Deep learning structural insights into heterotrimeric alternatively spliced P2X7 receptors

Area of Science:

• Molecular Biology
• Structural Biology
• Biophysics

Background:

• P2X7 receptors (P2X7Rs) are ATP-gated ion channels formed by three subunits.
• Alternative splicing of the P2RX7 gene generates various P2X7R subunit structures, influencing receptor function.
• Understanding the structure of these heterotrimeric receptors is essential for elucidating their biological roles.

Purpose of the Study:

• To validate AlphaFold2-Multimer (AF2M) for predicting P2X7R structures.
• To generate structural models of heterotrimeric P2X7Rs incorporating alternatively spliced variants.
• To provide a structural basis for understanding the function of diverse P2X7R assemblies.

Main Methods:

• Utilized AlphaFold2-Multimer (AF2M) for predicting trimeric P2X7 receptor structures.
• Validated AF2M predictions against cryo-EM data of wild-type rat P2X7A receptors (PDB ID: 6U9V).
• Generated and analyzed models of heterotrimeric P2X7Rs with various splice variants (P2X7B, P2X7E, P2X7J, P2X7L) using AF2M.

Main Results:

• AF2M accurately predicted P2X7R structures and identified high-quality models.
• Validated AF2M models showed stability through molecular dynamics simulations.
• Structural models revealed missing key residues in splice variants compared to wild-type P2X7A, explaining functional differences.

Conclusions:

• AF2M is a reliable tool for predicting P2X7R structures, including complex heterotrimeric forms.
• The generated models offer structural insights into alternatively spliced P2X7Rs.
• These structure-based models facilitate further research into P2X7R function and therapeutic targeting.