Abstract
Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is a cerebral small vessel disease that results from mutations in NOTCH3. How mutations in NOTCH3 ultimately result in disease is not clear, although there is a predilection for mutations to alter the number of cysteines of the gene product, supporting a model in which alterations of conserved disulfide bonds of NOTCH3 drives the disease process. We have found that recombinant proteins with CADASIL NOTCH3 EGF domains 1 to 3 fused to the C terminus of Fc are distinguished from wildtype proteins by slowed mobility in nonreducing gels. We use this gel mobility shift assay to define the effects of mutations in the first three EGF-like domains of NOTCH3 in 167 unique recombinant protein constructs. This assay permits a readout on NOTCH3 protein mobility that indicates that (1) any loss of cysteine mutation in the first three EGF motifs results in structural abnormalities; (2) for loss of cysteine mutants, the mutant amino acid residue plays a minimal role; (3) the majority of changes that result in a new cysteine are poorly tolerated; (4) at residue 75, only cysteine, proline, and glycine induce structural shifts; (5) specific second mutations in conserved cysteines suppress the impact of loss of cysteine CADASIL mutations. These studies support the importance of NOTCH3 cysteines and disulfide bonds in maintaining normal protein structure. Double mutant analysis suggests that suppression of protein abnormalities can be achieved through modification of cysteine reactivity, a potential therapeutic strategy.