Spatially informed clustering, integration, and deconvolution of spatial transcriptomics with GraphST.

Spatial transcriptomics technologies generate gene expression profiles with spatial context, requiring spatially informed analysis tools for three key tasks, spatial clustering, multisample integration, and cell-type deconvolution. We present GraphST, a graph self-supervised contrastive learning method that fully exploits spatial transcriptomics data to outperform existing methods. It combines graph neural networks with self-supervised contrastive learning to learn informative and discriminative spot representations by minimizing the embedding distance between spatially adjacent spots and vice versa. We demonstrated GraphST on multiple tissue types and technology platforms. GraphST achieved 10% higher clustering accuracy and better delineated fine-grained tissue structures in brain and embryo tissues. GraphST is also the only method that can jointly analyze multiple tissue slices in vertical or horizontal integration while correcting batch effects. Lastly, GraphST demonstrated superior cell-type deconvolution to capture spatial niches like lymph node germinal centers and exhausted tumor infiltrating T cells in breast tumor tissue.

The APP intracellular domain promotes LRRK2 expression to enable feed-forward neurodegenerative mechanisms in Parkinson's disease.

Gain-of-function mutations in the leucine-rich repeat kinase 2 (LRRK2) gene are common in familial forms of Parkinson's disease (PD), which is characterized by progressive neurodegeneration that impairs motor and cognitive function. We previously demonstrated that LRRK2-mediated phosphorylation of ß-amyloid precursor protein (APP) triggers the production and nuclear translocation of the APP intracellular domain (AICD). Here, we connected LRRK2 to AICD in a feed-forward cycle that enhanced LRRK2-mediated neurotoxicity. In cooperation with the transcription factor FOXO3a, AICD promoted LRRK2 expression, thus increasing the abundance of LRRK2 that promotes AICD activation. APP deficiency in LRRK2G2019S mice suppressed LRRK2 expression, LRRK2-mediated mitochondrial dysfunction, α-synuclein accumulation, and tyrosine hydroxylase (TH) loss in the brain, phenotypes associated with toxicity and loss of dopaminergic neurons in PD. Conversely, AICD overexpression increased LRRK2 expression and LRRK2-mediated neurotoxicity in LRRK2G2019S mice. In LRRK2G2019S mice or cultured dopaminergic neurons from LRRK2G2019S patients, treatment with itanapraced reduced LRRK2 expression and was neuroprotective. Itanapraced showed similar effects in a neurotoxin-induced PD mouse model, suggesting that inhibiting the AICD may also have therapeutic benefits in idiopathic PD. Our findings reveal a therapeutically targetable, feed-forward mechanism through which AICD promotes LRRK2-mediated neurotoxicity in PD.