Purified CADASIL mutant NOTCH3 receptors could not assemble into large clumps by themselves, and did so only when mixed with the normal, wild-type receptor, according to a laboratory study published in The FASEB Journal on 31 March 2026 by a group at Leiden University Medical Center working with colleagues at Xi'an Jiaotong University. The team, which included the Leiden CADASIL clinical genetics group of Saskia Lesnik Oberstein and Julie Rutten alongside the cell biology laboratory of David Baker, made full-length NOTCH3 extracellular domains in human producer cells and measured their size by light scattering and by gel electrophoresis under non-denaturing conditions. The wild-type protein behaved as a two-part dimer. Two variants sitting in the fourth EGF-like repeat, C162W and C183R, were monomers on their own; a third variant, C1015R, looked like wild-type. When mutant and wild-type proteins were combined at equal amounts, high molecular mass assemblies appeared.
CADASIL, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy, is the most common inherited cause of small vessel stroke and of early vascular cognitive decline, and it is caused by cysteine-altering changes in the NOTCH3 gene that leave the receptor protein accumulating in the walls of small brain arteries. Because the condition is dominant, cells carry one altered and one normal copy of the gene, so both forms of the receptor are present. In the same paper, the two EGF-like repeat 4 variants significantly increased signalling driven by the NOTCH ligand JAGGED1 in stably transfected 293T cells, while signalling driven by a different ligand, DLL1, was not increased. Previously described peptides that target the receptor dimerisation region, used at 10 micromolar, blocked both the aggregation and the excess signalling; those peptides were first reported by the same laboratory in The FEBS Journal.
The result is consistent with earlier mouse work from a Paris group, published in the Journal of Clinical Investigation in 2024, which found more NOTCH3 aggregation and more arterial damage in mice carrying both normal and mutant Notch3 than in mice carrying the mutant alone. The Leiden authors present the interaction between the two forms of the receptor, rather than the mutant protein in isolation, as the point a drug might act on.
The counterpoint is the scale of what has been shown. This is bench chemistry and cultured cell work, in cells engineered to make large amounts of the receptor, with three of the more than two hundred reported CADASIL variants. One of those three, C1015R, did not fit the pattern, which the authors themselves note as a reason to test a wider range of mutations. No animal experiment, no measurement of how the peptides behave in a living body, and no dose has been reported.