Taking advantage of physiological proteolytic processing of the prion protein for a therapeutic perspective in prion and Alzheimer diseases

Abstract

Prion and Alzheimer diseases are fatal neurodegenerative diseases caused by misfolding and aggregation of the cellular prion protein (PrP^C) and the β-amyloid peptide, respectively. Soluble oligomeric species rather than large aggregates are now believed to be neurotoxic. PrP^C undergoes three proteolytic cleavages as part of its natural life cycle, α-cleavage, β-cleavage, and ectodomain shedding. Recent evidences demonstrate that the resulting secreted PrP^C molecules might represent natural inhibitors against soluble toxic species. In this mini-review, we summarize recent observations suggesting the potential benefit of using PrP^C-derived molecules as therapeutic agents in prion and Alzheimer diseases.

Keywords: :

• Alzheimer disease
• PrPN1
• neuroprotection
• prion
• prion disease
• shedding
• α-cleavage
• β-cleavage

Numerous neurodegenerative diseases including prion and Alzheimer diseases are caused by aberrant folding and aggregation of amyloid proteins into toxic intermediates. Blocking the formation or sequestrating these toxic intermediates may represent an interesting therapeutical avenue to reduce neurodegeneration occurring in these diseases.

Secreted Forms of PrP^C in Prion Diseases

Prion diseases are fatal transmissible neurodegenerative disorders caused by the misfolding and aggregation of the cellular prion protein (PrP^C) into a pathological isoform partially resistant to proteinase K termed PrP^{res 1}. Newly converted PrP^res can recruit other PrP^C molecules and cause their conversion into more PrP^res. The assembly of PrP^res molecules form infectious particles termed prions that are responsible for the transmission of prion diseases. The plasma membrane and the endosomal recycling compartment are the primary site for prion conversion.^2,3 Prion conversion is central to prion neurotoxicity.⁴ In humans, most prion diseases are sporadic and occur in the absence of mutation in the cellular prion protein (PrP^C) gene, PRNP, and in the absence of any contact with infectious prions. In familial cases, inherited dominant mutations in the PRNP gene cause the production of abnormal PrP^C molecules. Iatrogenic and acquired cases constitute a third category of prion diseases and result from accidental exposures to prions

PrP^C is translated as a precursor, and the mature protein is a 209 amino acid GPI anchored polypeptide present at the cell surface. PrP^C is composed of an unstructured N-terminal domain and a structured C-terminal domain with three α-helices and two short β-sheets.⁵ Both domain are essential for efficient conversion of PrP^C into PrP^{res 6,7}. PrP^C has a dual function in prion diseases. PrP^C is the substrate for prion conversion and its neuronal expression is required for the pathogenesis of prion diseases.^4,8,9 In addition, PrP^C is a cell surface receptor mediating neurotoxic signals for β-sheet soluble oligomers produced during prion conversion.^10,11 In contrast to its role in prion diseases, the physiological function of PrP^C is still unclear.

En route to the cell surface, a fraction of PrP^C undergoes an endoproteolytic cleavage, termed α-cleavage, between amino acid 110–111 and 112. This cleavage causes the release of PrPN1, a N-terminal moiety of 88 amino acids. The C-terminal counterpart, PrPC1, remains at the cell surface.^12,13 In the brain, up to 50% of PrP^C molecules undergo the α-cleavage and this physiological cleavage may be considered a major post-translational modification of this protein.¹³ The identity of the protease responsible for the α-cleavage is still unknown.¹⁴ A second physiological cleavage also termed shedding is performed by the metalloprotease ADAM10 at the cell surface and leads to the release of a full-length anchorless PrP^C isoform (shed PrP).^15-17 PrP^C shedding contributes to a significant proportion of secreted PrP^C molecules in cell culture and shed PrP^C was also detected in vivo.^16-19 A third cleavage between amino acid 89 and 90 also termed β-cleavage is activated by ROS and may be involved in the cellular response to oxidative stress.^12,20 This modification causes the release of PrPN2, and leaves the C-terminal PrPC2 moiety at the cell surface.

