DJ-1 mutations cause early-onset parkinsonism

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

Vincenzo Bonifati, Patrizia Rizzu, Peter Heutink and colleagues, with 15 co-authors including Alexis Brice, Giovanni Meco, Cornelia van Duijn and Ben Oostra (18 authors in all), led from the Erasmus Medical Center, Rotterdam. Funders were not checked.

What happened

The team searched two families in which relatives had inherited the same DNA segment from both parents, which narrows the hunt. In a Dutch family they found a large (14 kb) deletion that removes DJ-1's expression; in an Italian family, a single-letter change (L166P) at a highly conserved position, probably disrupting the protein's structure. Neither was found in 720 chromosomes from the general populations (per Alzforum's summary of the paper). Science 299:256–259, 10 January 2003 (online 21 November 2002). Per Alzforum, wild-type DJ-1 sat in the cytoplasm and nucleus of transfected cells, while the L166P version concentrated in mitochondria and nucleus. A follow-up screen of 100 early-onset patients (Hedrich et al. 2004) found only two DJ-1 carriers, both with a single faulty copy, against 17 patients with parkin mutations; a reference work puts DJ-1 at about 1% of early-onset Parkinson's.

Why it moved us toward a cure

It added a third gene (after SNCA and parkin, see landmark 4 and landmark 12) in which losing a protective function, not gaining a toxic one, causes disease. The paper itself called the protein's function unknown but pointed to the oxidative stress response, where cells are damaged by reactive oxygen molecules; a 2004 study found that oxidised DJ-1 moves to mitochondria (the cell's power units) and protects cells (Canet-Avilés et al.). That sits beside the energy-failure evidence in landmark 11 and the mitochondrial quality-control pathway in landmark 15. Later work disputes its exact job: Richarme et al. (2015) proposed that it repairs sugar-damaged proteins ("deglycase"), but a 2023 study argues it is instead a detoxifying enzyme (a glyoxalase) for methylglyoxal, a reactive by-product of sugar breakdown, and not a deglycase.

Outcome

Open. The genetic link stands, but it is rare, the molecular function is contested, and no DJ-1-targeting therapy was found in this check (a ClinicalTrials.gov search was not run).

Sources

• Bonifati et al., Science 2003;299(5604):256–259, DOI 10.1126/science.1077209 (record page with authors and date)
• Alzforum commentary thread on the paper (mutations, controls, online date Nov 2002)
• Hedrich et al., Neurology 2004;62:389–394, DOI 10.1212/01.wnl.0000113022.51739.88, PMID 14872018
• Canet-Avilés et al., PNAS 2004;101:9103–9108, DOI 10.1073/pnas.0402959101, PMID 15181200 · Richarme et al., J Biol Chem 2015;290:1885–1897, DOI 10.1074/jbc.M114.597815, PMID 25416785
• Atlas of Genetics and Oncology, PARK7
• Coukos et al., Biochemistry 2023;62:3126–3133, DOI 10.1021/acs.biochem.3c00325, glyoxalase versus deglycase
• Bonifati et al., PMID 12446870 (abstract)
• Not yet verified: the paper's full text (the "720 chromosomes", deletion size and protein localisation come from Alzforum's summary of the paper); whether any DJ-1-targeting drug is in trials; the claim that DJ-1 loss makes neurons more vulnerable in humans.