Prion protein aggresomes are poly(A)+ribonucleoprotein complexes that
induce a PKR-mediated deficient cell stress response
Kevin Goggin1, Simon Beaudoin1, Catherine Grenier, Andrée-Anne Brown, Xavier Roucou⁎
Department of Biochemistry, Faculty of Medicine, University of Sherbrooke, 3001 12èmeAvenue Nord, Sherbrooke, Québec, Canada J1H 5N4
Received 26 July 2007; received in revised form 16 October 2007; accepted 16 October 2007
Available online 20 November 2007
In mammalian cells, cytoplasmic protein aggregates generally coalesce to form aggresomal particles. Recent studies indicate that prion-infected
cells produce prion protein (PrP) aggresomes, and that such aggregates may be present in the brain of infected mice. The molecular activity of PrP
aggresomes has not been fully investigated. We report that PrP aggresomes initiate a cell stress response by activating the RNA-dependent protein
kinase (PKR). Activated PKR phosphorylates the translation initiation factor eIF2α, resulting in protein synthesis shut-off. However, other
components of the stress response, including the assembly of poly(A)+RNA-containing stress granules and the synthesis of heat shock protein 70,
are repressed. In situ hybridization experiments and affinity chromatography on oligo(dT)-cellulose showed that PrP aggresomes bind poly(A)+
RNA, and are therefore poly(A)+ribonucleoprotein complexes. These findings support a model in which PrP aggresomes send neuronal cells into
untimely demise by modifying the cell stress response, and by inducing the aggregation of poly(A)+RNA.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Prion protein; Aggresome; Stress response; PKR; Stress granule
Misfolded proteins are usually directed through chaperone-
mediated refolding pathways, or are degraded by the protea-
some. Cells have also evolved a structure termed aggresome,
where aggregated proteins may be sequestered. Aggresomes are
juxtanuclear cytoplasmic inclusion bodies that form around the
microtubule-organizing center or centrosome by way of dynein-
directed retrograde transport of proteins on microtubule tracks
[1,2]. Aggregated forms of proteins, including mutant forms of
cystic fibrosis transmembrane conductance regulator, super-
oxide dismutase, rhodopsin, Tcell receptor α, and presenilin-1,
have been shown to localize to aggresomes . Aggresomes
share both morphological and biochemical similarities with
inclusion bodies that characterize common neurodegenerative
diseases, including amyotrophic lateral sclerosis, Parkinson's
disease, and Alzheimer's disease. In addition to the aggregated
proteins, aggresomes contain ubiquitin, proteasomes, and heat
shock proteins .
Prion diseases are rare fatal neurodegenerative disorders,
which include Creutzfeldt–Jakob disease in humans, bovine
abnormal prion protein conformers (PrPSc) derived from normal
cellular host prion protein (PrPC) . The cause of neurode-
generation in these disorders is not well understood, and there is
much evidence that argues against the direct neurotoxicity of
PrPSc[5–7]. Some attention has recently turned toward explo-
ring mistrafficking and accumulation of PrP in the cytoplasm.
Inappropriate expression of PrP in the cytosol of cells led to the
formation of neurotoxic aggregates insoluble in non-ionic deter-
gents and partially resistant to proteinase K [8,9]. We recently
determined the molecular morphology of these aggregates and
PrP appears in the cytosol may involve retrotranslocation of
Available online at www.sciencedirect.com
Biochimica et Biophysica Acta 1783 (2008) 479–491
Abbreviations: CyPrPEGFP, cytoplasmic prion protein genetically fused to
EGFP; EGFP, enhanced green fluorescent protein; ER, endoplasmic reticulum;
Hsp, heat shock protein; poly(I:C), polyinosinic–polycytidylic acid; PKR,
RNA-dependent protein kinase; PrP, prion protein; SGs, stress granules
⁎Corresponding author. Tel.: +1 819 346 1110x12248; fax: +1 819 564 5340.
E-mail address: firstname.lastname@example.org (X. Roucou).
1These authors have contributed equally to this work.
