GCH1 variants raise risk

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

Niccolo Mencacci, Ioannis Isaias, Martin Reich, Nicholas Wood and colleagues, with the International Parkinson's Disease Genomics Consortium and the UCL-exomes consortium (Brain 137:2480–2492, September 2014, DOI 10.1093/brain/awu179, PMID 24993959). Funders included the Wellcome Trust/MRC, Parkinson's UK, the Michael J. Fox Foundation, NIH intramural programmes and others.

What happened

GCH1 makes GTP cyclohydrolase 1, an enzyme needed to produce dopamine. Severe loss of it causes DOPA-responsive dystonia, a childhood movement disorder that responds to levodopa.

The team first found four families with that disease in which relatives had adult-onset parkinsonism. Dopamine transporter scans were abnormal in the parkinsonian patients scanned, so they showed nigral dopamine-terminal loss like Parkinson's. They then searched exome data (the protein-coding DNA) from 1,318 Parkinson's cases and 5,935 controls. GCH1 variants were found in 10 of 1,318 cases (0.75%) against 6 of 5,935 controls (0.1%): odds ratio 7.5 (95% confidence interval 2.4 to 25.3). Eleven different variants were found in 16 people; four were known to cause dopa-responsive dystonia and seven were of undetermined significance.

Why it moved us toward a cure

It linked a dopamine-synthesis gene to loss of dopamine neurons, which suggests that long-standing low dopamine may itself make nigral cells vulnerable. A common variant at the GCH1 locus was also one of the six new loci in the 2014 pooled scan (landmark 237); the paper itself cited this only as an unpublished personal communication. The paper did not identify a drug target.

Outcome

Open. The exome cohort was enriched for early-onset and familial cases, had no independent replication in the paper, and could not test copy-number changes. No dopamine scans were available for the exome cases, so some could have been misdiagnosed dopa-responsive dystonia; dropping the one case the authors flagged left the result (P = 0.0003). The mechanism of cell loss is uncertain. The paper drew a Brain commentary (Weissbach and Klein) and five letters with the authors' reply in 2015 (contents not read). A web search found no independent replication.

Sources

• Mencacci et al., Brain 2014, DOI 10.1093/brain/awu179, PMC4132650
• Nalls et al., Nat Genet 2014, DOI 10.1038/ng.3043 (lists GCH1 among six novel loci, PMC4146673)
• Weissbach & Klein, Brain 2014;137:2402–4 (commentary)
• Not yet verified: later replication of the rare-variant association; the GCH1 signal in the 2019 meta-analysis; the content of the 2015 letters. No retraction or correction was found (Crossref, Europe PMC); PubPeer could not be reached.