Two-steps control of cellular prion physiology by the Extracellular Regulated Kinase-1 (ERK1)

Abstract

Cellular prion (PrP^c) undergoes a regulated α-secretase-like cleavage by the disintegrin ADAM17 similar to the one taking place on β-Amyloid Precursor Protein (βAPP). Because these cleavages give rise to biologically active fragments, understanding their regulation could be of importance. We have established that the Extracellular Regulated Kinase-1 (ERK1) controls PrP^c processing by modulating ADAM17 phosphorylation in a protein kinase C-dependent manner. Strikingly, we also demonstrated that ERK1 acts upstream to increase PrP^c promoter transactivation in an AP-1 dependent manner. Therefore, ERK1 exerts a dual control of both PrP^c metabolism and expression. Interestingly, α-secretase cleavage of βAPP appears to be independent of ERK1. I describe here similarities and differences in α-secretase-mediated PrP^c and βAPP processing pathways and discuss putative physiopathological implications.

Keywords: :

• ADAM proteases
• AP-1
• Cellular prion (PrP^c)
• Extracellular regulated kinase-1 (ERK1)
• N1 fragment
• p53
• processing
• protein kinase C (PKC) isoforms
• transcription
• α-secretase

The cellular prion (PrP^C) is a fascinating protein that has been at the center of a biological revolution.1 Thus, this protein is naturally occurring in animals and human beings and converts into a pathogenic form, prion scrapie (PrP^Sc) to trigger fatal diseases.2 The “revolution” stands in the concept that this conversion of PrP^C into PrP^Sc occurs in a self-propagating manner without yet identified co-factors or nucleic acids.3 This puzzling observation aroused numerous works aimed at better understanding how PrP^C aggregates, converts, disseminates, how the diseases are propagated and transmitted, what are the organs and species barriers, etc.4

What about the putative physiological role(s) of PrP^C? Numerous functions have been proposed but not yet firmly identified. Among others, PrP^C could participate in lymphocyte activation, cell-cell adhesion and cellular signaling (reviewed in ref. 5). More discussed were the observations concerning PrP^C function in the control of cell death. Thus, PrP^C could display both protective and trophic phenotypes6 while alternatively, PrP^C could also potentiate cell death and caspase-3 activation in several cell systems challenged with apoptotic stimuli.7–9 Some of the discrepancies could stand in the fact that PrP^C is physiologically processed by an α-secretase-like cleavage that yields both N- and C-terminal fragments that display distinct phenotypes. Thus, the α-secretase-derived N-terminal fragment, referred to as N1, exhibits a protective function related to its ability to control the proapoptotic tumor suppressor p5310 while the C-terminal counterpart C1 is clearly toxic and triggers p53-dependent cell death.11 Therefore, the survival/toxic functions ascribed to PrP^C could be linked to the balance between the levels of the two fragments generated upon proteolytic maturation and therefore, could be directly dependent upon the enzymatic machinery associated to the biological system in which PrP^C function is examined. The scheme is even complicated by the fact that the two fragments could functionally interact. Thus, N1 is able to protect against both PrP^C-and C1-associated staurosporine-stimulated caspase-3 activation. It is therefore of importance to understand the PrP^C physiopathological maturation process.

Our very first data on PrP^C processing was the result of a bet. Previous literature indicated that PrP^C underwent in vivo cleavage within its putative domain (sequence 106–126).12 Working on Alzheimer disease and more particularly on the physiopathological maturation of the β-amyloid precursor protein (βAPP),13–15 it was striking that the membrane cleavage of PrP^C was clearly reminiscent of the one taking place in the middle of the putative toxic domain of βAPP (i.e., the Aβ sequence). Therefore, in both cases, it appeared that this cleavage could be seen as an inactivating process. The bet was when we questioned whether PrP^C and βAPP could undergo such a cleavage by similar processes and activities. Thus, βAPP was reported to be proteolyzed within the Aβ domain by a series of proteases belonging to the disintegrins family, namely ADAM10 and ADAM17.16,17 However, there was no example of proteins linked to the membranes by a glycosylphosphatidyl inositol group (as is PrP^C) that could behave as a disintegrin substrate.18 Indeed, the bet was successful since we established that both PrP^C and βAPP undergo constitutive and regulated α-secretase cleavages by ADAM10 and ADAM17, respectively.19,20