0167-4889/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
misfolded PrP from the ER for degradation by proteasomes ,
or inefficient signal-peptide-mediated translocation into the
endoplasmic reticulum (ER) [11,12]. ER stress also prevents
nascent PrP molecules from being translocated into the ER
[13,14]. Interestingly, prion-infected N2aPD88 and GT-1 cells
displayed cytoplasmic aggresomes upon mild impairment of the
proteasome . Furthermore, evidence of aggresomal struc-
tures was also found in the brain of prion-infected mice .
Thus, prions may facilitate mistrafficking of PrP in the cytosol.
Clearly, any role of PrP aggresomes in prion diseases cannot
be delineated without further investigating their impact on cell
physiology. We have investigated if PrP aggresomes induce a
cellular stress by determining if cells producing aggresomes
exhibit a spontaneous stress response. The cell stress response is
characterized by three mechanisms. First, one of four kinases
(RNA-dependent protein kinase PKR, PKR-like endoplasmic
reticulum kinase PERK, heme-regulated inhibitor HRI, or ge-
neral control non-derepressible-2 GCN2) is activated and phos-
phorylates the translation initiation factor eIF2α at Ser51 .
This phosphorylation converts eIF2α from a substrate to a
factor responsible for regenerating eIF2–GTP. This results in a
limited availability of the ternary complex eIF2–GTPtRNA–
Met for the assembly of the 43S pre-initiation complex, and thus
a reduced rate of translation initiation . Second, large cyto-
plasmic aggregates of poly(A)+RNA termed stress granules
(SGs) containing stalled translation initiation complexes are
assembled in the cytoplasm . The sequestration of these
components may help cells to recover post-stress by replenish-
ing the cellular pool of mRNA without the need for new
transcription. Third, cells induce the synthesis of heat shock
proteins (Hsps) that participate in protein refolding, elimination
of misfolded proteins, and inhibition of apoptosis [19,20].
Here, we show that PrP aggresomes induce a cellular stress
response characterized by the activation of PKR, the phosphor-
ylation of eIF2α and a remarkable reduction in protein syn-
thesis. However, aggresomes inhibit the assembly of SGs
and the synthesis of Hsp70. This inhibition results from the co-
aggregation of poly(A)+RNA with PrP aggregates. These fin-
dings indicate for the first time that PrP aggresomes induce a
deficient cell stress response and display mRNA aggregation
2. Materials and methods
2.1. Antibodies, clones, and reagents
Primary antibodies used were monoclonal anti-EGFP (clone B-2, SantaCruz
Biotechnology), polyclonal anti-eIF2α (FL-135, SantaCruz Biotechnology),
polyclonal anti-phospho-eIF2α (Cell Signaling), anti-eIF4E (clone P-2,
SantaCruz, Biotechnology), anti-Hsp70 (clone C92F3A-5, Stressgen), anti-
HuR (clone 3A2, SantaCruz Biotechnology), anti-PKR (clone B-10, SantaCruz
Biotechnology), anti-phospho PKR Thr-446 (clone E120, Abcam), anti-prion
protein (clone 3F4, Chemicon), anti-TIAR (clone 6, Pharmingen). Secondary
antibodies were alexa Fluor 633 and 568 F(ab′)2 fragment of goat anti-mouse
IgG (Molecular Probes), peroxidase-linked anti-mouse and anti-rabbit IgG from
sheep (Amersham Biosciences).
Cloning of EGFP and CyPrPEGFPin pCEP4β (Invitrogen) was described
previously [10,21]. The CyPrP sequence is from human origin. Translation
initiation factor eIF2α was amplified from human cDNAs kindly provided by
Dr Jana Stankova (University of Sherbrooke, QC Canada). The PCR product
was introduced in the HindIII and BamHI sites of pCEP4β. Mutation S51D
was created by the QuickChange procedure using primers eIF2α-forward
5′-gattcttcttagtgaattagacagaaggcgtatccgttc-3′ and eIF2α-reverse 5′-gaacggatacg-
ccttctgtctaattcactaagaagaatc-3′. The mutant construct was sequenced and its
expression verified by Western blotting. The construct encoding GFP-250 was
kindlyprovidedbyDrElisabethSztul(University ofAlabama atBirmingham,AL
USA). All reagents were obtained from Sigma-Aldrich, unless otherwise stated.
PKR inhibitor and its inactive analog were purchased from Calbiochem.
