Gain in toxic function of stefin B EPM1 mutants aggregates: correlation between cell death, aggregate number/size and oxidative stress.

EPM1 is a rare progressive myoclonus epilepsy accompanied by apoptosis in the cerebellum of patients. Mutations in the gene of stefin B (cystatin B) are responsible for the primary defect underlying EPM1. Taking stefin B aggregates as a model we asked what comes first, protein aggregation or oxidative stress, and how these two processes correlate with cell death. We studied the aggregation in cells of the stefin B wild type, G4R mutant, and R68X fragment before (Ceru et al., 2010, Biol. Cell). The present study was performed on two more missense mutants of human stefin B, G50E and Q71P, and they similarly showed numerous aggregates upon overexpression. Mutant- and oligomer-dependent increase in oxidative stress and cell death in cells bearing aggregates was shown. On the other hand, there was no correlation between the size and number of the aggregates and cell death. We suggest that differences in toxicity of the aggregates depend on whether they are in oligomeric/protofibrillar or fibrillar form. This in turn likely depends on the mutant's 3D structure where unfolded proteins show lower toxicity. Imaging by transmission electron microscopy showed that the aggregates in cells are of different types: bigger perinuclear, surrounded by membranes and sometimes showing vesicle-like invaginations, or smaller, punctual and dispersed throughout the cytoplasm. All EPM1 mutants studied were inactive as cysteine proteases inhibitors and in this way contribute to loss of stefin B functions. Relevance to EPM1 disease by gain in toxic function is discussed.

Intracellular aggregation of human stefin B: confocal and electron microscopy study.

BACKGROUND: Protein aggregation is a major contributor to the pathogenic mechanisms of human neurodegenerative diseases. Mutations in the CSTB (cystatin B) gene [StB (stefin B)] cause EPM1 (progressive myoclonus epilepsy of type 1), an epilepsy syndrome with features of neurodegeneration and increased oxidative stress. Oligomerization and aggregation of StB in mammalian cells have recently been reported. It has also been observed that StB is overexpressed after seizures and in certain neurodegenerative conditions, which could potentially lead to its aggregation. Human StB proved to be a good model system to study amyloid fibril formation in vitro and, as we show here, to study protein aggregation in cells. RESULTS: Endogenous human StB formed smaller, occasional cytoplasmic aggregates and chemical inhibition of the UPS (ubiquitin-proteasome system) led to an increase in the amount of the endogenous protein and also increased its aggregation. Further, we characterized both the untagged and T-Sapphire-tagged StB on overexpression in mammalian cells. Compared with wild-type StB, the EPM1 missense mutant (G4R), the aggregate-prone EPM1 mutant (R68X) and the Y31 StB variant (both tagged and untagged) formed larger cytosolic and often perinuclear aggregates accompanied by cytoskeletal reorganization. Non-homogeneous morphology of these large aggregates was revealed using TEM (transmission electron microscopy) with StB detected by immunogold labelling. StB-positive cytoplasmic aggregates were partially co-localized with ubiquitin, proteasome subunits S20 and S26 and components of microfilament and microtubular cytoskeleton using confocal microscopy. StB aggregates also co-localized with LC3 and the protein adaptor p62, markers of autophagy. Flow cytometry showed that protein aggregation was associated with reduced cell viability. CONCLUSIONS: We have shown that endogenous StB aggregates within cells, and that aggregation is increased upon protein overexpression or proteasome inhibition. From confocal and TEM analyses, we conclude that aggregates of StB show some of the molecular characteristics of aggresomes and may be eliminated from the cell by autophagy. Intracellular StB aggregation shows a negative correlation with cell survival.