The prion protein (PrP) is a cell-surface glycoprotein that has been linked to various neurodegenerative diseases. It seems that the cytoplasm provides an environment which promotes the conversion of PrP to PrPSc — a form of PrP that is associated with transmissible spongiform encephalopathies — but does PrP ever exist in the cytoplasm under normal circumstances? In the Proceedings of the National Academy of Sciences, Ma and Lindquist now report that PrP can accumulate in the cytoplasm, which has implications for the pathogenesis of prion-related diseases.

It is difficult to detect misfolded or mistargeted proteins in the cytoplasm because they are efficiently degraded by the proteasome, so Ma and Lindquist studied the effect of proteasome inhibitors on the cellular distribution of PrP. In untreated neuroblastoma cells, the authors found that PrP was localized mainly to the cell surface, but treatment with a proteasome inhibitor caused substantial amounts of PrP to accumulate internally.

Ma and Lindquist showed that antibodies against a cytoplasmic form of Hsp70 — Hsc70 — strongly colocalized with intracellular PrP aggregates in inhibitor-treated cells, but not in untreated cells, and concluded that PrP can accumulate in the cytoplasm when proteasome activity is compromised.

The authors found that the internally accumulated PrP was unglycosylated and lacked both the amino- and carboxy-terminal signal sequences, indicating that it had been fully processed in the endoplasmic reticulum and had arrived in the cytoplasm through retrograde transport.

Using various cell lines, protease inhibitors and transfection procedures, Ma and Lindquist showed that cytoplasmic accumulation of PrP is a general consequence of proteasome inhibition rather than the result of using a particular method. They also showed, by overexpressing PrP, that this accumulation is not the result of proteasome inhibition per se.

The authors proposed that if cytoplasmic PrP accumulation is involved in disease, it should occur more readily in cells expressing disease-related PrP mutants. They therefore created a point mutation in mouse PrP, which corresponds to one of the most common human PrP mutations associated with transmissible spongiform encephalopathies.

In the absence of proteasome inhibitors, the authors found that a substantial fraction of mutant, but not wild-type, PrP accumulated in the cytoplasm, and that, compared to wild-type PrP, a smaller fraction of the mutant protein had passed through the ER quality-control system. After proteasome inhibition, both wild-type and mutant PrP accumulated in the cytoplasm, although the mutant protein accumulated to higher levels.

These data indicate that PrP does appear in the cytoplasm under normal circumstances, a conclusion that was also reached by Yedidia and colleagues in a recent EMBO Journal paper. In unpublished experiments, Ma and Lindquist have shown that cytoplasmic PrP is selectively toxic to neuronal cells, and that a fraction of PrP can convert to PrPSc, depending on the rate of cytoplasmic accumulation. These results indicate that cytoplasmic accumulation of PrP might contribute to the pathogenesis of prion diseases.