If the number of diseases associated with protein aggregates is anything to go by, keeping proteins in their native state is not a simple task. One of the many molecules that have evolved to deal with the problem is an enzyme called protein disulphide isomerase (PDI). Ensconced in the endoplasmic reticulum (ER), PDI ensures that disulphide bonds in secretory proteins are properly cross-linked. Because disulphide bonds provide the only opportunity for protein side chains to covalently interact with each other, PDI performs a uniquely important task that raises an equally important question: Is PDI activity ever compromised, and what happens if it is?
In last Thursday’s Nature, Stuart Lipton at the Burnham Institute for Medical Research in La Jolla, California; Eliezer Masliah at the University of California, San Diego; Yasuyuki Nomura at Hokkaido University, Sapporo, Japan; and their colleagues report that inactive forms of PDI are present in brain samples taken from Alzheimer and Parkinson disease patients. More specifically, first author Takashi Uehara and coworkers have found that thiol groups at the active site of PDI are chemically modified by addition of a nitric oxide (NO) group. Because the authors also found that this modification both inactivates PDI and exacerbates accumulation of ubiquitinated proteins in cells, the findings hint that inactivation of the isomerase may compound the pathology of neurodegenerative diseases.
Using in vitro reactions, Uehara and colleagues first found that PDI could be inactivated by NO donors such as S-nitrosocysteine (SNOC). The isomerase becomes S-nitrosylated at any of four cysteine thiols present in two active site domains that lie at the N- and C-terminals of the protein. To find out if this modification has any physiological significance, the authors looked for S-nitrosylated PDI (SNO-P) in dopaminergic SH-SY5Y cells that had been treated with rotenone—this mitochondrial inhibitor, which induces a Parkinson disease-like pathology in animals, also leads to increases in NO. Finding that SNO-P does indeed form in the rotenone-treated cells, the authors then tested the human brain samples.
Of course, rotenone is not likely to be the cause of SNO-P in the human brain, but there are plenty of other factors that might increase NO production, one of them being excessive stimulation of N-methyl-D-aspartate (NMDA) receptors. When the researchers exposed primary cortical neurons to NMDA, they detected SNO-P, polyubiquitinated proteins, and signs of an activated unfolded protein response (UPR)—up-regulation of UPR proteins CHOP and XBP-1 (see ARF related news story). These events could all be prevented by overexpressing active PDI or treating the cells with NO blockers. The authors also found that wild-type, but not an isomerase-negative PDI, could attenuate Lewy body-like inclusions when synphilin was overexpressed in SH-SY5Y cells. And they found that this protection from synphilin aggregation was abolished by NO or SNOC.
All told, these findings suggest that PDI may be more than a mere housekeeping protein. It may also help protect cells from various forms of stress. In fact, to see just how versatile PDI might be, Uehara and colleagues overexpressed the isomerase in SH-SY5Y cells that were treated with either the ER toxin thapsigargin or the proteasome inhibitor MG132. In both cases the isomerase reduced the number of cells that underwent apoptosis by about half.
“Our data demonstrate a previously unrecognized relationship between NO and protein misfolding in degenerative disorders, showing that PDI can be a target of NO after mitochondrial insult in cellular models of PD and in human neurodegenerative diseases,” write the authors. The data also suggest yet another way that NMDA receptors, which have been linked to calcium toxicity, can contribute to cell death.—Tom Fagan