Hedgehog signalling: overcoming a prickly problem to reveal an oxysterol target

Lipidomics Gateway (25 January 2012) [doi:10.1038/lipidmaps.2012.2]

Oxysterols can directly activate the transmembrane protein Smoothened through an allosteric mechanism to induce Hedgehog pathway signalling.

Macmillan Image Bank

The Hedgehog (Hh) signalling pathway is one of a core set of pathways that is indispensible for embryonic development but that, when dysregulated, can lead to cancer, emphasizing the importance of understanding the underlying mechanisms of signal transduction. Binding of Hh ligands to the transmembrane protein Patched1 (Ptc1) inactivates Ptc1, allowing activation of another transmembrane protein, Smoothened (Smo), to signal downstream to induce transcriptional changes.

Specific oxysterols are known to activate Hh signalling and, as eliminating Smo can abrogate this effect, these bioactive lipids are likely to function at the level of Ptc1, Smo or an as-yet-undetermined intermediate step. Precisely how, though, has remained elusive. A particular problem is that sterols can induce cellular responses by directly affecting proteins or by altering the properties of lipid membranes, thereby indirectly influencing constituent proteins. However, using a combination of biochemical and pharmacological approaches, Nachtergaele et al. have overcome the challenge of distinguishing between these two potential mechanisms of action to show that oxysterols bind directly to Smo, allosterically activating it to induce downstream signalling.

Iso-octyl chain (IOC)-hydroxylated oxysterols are the strongest inducers of Hh signalling, and the researchers found that 20(S)-hydroxycholesterol (20(S)-OHC) induced Hh pathway activation more potently and effectively than seven other oxysterols that had a hydroxyl group at a different position on the IOC (as well as the 3-hydroxyl in cholesterol), indicating a high level of regioselectivity. Hh activation by oxysterols was also stereoselective, as 20(R)-OHC, an epimer of natural 20(S)-OHC (the only stereoisomer of 20(S)-OHC found in nature), failed to activate Hh signalling (measured by reporter gene transcription and the accumulation of Smo in the primary cilium). Furthermore, whereas natural 20(S)-OHC and its enantiomer, ent-20(S)-OHC, had similar effects on synthetic lipid bilayers, ent-20(S)-OHC was completely inactive in Hh pathway assays, indicating enantioselectivity and providing further evidence that oxysterols function through a specific protein target.

Cyclopamine is a small-molecule inhibitor of Hh signalling that binds directly to Smo. Nachtergaele et al. observed that the half-maximal inhibitory concentration (IC₅₀) of cyclopamine was almost 10-fold lower when activated by natural 20(S)-OHC than when activated by the ligand Sonic Hh, indicating that these two agonists activate Hh signalling by different mechanisms. The fact that increasing doses of natural 20(S)-OHC did not alter the IC₅₀ of cyclopamine indicates that natural 20(S)-OHC and cyclopamine are unlikely to compete for the same binding site. However, increasing doses of cyclopamine reduced the maximum level of Hh pathway activity induced by saturating concentrations of natural 20(S)-OHC, indicating that cyclopamine noncompetitively inhibits natural 20(S)-OHC. This being the case, natural 20(S)-OHC should show allosteric interactions with Smo activators that compete with cyclopamine, such as SAG or purmorphamine. Indeed, each of these Smo agonists synergized with natural 20(S)-OHC to induce Hh pathway activation.

Finally, Nachtergaele et al. confirmed biochemically that Smo is the receptor for natural 20(S)-OHC by demonstrating that 20(S)-OHC immobilized on magnetic beads could capture recombinant, tagged Smo expressed in Smo-deficient cells. This binding was inhibited in a dose-dependent manner by adding free natural 20(S)-OHC; similarly, natural 20(S)-OHC eluted recombinant Smo from the beads, whereas its enantiomer, ent-20(S)-OHC, failed to do so.

Nachtergaele et al. have therefore provided biochemical and pharmacological evidence that natural 20(S)-OHC interacts directly with Smo to activate Hh signalling, and that the site of interaction is distinct from that to which cyclopamine and SAG bind, meaning that SAG and natural 20(S)-OHC show a positive allosteric interaction. From a therapeutic perspective — as the authors highlight — Smo might therefore be amenable to allosteric regulation, providing a previously unexplored strategy to add to the numerous current drug discovery efforts aimed at targeting Hh signalling.

Katrin Legg