2.2. Cell culture, transfections, and treatment
Mouse N2a neuroblastoma, human embryonic kidney 293T, human cervical
Hela, and human neuroblastoma SK-N-SH were maintained in Dulbecco's
modified Eagle's medium plus 10% fetal bovine serum (Wisent). Human
neuroblastoma BE(2)M17 cells were maintained in Optimem plus 10% fetal
bovine serum (Invitrogen). Transfections were carried out using lipofectamine
according to the manufacturer's protocol (Invitrogen).
Cells were treated with different stress treatments as follows: 0.5 mM
sodium arsenite for 30 min; heat shock, 42 °C for 1 h, 1 μM thapsigargin, 2 mM
dithiotreitol, for 1 h; 1 μg/ml polyinosinic–polycytidylic acid [poly(I:C)] with
lipofectamine for 1 h; Dulbecco's modified Eagle's medium without leucine in
the presence of 1 μM MG132. For induction of Hsp70, cells were allowed to
recover at 37 °C for the indicated period of time.
2.3. Apoptosis assays
Cells were transiently transfected with empty vector, EGFP or CyPrPEGFP.
After 24 h, cells were incubated in the presence of sodium arsenite at the
indicated concentrations for 30 min. Arsenite was removed and cells were
returned at 37 °C for 8 h. Cells were then fixed and processed for nuclei staining
as previously described . Apoptosis was measured by counting cells
displaying apoptotic nuclei versus total number of cells.
2.4. Immunofluorescence, in situ fluorescence staining and
Cells grown on coverslips were fixed and processed for immunofluores-
cence as previously described . Primary antibodies dilutions were as
follows: eIF2α (1/100), phospho-eIF2α (1/100), eIF4E (1/100), HuR (1/1500),
Hsp70(1/50), G3BP (1/1000). Secondary antibodies were diluted 1/1000. For in
situ staining, permeabilized cells were incubated for 10 min with 2× SSC, and
hybridized with 1 nM of an end-labelled biotinylated oligo-dT (50 nucleotide,
IDT) overnight at 40 °C. In control experiments, permeabilized cells were
incubatedfor10 min with3 NNaOH, and washedthreetimesfor 5 minwithPBS.
After washingtwice with2×SSC and once with 0.5× SSC,cells were equilibrated
in 1× PBS containing 1 mg/ml BSA. Cells were incubated with 2 μg/ml Alexa
Fluor 633-labelled strepavidin (Molecular Probes) in 1× PBS containing 1 mg/ml
BSA. After 1 h incubation, cells were washed and mounted as previously
visible/epifluorescence inverted microscope (Nikon Corporation, Japan)
equipped with band pass filters for fluorescence of Hoechst (Ex. D340/40: Em.
D420), GFP (Ex. D450/40: Em. D500/50) and tetramethylrhodamine isothio-
cyanate (TRITC) (Ex. D528/25: Em. D590/60) (Nikon Corporation). Photo-
micrographs of 1344×1024 pixels were captured using either 60× or 100× oil
immersion objectives and Orca cooled color digital camera (Hamamatsu Pho-
tonics, Japan). Images were processed using NIS Elements AR software (Nikon
Corporation). Within the same figure, all pictures were taken with the same
2.5. Confocal and image analysis
Cells were examined with a scanning confocal microscope (FV1000,
Olympus, Tokyo, Japan) coupled to an inverted microscope with a 63× oil
immersion objective (Olympus). Specimens were laser-excited at 488 nm
480K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
between the emitted EGFP and Alexa Fluor 633 fluorescences were collected
sequentially at wavelengths 525–550 and N590 nm respectively. Serial
horizontal optical sections of 512×512 pixels with 2 times line averaging were
taken at 0.11 μm intervals through the entire thickness of the cell (optical
resolution: lateral — 0.18 μm; axial — 0.25 μm). Images were acquired during
the same day, typically from 5 cells of similar size from each experimental
condition using identical settings of the instrument. For illustration purposes
images were pseudocolored according to their original fluorochromes, merged
(FluoView software, Olympus), then cropped and assembled (Adobe Photoshop
software, Adobe Systems, Mountain View, CA).