PrP^C proteolytic cleavages may have important functions.^14,20,21 In particular, they may be considered as endogenous mechanisms of defense against prion disease for several reasons (Fig. 1). First, they reduce the expression of GPI-anchored PrP^C which is essential for efficient prion conversion in cell cultures, the development of prion diseases in mice models, and central nervous system invasion after peripheral prion infection.^4,8,9,22-24 GPI-anchored PrP^C is also a cell surface receptor for prion soluble oligomers and relay intracellular toxic signals.^10,11 Releasing PrP^C from the cell surface is expected to interfere with these neurotoxic signals. Second, constitutive secretion of PrP^C using a genetic fusion between PrP^C and immunoglobulin Fc gamma interferes with the propagation of PrP^res in experimentally-infected mice and delays onset of disease after experimental inoculation with prions.²⁵ The mechanism involves the capture and neutralization of PrP^res. In agreement with this result, recombinant PrP reduces prion conversion in vitro by binding to PrP^res and preventing PrP^res/PrP^C interactions.⁶ Third, PrPC1 produced after PrP^C α-cleavage is not a substrate for prion conversion. In contrast, PrPC1 acts as a dominant-negative inhibitor of PrP^res mediated-conversion suggesting that an incomplete PrP^C molecule at the cell surface can trap PrP^res into a non-convertible complex.²⁶ Fourth, the observation that two disease-associated PrP^C mutants do not undergo ROS-mediated β-cleavage suggests that PrPN2 and/or PrPC2 have an important role against oxidative stress in prion diseases.²⁰ In agreement with the protective role of PrP^C α-cleavage, a decrease in PrPC1 levels has been observed in animal models of prion diseases and in a human prion disease.^13,26,27 This decrease is mediated by an increased activity of 3-phosphoinositide–dependent kinase-1 (PDK1) in prion-infected neurons, and siRNA-mediated silencing of PDK1 delays mortality in prion-infected mice.²⁸ Also, the annual incidence of prion diseases is approximately 1 in 1 000 000 persons.²⁹ Thus, these disorders are among the most rare neurodegeneratives diseases, and physiological proteolytic cleavages of PrP^C may in part explain this very low incidence rate.

Figure 1. PrP^C α-cleavage display hypothetical neuroprotective functions in prion and Alzheimer diseases. In prion diseases (left), endogenously produced PrPN1 (green) traps PrP^res toxic species, inhibits their conversion and thus, reduces their toxicity. In addition, PrPC1 is a dominant-negative inhibitor of PrP^res formation. In Alzheimer disease (right), PrP^C α-cleavage is increased. PrPN1 molecules trap Aβ oligomers (Aβ_o) and reduce Aβ-mediated toxicity by competition with Aβ_o receptors at the cell surface. Furthermore, the lack of Aβ binding sites on PrPC1 prevents PrPC1 from mediating Aβ_o toxicity. In a curative approach, therapeutic PrPN1 or PrPN1-derived drugs (purple) could be administered to supplement endogenous PrPN1 in both prion and Alzheimer diseases.

All these evidences point to two possible therapeutic avenues in prion diseases. As a first strategy, PrP^C-derived molecules may be used to capture PrP^res, prevent PrP^C/PrP^res interactions at the cell surface and reduce prion conversion. In addition, it is tempting to take advantage of the physiological cleavages of PrP^C to try and activate the involved proteases. We propose that PrP^C α-cleavage is the most promising therapeutic target because it removes the substrate for prion conversion. In contrast, too much PrP^C shedding may be neurotoxic.³⁰

Secreted Forms of PrP^C in Alzheimer Diseases

Alzheimer disease accounts for an estimated 60 to 80% of the current 35 million individuals living with dementia worldwide.³¹ An important area of research into the molecular mechanisms of Alzheimer disease is aiming at understanding how genes and their products modulate the pathogenesis of Alzheimer disease and finding therapeutic avenues.

Accumulation of amyloid β-peptide (Aβ) in extracellular plaques is a key feature of Alzheimer disease.³² Aβ is derived by sequential proteolysis of the β-amyloid precursor protein. Synaptic dysfunction occurs early in the disease and there is a clear correlation between synaptic failure and dementia, and Aβ oligomers levels in the brain.^33-35 PrP^C was identified as a high affinity receptor for Aβ oligomers in an unbiased approach.³⁶ Importantly, PrP^C/Aβ interactions were independently confirmed by different laboratories.^37-40 The role of PrP^C in Aβ-mediated toxicity in Alzheimer disease is currently a matter of contention. Some groups propose that PrP^C mediates synaptic failure by Aβ oligomers,^36,41-49 while other laboratories challenge the function of PrP^C in Aβ-mediated toxicity.^50-53

Both charged region of PrP^C (aa. 23–27 and 95–110) are essential for optimal interaction between PrP^C and Aβ.^38,39 Most importantly, these two regions are only present on PrPN1 and shed PrP. In several paradigms, these secreted PrP^C molecules show promising activity against Aβ oligomers (Fig. 1). First, PrPN1 and shed PrP produced recombinantly or derived from cells display neuroprotective functions against Aβ mediated-toxicity.^11,39,54-56 Second, both PrPN1 and to a lesser extent shed PrP, inhibit Aβ fibrillization, a critical step in the formation of toxic species of Aβ.^39,56,57 The absence of the second charged region on PrPN2 and the lower efficiency of shed PrP to inhibit Aβ fibrillization suggest that PrPN1 could represent the most efficient molecule interfering with Aβ fibrillization and toxicity. Third, PrPC1 does not bind Aβ oligomers which would also lead to a reduction of the neurodegeneration caused by Aβ. Again, it will be of great interest to use recombinant PrPN1 in vivo or to identify drugs capable of inducing PrP^C α-cleavage in order to stop or reduce the development of Alzheimer disease.

