Parkin/PINK1–STING paper retracted

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

Danielle Sliter, Richard Youle and colleagues at the US National Institutes of Health, with Christine Klein and Max Borsche (Lübeck) and others.

What happened

Mice lacking parkin or PINK1 (genes behind recessive Parkinson's; landmark 12, 15) show few Parkinson's features. The 2018 paper reported that, after exhaustive exercise, such mice mounted a strong inflammatory response, and so did parkin-lacking mice that also accumulate mitochondrial DNA mutations. Removing STING, a sensor that triggers interferon responses to DNA in the wrong place, reversed the inflammation and, in aged mutator mice, rescued dopamine neurons and movement. People with one or two faulty PRKN copies were also reported to have raised cytokines. Nature 561:258-262 (online 22 Aug 2018, issue of 13 Sep 2018).

On 30 July 2025 Nature published a retraction note. It says the data in Fig. 3a-c were discrepant from the complete raw data and so not reliable, and that the data in Fig. 1d were not reproducible. One author, Jonathon Burman, is deceased; the remaining authors agreed to the retraction.

Why it moved us toward a cure

Mainly as a caution. It claimed a bridge from mitochondrial quality control to inflammation, and so was a natural basis for ideas such as blocking STING. The idea that unremoved mitochondrial damage provokes immune activation is still plausible and is studied in other ways (landmark 55), but this paper can no longer be relied on to support it, and any programme built on its specific claims needs fresh evidence.

Outcome

Dead end for this paper. The hypothesis it addressed is untested rather than disproven.

Sources

• Sliter et al., Nature 2018;561:258-262, DOI 10.1038/s41586-018-0448-9 (PubMed 30135585, marked retracted; abstract)
• Retraction Note, Nature 2025;644:1116, 30 Jul 2025, DOI 10.1038/s41586-025-09346-8 (publisher page, read directly; PubMed 40739362)
• Not yet verified: which conclusions, if any, survive in other labs' independent work; whether any STING-targeting drug entered Parkinson's trials.