChR, alongside improved selectivity and stability. Structural analysis using cryo-EM resolved the 3.3 Å resolution structure of α7 nAChR bound to [S8R]SS1. This revealed a unique binding mode stabilized by hydrogen bonds, hydrophobic interactions, and glycan contacts. Computational modeling further elucidated the inhibitory mechanism by identifying hybrid receptor conformations (closed/desensitized) induced by the peptide. This integrated approach represents a significant advancement in peptide discovery.
[ΔQP,S8R]SS1 achieved an IC50 = 49.2 nmol/L for α7 nAChR antagonism, demonstrating nanomolar potency and improved selectivity.
Key Findings
- A deep learning model screened 689 disulfide-poor conopeptides to identify novel α7 nAChR antagonists.
- The conopeptide SS1 was discovered and optimized to [ΔQP,S8R]SS1 via structure-activity relationship studies.
- [ΔQP,S8R]SS1 demonstrated nanomolar potency against α7 nAChR with an IC50 = 49.2 nmol/L.
- The optimized peptide exhibited enhanced selectivity and improved stability.
- Cryo-EM resolved the 3.3 Å resolution structure of α7 nAChR bound to [S8R]SS1, revealing a unique binding mode.
Why It Matters
This study presents a transformative deep learning-to-experiment framework that significantly accelerates the discovery and optimization of nature-inspired peptide therapeutics, particularly for challenging targets like α7 nAChR. For peptide users and biohackers, this methodology highlights the potential of AI-driven design to yield highly potent and selective compounds, potentially reducing the trial-and-error in peptide development. The discovery of [ΔQP,S8R]SS1 provides a novel, minimalist antagonist with nanomolar potency, offering a new pharmacological tool for studying α7 nAChR function and potentially serving as a lead compound for future drug development. This approach could be broadly applicable, enabling faster translation of natural product diversity into usable protocols and therapeutic candidates.
conopeptide
alpha7-nachr
deep-learning
drug-discovery
antagonist
in-vitro