David R Brown

1. PRION DISEASES

The names bovine spongiform encephalopathy (BSE), Creutzfeldt-Jakob disease (CJD), and scrapie are now household names despite the fact that the human disease (CJD) remains a disease of very low incidence, accounting for 1 in 106 deaths per annum worldwide. BSE and CJD are examples of prion diseases, fatal neurodegenerative conditions (1). Before the BSE epidemic, prion diseases were infamous because the cause could not be linked to a known pathogen such as a virus or bacterium but to a single protein termed the prion protein. Prusiner's protein-only hypothesis for the cause of prion diseases (2) has become quite famous in recent years and won him the Nobel Prize (3) despite the lack of decisive proof of the theory that the altered isoform is all that is need to induce prion diseases. There is continued support for the hypothesis that "transmissible spongiform encephalopathies" are not just the result of conformationally corrupted protein (PrPSc) passing from individual to individual. At the far end of this opposition are those who believe that there is a hidden virus that stimulates production of the abnormal prion protein but is the real culprit. At the other end are those who say that the "protein-only" cause is inadequate as tests show that recombinant prion protein forced to take on a similar conformation to that found in disease does not induce prion disease when injected into (4,5). The evidence that confounds the idea that PrPSc is not sufficient to cause disease is growing. It is now possible to isolate PrPSc from mouse brain that has much lower infectiv-ity than a standard PrPSc preparation (6). However, infectivity can be restored to this protein by the addition of heparin sulfate (6). Nevertheless, mice that cannot express the prion protein because of genetic ablation cannot be infected with the disease (7). Thus, only those that express the prion protein can get prion disease. Despite the possibility that a virus might emerge as the cause of prion disease, the diseases themselves are inseparable from the prion protein. Therefore, understanding what the prion protein does and how it is important to cellular metabolism is central to the whole enigma of the prion diseases.

2. THE PRION PROTEIN

The genetic code of the prion protein was identified only after the isolation of the abnormal isoform (PrPSc) from infected brains. Discovery of the gene led to the realization that there was a normal brain protein involved in the disease (1,3). However, what this protein actually does in the brain has

From: Handbook of Copper Pharmacology and Toxicology Edited by: E. J. Massaro © Humana Press Inc., Totowa, NJ

Fig. 1. The primary structure of the mouse prion protein. This protein is anchored to the cell membrane by a glycosylphospholipid (GPI) anchor. The signal peptide for entry into the endoplasmic reticulum and the GPI signal peptide are cleaved off before the protein reaches the cell surface. Glycosylation can occur at one two or none of the asparagine residues indicated. A hydrophobic region envelopes a cleavage point where the protein is cleaved during normal metabolic breakdown. A disulfide bond links two regions of the protein that form separate a-helices in the three-dimensional structure of the protein. The complete octarepeats can bind up to four copper atoms. Most mammals also have an incomplete repeat prior to this.

Fig. 1. The primary structure of the mouse prion protein. This protein is anchored to the cell membrane by a glycosylphospholipid (GPI) anchor. The signal peptide for entry into the endoplasmic reticulum and the GPI signal peptide are cleaved off before the protein reaches the cell surface. Glycosylation can occur at one two or none of the asparagine residues indicated. A hydrophobic region envelopes a cleavage point where the protein is cleaved during normal metabolic breakdown. A disulfide bond links two regions of the protein that form separate a-helices in the three-dimensional structure of the protein. The complete octarepeats can bind up to four copper atoms. Most mammals also have an incomplete repeat prior to this.

remained a mystery for the last 15 yr. The prion protein (PrPC) is a glycoprotein expressed (Fig. 1) on the surface of many cell types (8-12). The protein is linked to the cell membrane by a glycosal phosphatidyl insoitol anchor (13). It has one or two sugar chains linked close to the C-terminus but may also exist in a nonglycosilated form. PrPC is probably expressed by all vertebrates. Many mammalian and avian genes have been sequenced and, recently, the coding sequence for turtle prion protein has also been described (14,15). One region of the protein that encapsulates a normal metabolic splice site is so precisely conserved and is so unique among protein sequences that it must represent a functional domain of the protein essential to normal activity of the protein. Two separately derived "strains" of mice in which protein expression has been knocked out (Zrchl, Npu) were examined for gross disturbances in behavior and development (16,17). None were found, and on the basis of this, some experts suggested that the protein had a redundant function or no function at all. However, why would the sequence of the prion protein be so highly conserved from turtle to man? Possibly its function is so essential that, like many such proteins, normal metabolism has mechanisms to compensate for its loss.