2.6. Pull-down assays of PrP aggresomes with oligo(dT)-cellulose
Cells (106in 6-well plates) were rinsed twice with 2 ml cold PBS, and lysed
EDTA; 0.5 mM DTT; 1 mini EDTA-free protease inhibitor tablet (Roche) per
10 ml] for 15 min on ice. The lysate was centrifuged at 10,000 ×g for 10 min at
4 °C. Forty microliters were kept aside (total input), and the remaining 200 μl
were incubated for 5 min at room temperature with 10 mg of oligo(dT)-cellulose
(GE Healthcare) that had been previously pre-equilibrated in buffer A. After
centrifugation at 2000 ×g for 1 min, supernatant containing unbound proteins
was transferred into a new tube, and the resin was washed twice with 1 ml of
buffer A. The beads were then washed three times with 1 ml of buffer A, and the
bound proteins eluted by incubating for 5 min at 95 °C in SDS-PAGE sample
(dT)-cellulose was pre-incubated with 5 mg polyadenylic acid (polyA, Sigma)
Fig. 1. Translational inhibition in cells producing PrP aggresomes. (A) Total
protein production was measured by analyzing newly synthesized proteins in
and cysteine (Cys)-deficient medium. Where indicated, medium contained
30 μg/ml cycloheximide (CHX). Proteins were separated by 10% SDS-PAGE
and stained with Coomassie blue prior to analysis of radioactivity with a
phosphorimager. The position of the molecular mass markers is indicated on the
right. (B) Cpm levels from autoradiograms of 5 independent experiments were
measured with an Instant Imager. Protein synthesis is expressed as percentage of
cpm incorporated by proteins from cells transfected with empty vector. Means
and SD are shown.
Fig. 2. Phosphorylation of eIF2α in cells with PrP aggresomes. (A) Cells
transiently transfected with empty vector, vector encoding EGFP or CyPrPEGFP
were lysed and processed for immunoblotting using anti-eIF2α antibodies
recognizing eIF2α phosphorylated at Ser51 (P-eIF2α), or both non-phosphory-
obtained with cells treated with a sublethal dose of arsenite (0.5 mM, 30 min).
Molecular weight standards are shown on the right. (B) Immunofluorescence of
N2a cells transiently transfected with CyPrPEGFP, EGFP, empty vector, or cells
treated with arsenite (0.5 mM, 30 min), with an antibody recognizing phos-
phorylated eIF2α at Ser51 (red). Green (left panels) and red channels (middle
right panels). Scale bar: 10 μm. Original magnification ×100.
481K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
incubated with the cell lysates prior to pull-down with oligo(dT)-cellulose. In
control experiments, buffer A did not contain KCl. Pull-down of CyPrPEGFPor
EGFP was analyzed by immunoblotting using anti-EGFP antibodies.
2.7. Metabolic labelling
106cells were transfected either with the empty vector, EGFP or CyPrPEGFP
and incubated for 24 h at 37 °C. Cells were washed twice with PBS and
incubated for 20 min in starvation media (DMEM without methionine and
cysteine, Gibco). Cells were then pulsed with 25 μCi/ml35S labelling mix (Easy
Tag Express protein labelling mix, NEG772, Perkin Elmer) for 1 h at 37 °C. In
control experiments, the translation inhibitor cycloheximide was added at 30 μg/
ml prior to starvation. After labelling, cells were washed twice with PBS,
scraped, collected in 1 ml PBS and centrifuged for 5 min at 5000 rpm at 4 °C.
Pellets were lysed with 100 μl of buffer B [10 mM Tris pH 8.0, 100 mM NaCl,
0,5% Nonidet-P40, 0,5% sodium deoxycholate and 1 mini EDTA-free protease
inhibitor tablet (Roche) per 10 ml]. Proteins were dosed with BCA protein assay
(Pierce) and 50 μg of total proteins were loaded on 10% polyacrylamide
denaturing gels. After electrophoresis, gels were stained with Coomassie blue
and destained overnight. Gels were dried and cpm and total counts were
measured over a 1 h exposure time on an Instant Imager (Packard Instruments).
Gels were then exposed 24 h to a Phosphor screen (GE Healthcare) and scanned
on a Storm 860 Imager (Molecular Dynamics).
