Detailed 3D structures of LRRK2

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Deniston, Salogiannis, Mathea and colleagues, with Samara Reck-Peterson and Andres Leschziner (UC San Diego) as co-corresponding authors and Stefan Knapp (Goethe University Frankfurt) among the authors (Nature 588:344-349, online 19 August 2020). Watanabe, Buschauer, Böhning, with Elizabeth Villa (UC San Diego) as senior author (Cell 182:1508-1518, online 11 August 2020). Myasnikov, Zhu, Hixson and colleagues, with Ji Sun (St. Jude Children's Research Hospital) as senior author (Cell 184:3519-3527, online 8 June 2021). Funding: the Michael J. Fox Foundation supported the Deniston paper (grants 11425, 11425.02 and 18321) and the Watanabe paper (grant 11425, alongside an NIH New Innovator Award); the Myasnikov paper lists NIH grants (HL143037, AG053987) among its funders.

What happened

LRRK2 is a very large protein with several parts, including a kinase (the active enzyme) and a GTPase (a molecular switch). The Deniston paper described structural information on the enzyme as limited before 2020.

• Watanabe 2020. Cryo-electron tomography (freezing cells and imaging them in 3D) showed LRRK2 carrying a pathogenic mutation, assembled into a right-handed double helix around microtubules (the cell's internal rails). The pathogenic mutation was I2020T. The map was at 14 angstrom resolution, coarse but in cells. Two contact surfaces between LRRK2 units were found; mutating one stopped it binding microtubules.
• Deniston 2020. A 3.5 angstrom structure of the catalytic half of LRRK2, and an atomic model of LRRK2 on microtubules built from the Watanabe data. LRRK2's catalytic half alone formed filaments and blocked the motor proteins kinesin 1 and dynein in a test tube. Kinase inhibitors that stabilise an open conformation relieved this and reduced filament formation in cells, while inhibitors that stabilise a closed conformation did not.
• Myasnikov 2021. Structures of full-length human LRRK2 by cryo-electron microscopy, at 3.7 angstrom (single molecule) and 3.5 angstrom (dimer), with the C-terminal part at 3.1 angstrom. The kinase was captured in an inactive state, also seen with the common G2019S mutation. The authors identified a dimer surface and showed that single changes at this interface stopped filament formation in cells.

Why it moved us toward a cure

The Deniston paper described structural information as limited. These maps of the protein from landmark 7 showed how mutations act, and suggested different inhibitors could have different effects, which matters for drug choice. The full-length paper said the inactive kinase shape gives a framework for designing inhibitors that favour a particular shape.

Outcome

Advanced as a basic-science step. Whether any current drug was designed from these structures was not checked. A Mov Disord commentary on the Deniston paper exists (2020;35:2162); no correction or retraction is recorded for any of the three papers (Crossref).

Sources

• Deniston et al., Nature 2020;588:344–349, DOI 10.1038/s41586-020-2673-2
• Watanabe et al., Cell 2020, DOI 10.1016/j.cell.2020.08.004
• Myasnikov et al., Cell 2021, DOI 10.1016/j.cell.2021.05.004
• Abstracts and full texts read via Europe PMC and PMC (PMC7726071, PMC7869717, PMC8887629).
• Not yet verified: whether earlier low-resolution LRRK2 structures existed (I dropped the "first" in the title for that reason); which, if any, inhibitors were designed from these structures.