#AI reads Urine#Multi-cohort, cross-species urinary proteomics reveals signatures of LRRK2 dysfunction in Parkinson’s disease

This study focuses on developing a non-invasive detection method for Parkinson's disease, particularly the type associated with LRRK2 gene mutations. By analyzing urine proteins from 1,215 individuals across three cohorts, it identified 177 proteins linked to LRRK2 mutations, involving pathways like lysosomal function and glycosphingolipid metabolism. A 30-protein panel screened via machine learning achieves high accuracy (average AUC of 0.91) in detecting LRRK2 G2019S mutations and is applicable to diverse populations. Verified in rat experiments, urinary protein changes in rats are similar to those in humans, and abnormal levels reverse with LRRK2-inhibiting drugs, enabling treatment effect monitoring. Additionally, 24 brain-derived proteins are detectable in urine, reflecting brain-related pathological changes, while other Parkinson’s-related gene mutations (e.g., GBA) cause opposite urinary protein alterations to aid in etiological differentiation. This research provides a non-invasive, easily promotable solution for early mutation detection, disease monitoring, and therapeutic efficacy evaluation of Parkinson's disease.

https://doi.org/10.1101/2025.11.06.686787 bioRxiv preprint

Youhe Gao

Statement: During the preparation of this work the author(s) used Doubao / AI reading for summarizing the content. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article.

For earlier AI Reads Urine articles：

https://www.keaipublishing.com/en/journals/advances-in-biomarker-sciences-and-technology/ai-reads-urine/