One injection of fibrils spreads pathology

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

Kelvin Luk, Victoria Kehm, Jenna Carroll, Bin Zhang, Patrick O'Brien, John Trojanowski and Virginia Lee (Perelman School of Medicine, University of Pennsylvania). The paper's funding statement lists NIH grant NS053488, the JPB and Benaroya Foundations and the Jeff and Anne Keefer Fund. The cell-culture version came a year earlier from Laura Volpicelli-Daley, Luk and colleagues in the same lab.

What happened

In 2011 the group showed that tiny synthetic fibrils added to cultured neurons were taken up and made the neurons' own alpha-synuclein clump into Lewy-like inclusions, eventually impairing and killing the cells (Neuron 72:57-71). In 2012 they injected such fibrils once into the striatum (a movement-control region) of ordinary, non-transgenic mice. Clumps appeared at the injection site within 30 days, then in connected regions (ventral striatum, thalamus, cortex) over the next months. In the substantia nigra, inclusions matured from pale deposits into dense Lewy-body-like ones. There was no neuron loss at 30 days. Dopamine-neuron counts fell about 15% at 90 days and about 35% at 180 days, striatal dopamine dropped, and the mice declined on rotarod and wire-hang tests (wire-hang performance fell 53% from baseline at 30 days and 81% at 180 days). Open-field, Y-maze and tail-suspension tests showed no change. Group sizes were small: 3 to 4 mice per group for neuron counts, and 6 to 17 injected mice per time point for behaviour. The neighbouring ventral tegmental area was largely spared. Science 338:949-953, 16 Nov 2012.

Why it moved us toward a cure

It turned the prion-like idea from the graft findings (landmark 8) into a reproducible experiment: misfolded synuclein alone can start the whole sequence, from spread to cell death, in a normal brain. It showed that clumping, not just gene mutation, can drive neuron loss, and it made a testable target of "stop the spread." The model underlies preclinical work on antibodies and on synuclein-lowering drugs (landmark 29) and fits the staged climb in landmark 6.

Outcome

Advanced. The seeding model is widely used. The same misfolded-seed behaviour is what the spinal-fluid test detects (landmark 9). Open question: whether blocking spread slows human disease, which the Phase 3 antibody trial will test. Fibrils are an artificial trigger, and the mice show only partial loss, not the full human disease. An independent group in Tokyo (Masuda-Suzukake 2013, Brain) reproduced the seeding in wild-type mice with fibrils, and also with insoluble synuclein from dementia-with-Lewy-bodies brains. No correction, retraction or integrity flag was found for Luk 2012 or the other papers cited here.

Sources

• Luk et al., Science 2012;338:949-953, DOI 10.1126/science.1227157 (open-access author manuscript)
• Volpicelli-Daley et al., Neuron 2011;72:57-71, DOI 10.1016/j.neuron.2011.08.033, PubMed 21982369
• Luk et al., PNAS 2009;106:20051-20056, DOI 10.1073/pnas.0908005106, PubMed 19892735 (the cell-culture seeding paper)
• Masuda-Suzukake et al., Brain 2013;136:1128-1138, DOI 10.1093/brain/awt037 (independent replication)
• ScienceDaily on Luk 2012 (press release gives rounder figures, "one-sixth" at 3 months and "about half" at 6 months, that differ from the paper's 15% and 35%; the paper's numbers are used above)
• Not yet verified: the fibril dose (given only in the supplement, not read); per-group mouse numbers beyond the ranges above.