How a shape-shifting Parkinson's protein could unlock the next generation of treatments

Mutations in a single protein, called LRRK2, are among the most common genetic causes of Parkinson's disease, and at least four clinical trials are already in flight trying to quiet the protein down. The trouble is that drug designers have never been able to see LRRK2 clearly enough to design against. Researchers at Weill Cornell Medicine have now resolved the protein in both its "on" and "off" states, and the work does not finish the race so much as redraw the next lap: the unsolved problem is no longer visibility, it is selectivity.

LRRK2 sits at the center of a long-running therapeutic hunt. Mutations that push the protein into an overactive state are among the most common genetic drivers of Parkinson's, and even patients who carry no such mutation often show elevated LRRK2 activity. With at least four clinical trials already in flight against the protein, a clearer picture of its structure has been on every Parkinson's chemist's wish list.

LRRK2 is a large protein built from seven interlocking domains, and two of those domains do most of the disease-relevant work. One domain binds a small molecule called GTP. Like many such molecular switches, the protein is generally "on" when GTP is bound and "off" once that GTP has been converted into GDP. A second domain, the kinase, then attaches phosphate groups to other proteins, and it is that phosphorylation step that runs too hot in Parkinson's. Until now, drug designers had to work from an inferred picture of how those two domains sit together, and an inferred picture is not a map.

The new structures, published Aug. 10 in Cell, give chemists the actual map. Using electron microscopy and biochemistry, the team captured LRRK2 in both its active and inactive conformations and watched how the GTP-binding and kinase domains rearrange when the switch flips. That visibility turns a structural biology result into a design input. With both states resolved side by side, chemists can compare pockets directly and design molecules that prefer one conformation over the other, rather than screening blind.

The work was co-led by Dr. Samara Reck-Peterson. The team's contribution is not a drug. It is the shape of the target those drugs are aimed at.

The remaining problem is tissue selectivity. LRRK2 is not just a brain protein. It also does normal work in the immune system, in the lungs, and in the kidneys, where the same kinase activity that drives disease in neurons has legitimate housekeeping jobs. A drug that silences LRRK2 everywhere would silence it in the wrong places, which is why selectivity has been the open gate for the whole class. The new structures give chemists a way to design molecules that quiet the disease-driving "on" state in the brain while leaving the protein's other jobs intact. The four ongoing trials are now being asked to determine whether the chemistry can be made selective.

The structures land while LRRK2-targeting drugs are already in patients. Those trials were started against a target whose full active shape was a working assumption. The next readouts from those trials, and the next round of structure-guided compounds, are the test of whether the new map changes what those medicines can do.