Knockout mice keep their neurons

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

Matthew Goldberg, Jie Shen and colleagues (Brigham and Women's Hospital and Harvard Medical School) for parkin (2003), DJ-1 (2005) and PINK1 (Kitada 2007); Fabiola Perez and Richard Palmiter (University of Washington) for an independent parkin line (2005). Funders were not checked.

What happened

People who inherit two faulty copies of PRKN (landmark 12), PINK1 (landmark 15) or DJ-1 (landmark 33) get early-onset Parkinson's. Researchers deleted each gene in mice.

• Parkin, Goldberg 2003 (J Biol Chem): mice were viable with normal brain structure. Dopamine outside cells in the striatum was raised and there were behavioural deficits, but the number of nigral dopamine neurons was normal up to 24 months of age.
• Parkin, Perez & Palmiter 2005 (PNAS): a second line showed no robust parkinsonism, with normal striatal catecholamines and no nigrostriatal dysfunction. They reported that some earlier differences were not reproducible across genetic backgrounds.
• DJ-1, Goldberg 2005 (Neuron): normal numbers of nigral neurons, but reduced evoked dopamine release and hypoactivity.
• PINK1, Kitada 2007 (PNAS): numbers of dopamine neurons and striatal dopamine unchanged; evoked dopamine release and striatal plasticity were impaired.

Why it moved us toward a cure

The negative result is a structural obstacle. A drug that protects neurons cannot be tested for survival in a mouse that does not lose them. It pushed the field towards other routes: toxin and synuclein-fibril models (landmark 43), human patient-derived neurons (landmark 47) and gene-edited human cell lines (landmark 272). One later experiment supports the idea that a second stress is needed: parkin-knockout mice that also accumulate mitochondrial DNA mutations ("Mutator" mice) did lose nigral dopamine neurons and developed an L-dopa-reversible movement problem (Pickrell et al., Neuron 2015).

Outcome

Open. Dopamine-signalling phenotypes in knockouts remain useful, but absence of cell death in these lines is the standing limitation; the Mutator cross shows one way round it. LRRK2 knockout and mutant mice were not checked here.

Sources

• Goldberg et al., J Biol Chem 2003;278:43628–43635, DOI 10.1074/jbc.M308947200
• Perez & Palmiter, PNAS 2005;102:2174–2179, DOI 10.1073/pnas.0409598102
• Goldberg et al., Neuron 2005;45:489–496, DOI 10.1016/j.neuron.2005.01.041
• Kitada et al., PNAS 2007;104:11441–11446, DOI 10.1073/pnas.0702717104
• Pickrell et al., Neuron 2015;87:371–381, DOI 10.1016/j.neuron.2015.06.034, PMID 26182419
• Not yet verified: full texts (abstracts only); LRRK2 knockout and G2019S knock-in mice; PINK1 or DJ-1 versions of the second-stress experiment; funders.