Yet, even this picture has been blurred when other strains of prion-protein-deficient mice (Zrch2, Ngs, Rcm0) were found to develop late-onset motor disturbances and the loss of Purkinje cells in the cerebellum (18-20). A recent article has suggested that these other strains of PrPC-deficient mice become ill because another protein, termed doppel, with a small degree of homology (approx 25%) to the prion protein is highly expressed in these mice (19). This expression is possibly driven by the prion protein promoter running directly into the doppel reading frame, which is directly in tandem with that of the prion protein. Whatever the role of doppel in causing the phenotype of these PrPC-deficient mice, the late-onset pathology is abrogated by reintroducing prion protein expression (20). It is possible that renewed prion protein expression has a negative feedback effect on the prion protein promoter, inhibiting doppel expression. Regardless of this, the implication is that prion protein expression has a positive function in preventing disease. A close study of the doppel protein with NMR has shown stronger homology to PrPC at the secondary-structure level than the primary-

sequence homology would suggest (21). These nuclear magnetic resonance (NMR) results show a similar globular domain containing three helical domains and a small amount of p-sheet structure. However, unlike PrPC, doppel contains two disulfide bridges and is more heavily glycosilated (22). There is also evidence from my group's own work that doppel, unlike PrPC, is unable to bind Cu, which is not surprising because it lacks the octameric repeat region involved in Cu binding to PrPC.

In Rcm0 mice and other PrP knockout mice with late-onset neurodegeneration, the increased doppel expression is ectopic (19-22). In wild-type mice, doppel is expressed predominantly in other regions such as heart and testes. Cell death in mice overexpressing doppel in the brain is possibly related to increased production of nitric oxide. Rcm0 mice show increased levels of the enzymes that generate nitric oxide (iNOS and nNOS) and signs of oxidative and nitroxic damage to lipids (23). Despite these interesting investigations there is little evidence that doppel plays any role in prion disease. Studies from the lab of Aguzzi suggest that mice that lack doppel expression are just as sensitive to infection by scrapie and show similar degrees of neurodegeration. These results came from transplantation studies in which tissue from the brains of doppel knockout mice were implanted into the brains of PrP knockout mice (24). Thus, it remains to be determined what the normal function of the doppel protein is and whether doppel expression will affect disease progression in prion disease.

The initial suggestion that knocking out prion protein expression has no implication has also proven to be false for the two strains of PrPC-deficient mice that do not develop late-onset disease. At the level of the whole animal, there are behavioral differences related to changes in circadian rhythms (25). At the level of the nervous system, there are changes in electrophysiological parameters (26,27). PrPC-deficient mice are also more sensitive to kindling agents that cause fits (28). Although there is contradictory evidence from some investigators working with slice preparations at room temperature (29), there is evidence that parameters such as long-term potentiation and GABA-type inhibitory currents are abnormal in PrPC-deficient slices at physiological temperatures (26). Other parameters also differ, as do responses to stress-inducing agents such as exogenous copper and hydrogen peroxide (30,31). Down at the level of single cells, PrPC-deficient cells are less viable in culture than wildtype cells and are more susceptible to oxidative damage and toxicity from agents such as copper and cytosine arabinoside (32-36). Astrocytes show changes in ability to take up glutamate (37) and microglia are less responsive to activating substances (11). Therefore, at all levels, PrPC-deficient mice show a clear phenotype, indicating that they are more sensitive to various kinds of stresses implying that PrPC has an important function protecting cells from environment assaults. Furthermore, there is now evidence that prion protein expression increases when the brain is stressed by oxidative damage (38). Brains of patients with Alzheimer's disease show a 10-fold increase in the level of prion protein expression.

0 0

Post a comment