Back

Structure-based design of a chemical probe set for the 5-HT5A serotonin receptor

Kaplan, A. L.; Strachan, R. T.; Braz, J. M.; Craik, V.; Slocum, S.; Mangano, T.; Amabo, V.; O'Donnell, H. R.; Lak, P.; Basbaum, A.; Roth, B. L.; Shoichet, B. K.

2021-12-07 biophysics
10.1101/2021.12.06.471449 bioRxiv
Show abstract

The 5-HT5A receptor (5-HT5AR), for which no selective agonists and only a few antagonists exist, remains the least understood serotonin (5-HT) receptor. A single commercial antagonist (SB-699551) has been widely used to investigate central nervous system (CNS) 5-HT5AR function in neurological disorders, including pain. However, because SB-699551 has affinity for many 5-HTRs, lacks inactive property-matched controls, and has assay interference concerns, it has liabilities as a chemical probe. To better illuminate 5-HT5AR function, we developed a probe set through iterative rounds of molecular docking, pharmacological testing, and optimization. Docking over six million lead-like molecules against a 5-HT5AR homology model identified five mid-M ligand starting points with unique scaffolds. Over multiple rounds of structure-based design and testing, a new quinoline scaffold with high affinity and enhanced selectivity for the 5-HT5AR was developed, leading to UCSF678, a 42 nM arrestin-biased partial agonist at the 5-HT5AR with a much more restricted off-target profile and decreased assay liabilities vs. SB-699551. Site-directed mutagenesis supported the docked pose of UCSF678, which was also consistent with recent published 5-HTR structures. Surprisingly, property-matched analogs of UCSF678 that were either inactive across 5-HTRs or retained affinity for UCSF678s off-targets revealed that 5-HT5AR engagement is nonessential for alleviating pain in a mouse model, contrary to previous studies using less-selective ligands. Relative to SB-699551, these molecules constitute a well-characterized and more selective probe set with which to study the function of the 5-HT5A receptor. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=113 SRC="FIGDIR/small/471449v1_ufig1.gif" ALT="Figure 1"> View larger version (28K): org.highwire.dtl.DTLVardef@8fb24dorg.highwire.dtl.DTLVardef@2dd84org.highwire.dtl.DTLVardef@1d3346org.highwire.dtl.DTLVardef@1117331_HPS_FORMAT_FIGEXP M_FIG Table of Contents Graphic C_FIG

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.