Research
Four questions about the inflamed brain
Our work sits where RNA biology meets neuroimmunology: understanding how glial cells go wrong in disease, and building molecular tools that can gently steer them back.
Theme 01
RNA editing & ADAR sensors
Human cells carry a natural RNA-editing enzyme called ADAR, which chemically converts the RNA letter A into I — a change the cell reads as if it were G. We build "RNA sensors" that use this native machinery to rewrite single letters inside RNAs from disease-relevant genes, without touching the genome.
In our recent preprint, we showed that this strategy can precisely target subpopulations of inflammatory C3-positive astrocytes in human iPSC-derived models. Because the approach edits RNA rather than DNA, the change is temporary and tunable — a safety profile we believe is right for diseases of the central nervous system.
Theme 02
Reactive astrocytes & complement signaling
Astrocytes are the brain’s most abundant support cells, and when the brain is injured or diseased they enter "reactive" states. Some of these states help, others amplify damage. Our Nature Communications work showed that reactive astrocytes transduce inflammation across a human blood-brain barrier model through a TNF–STAT3 signaling axis.
We are now dissecting the complement cascade — especially C3 — as both a marker and a driver of harmful astrocyte states, with the long-term goal of selectively silencing the harmful programs while keeping the helpful ones.
Theme 03
Human iPSC, organoid & barrier models
Much of the human brain is inaccessible to direct experiments, so we rebuild its parts in the lab. Using induced pluripotent stem cells we generate cortical organoids, oligodendroglia-producing fused organoids, and stem-cell-derived blood-brain barrier models that recapitulate key human biology.
Alongside the biology, we engineer the culture environment itself — biofunctionalized gelatin hydrogels and parylene-coated microdevices — so that these models mature faster and behave more like the tissue they stand in for.
Theme 04
CRISPR functional genomics
To find the genes that control astrocyte inflammation and barrier function, we run CRISPR-based screens. A CRISPRi screen published in Nature Neuroscience revealed regulators of distinct inflammatory reactive states in human iPSC-derived astrocytes.
Arrayed knockout screens extend this to specific questions — such as which genes control GLUT1 expression at the blood-brain barrier — turning the models above into discovery engines for drug targets.