PrP^C N-terminal Domain/β-sheet Oligomers Interactions: Therapeutic Perspectives and Remaining Challenges

In recent years, unanticipated molecular cross talks between prion diseases and Alzheimer disease have emerged. PrP^C may be a cell surface receptor for toxic soluble β-sheet oligomers, including Aβ oligomers, and may have an important role in synaptic dysfunction in Alzheimer disease. In animal models affected by both prion and Alzheimer disease, the pathogenesis of both diseases is clearly accelerated, suggesting a molecular connection between these two disorders.⁵⁸ The molecular mechanisms underlying these cross talks are incompletely understood and are debated in the literature. However, binding of β-sheet oligomers to PrP^C N-terminal domain is undisputed, and the inhibition of Aβ oligomers toxicity by PrPN1 has been demonstrated in different laboratories.^11,39,54-56

Two forms of PrP^C N-terminal domain are naturally released by α- and β-cleavage, and there is some evidence that both α- and β-cleavage levels are increased in an animal model of Alzheimer disease.⁵⁹ We have shown that PrP^C α-cleavage is increased in post-mortem human Alzheimer disease brains.⁶⁴ Thus, increased secretion of PrP^C N-terminal domain may represent an endogenous compensatory mechanism aimed at interfering with Aβ-mediated toxicity.

In contrast to PrPN2, a neuroprotective activity of PrPN1 against β-sheet oligomers, including Aβ and PrP^res oligomers and other insults was identified in different experimental settings,^{11,39,54-56,60} and PrN1 is likely a promising start point molecule to develop therapeutic drugs. The protease responsible for PrP^C α-cleavage is unknown and modulating the release of PrPN1 in the short-term represents a difficult challenge. In contrast, PrPN1 or peptides derived from PrPN1 may represent a unique opportunity to develop therapeutic molecules against prion and Alzheimer diseases. Such molecules would act as molecular traps and prevent β-sheet oligomers from binding to the cell surface and induce neuronal degeneration. Although PrP^C N-terminal domain interacts with a broad range of ligands, unwanted side effects are unlikely because binding to PrP^C N-terminal domain induces a molecular mechanism only when the N-terminal domain is physically connected to the C-terminal domain.⁶¹

Several issues should be addressed before envisaging a specific therapeutic avenue. First, how does PrPN1 inhibit PrP^res conversion and Aβ fibrillization? A simple mechanism could involve the formation of PrPN1/PrP^res or PrPN1/Aβ complexes trapping PrP^res and Aβ molecules and preventing further conversion and toxic effects. In agreement for this mechanism, PrPN1 association with Aβ causes a conformational change resulting in the formation of amorphous and insoluble aggregates that are not compatible with the amyloidogenic pathway leading to the assembly of Aβ oligomers and fibrils.⁶⁴ Another issue relates to the fate of PrPN1/PrP^res or PrPN1/Aβ complexes. We have provided evidence for the presence of PrPN1/Aβ complexes in the guanidine-extractable fraction presumably representing insoluble amyloid plaques in post-mortem brain tissues of Alzheimer disease patients.⁶⁴ Interestingly, non-identified PrP^C molecules were found in Alzheimer disease extracellular plaques.⁶² These plaques were mainly recognized by antibodies targeting the N-terminal moiety of PrP^C, and it is tempting to speculate that PrP^C molecules found in these extracellular plaques of Alzheimer disease might be PrPN1. An important issue is the generation of a transgenic mouse constitutively expressing PrPN1. This mouse model is of paramount importance to test the therapeutic effect of PrPN1 in Alzheimer disease and prion diseases. However, PrPN1 is an unstructured polypeptide and targeting such polypeptide in the secretory pathway will be technically challenging.⁶³

In conclusion, recent data point to an emergent role of PrP^C-derived natural proteolytic products as a source for new therapeutic drugs in prion and Alzheimer diseases. In particular, there is clear evidence that the fragment resulting from PrP^C α-cleavage, PrPN1 has protective activity in vivo and has strong therapeutic potential. We predict that the development of a transgenic mouse expressing PrPN1 will be crucial importance to make a Go/No-Go decision in PrPN1-based drug development.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

This research was supported in part by a grant from the Fonds de Recherche en Santé, Québec and the Canadian Institutes for Health Research (MOP-89881) to X.R. We thank Julie Motard and Guillaume Tremblay for critical reading of the manuscript.

10.4161/pri.27438