We recently established that the Extracellular Regulated Kinase-1 (ERK1) controls the regulated ADAM17-mediated cleavage of PrP^C21 but also acts upstream and contributes to the regulation of the PrP^C promoter transactivation and mRNA levels.21 Thus, N1 production could be altered by both genetic or pharmacological manipulations of ERK1 or its upstream kinase MEK as well as by means of constitutively active or dominant negative ERK-/MEK-liked constructs.21 We showed that this ERK-associated control of PrP^C-mediated α-secretase cleavage was due to ERK1 ability to phosphorylate, and thereby to increase, ADAM17 enzymatic activity.21 Both protein kinase C-stimulated and muscarinic receptors-dependent α-secretase processing of PrP^C were fully dependent of ERK1 pathway.21

Interestingly, we also observed an ERK1-dependent modulation of endogenous PrP^C expression that we ascribed to ERK1-associated increase in PrP^C promoter transactivation.21 In agreement with previous works showing that ERK1 triggers the activation of the AP-1 transcription factor,22 we delineated an AP-1 consensus site on the PrP^C promoter region, the mutation of which fully abolished ERK1-mediated PrP^C promoter transactivation.21 The above set of data indicated that ERK1 could modulate PrP^C physiology at two distinct steps.

How could we rule out the possibility that ERK1-mediated control of N1 production could merely reflect the ability of the kinase to modulate PrP^C transcription/expression and thus ADAM17 substrate? Three lines of data indicated that EK1-mediated control of PrP^C expression and processing were independent events. First, ERK1 modulates ADAM17 activity as indicated by fluorimetric mesurements of the membrane embedded disintegrin activity on intact cells. Second, the expression of constitutively active MEK1 still triggers an increase of PrP^C expression and promoter transactivation in fibroblasts devoid of ADAM17. Third, ERK-1 still modulates N1 in cells overexpressing a cmv-driven PrP^C cDNA construct, i.e., in absence of endogenous promoter and therefore putative ERK-1/(AP-1)-linked responsive elements. Altogether, the above experiments undoubtedly support the conclusion that the ERK1 pathway independently controls both ADAM17-dependent N1 secretion and PrP^C promoter transactivation and expression (Fig. 1).

It is interesting to note that although βAPP and PrP^C share common proteolytic denominators,23 there exist also a few differences that could be delineated. Thus, while the protein kinase C (PKC) isoforms α, δ and ε contributed to the regulated α-secretase cleavage of PrP^C,24 PKCδ did not participate to the PKC-stimulated βAPP processing.25 Furthermore, unlike for PrP^C, βAPP regulated α-secretase cleavage remained independent of ERK1 pathway.25 The above data indicate that PrP^C and βAPP processing could be modulated/activated by pharmacological approaches aimed at selectively potentiating the α-secretase cleavage of a given protein without directly interfering with the processing of the other.

Figures and Tables

Figure 1 Regulated α-secretase pathways of cellular prion and βAPP processing. M1 and M3 muscarinic receptors can be stimulated by acetylcholine and blocked by atropine. In the case of PrP^C, subsequent to this agonist-receptor interaction, PKC stimulation triggers ERK1 phosphorylation and Thr735 phosphorylation of ADAM17, thereby increasing N1 production. Concomitantly, ERK1 increases PrP^C promoter transactivation via AP-1 and thereby enhances PrP^C expression. In the case of βAPP, PKC triggers ADAM17 phosphorylation by a yet unidentified intermediate effector that is distinct of ERK1. Both α-secretase cleavages yield fragments, N1 and sAPPα with similar protective properties.

Acknowledgments

I wish to thank Franck Aguila for artwork. This work was supported by the Fondation pour la Recherche Médicale and the Conseil Général des Alpes Maritimes.