For analysis of protein by Western blotting, cells were lysed on ice in lysis
buffer (1% NP-40, 50 mM Tris–HCl, pH 7.6, 150 mM NaCl, and 2 mM EDTA),
clarified by centrifugation at 3000 ×g for 5 min in a microcentrifuge at 4 °C,
separated by SDS-PAGE, transferred to PVDF (Millipore) membranes, and
blotted with respective primary and secondary antibodies. Chemiluminescence
was detected using the ECL reagents (Perkin Elmer). Membranes were stripped
by incubating 30 min in 0.2 N NaOH, washed, blocked, and probed again as
described above. Dilutions of primary antibodies were as follows: anti-EGFP (1/
(1/500), and anti-phospho-PKR-Thr446 (1/1000). Secondary antibodies were
3.1. PrP aggresomes induce a translational arrest and the
phosphorylation of eIF2α
In response to diverse environmental stress conditions,
eukaryotic cells reduce their protein synthesis in order to
conserve anabolic energy for the repair of stress-induced da-
mage. We determined if PrP aggresomes reproduce a cellular
stress and induce a translational arrest. In order to reproduce the
formation of PrP aggresomes in the absence of infectious prions
or any metabolic inhibitors, N2a cells were transiently trans-
fected with an EGFP-tagged cytoplasmic PrP construct termed
CyPrPEGFPthat does not contain the NH2- and COOH-terminal
signal peptides of PrPC. At 24 h after transfection, total
proteinproduction was analyzed by metabolic labelling. Fig. 1A
shows that efficient translation is observed in mock-transfected
cells (lane 1). In contrast, the level of protein translation was
considerably reduced in CyPrPEGFP-expressing cells (lane 3).
Such inhibition did not result from the overexpression of a
cytoplasmic recombinant protein as transfectants expressing
EGFP had a translation efficiency similar to mock-transfected
cells (compare lanes 1 and 2). PrPEGFPdid not affect protein
to N2a cells since a significant decrease in protein translation
wasalso observedinhuman embryonickidney293Tcells(lanes
5–7). The translational inhibitor cycloheximide almost com-
pletely abolished protein synthesis (lanes 4 and 8). Coomassie
staining confirmed that variations in levels of
proteins did not result from variations in protein loading
(Fig. 1A). Cpm counts of
that CyPrPEGFPreduced protein synthesis by more than 70%
A key determinant of stress-induced translational arrest is the
phosphorylation of eIF2α at residue Ser51 that prevents the
assembly of the 43S pre-initiation complex . Therefore, we
monitored the phosphorylation of eIF2α with an antibody that
detects endogenous eIF2α only when phosphorylated at Ser51.
Fig. 2A shows that undetectable levels of phospho-eIF2α were
transfected cells (lane 3). The phosphorylation of eIF2α did not
result from the overexpression of a recombinant cytoplasmic
protein since eIF2α was not phosphorylated in transfectants
expressing EGFP (lane 2). Maximum level of phospho-eIF2α
was achieved by treating cells with a sublethal dose of arsenite to
phospho-eIF2α was not a consequence of a variation in total
protein levels, immunoblots were stripped and probed with an
antibody that recognizes both phosphorylated and non-phos-
phorylated forms of eIF2α. Levels of total eIF2α were similar in
all conditions (Fig. 2A). The phosphorylation of eIF2α was
in cells expressing CyPrPEGFP, whilst no signal was detected in
untransfected and EGFP-transfected cells (Fig 2B). Together,
these results strongly suggest that PrP aggresomes inhibit protein
synthesis through phosphorylation of eIF2α at Ser51.
35S-labelled proteins confirmed
3.2. PKR mediates PrP aggresomes-induced phosphorylation
Once we had established that eIF2α was phosphorylated, we
PrP aggresomes. Four eIF2α kinases have been characterized
that recognize a different set of stress conditions . Heme-
regulated inhibitor HRI is activated by heat and oxidative
stresses; RNA-dependent protein kinase PKR is activated by
double-stranded RNAs and participates in an anti-viral defence
mechanism that is mediated by interferon; general control non-
derepressible-2 GCN2 is induced during amino acid deprivation
and upon inhibition of proteasomes; and PKR-like endoplasmic
reticulum kinase PERK is activated in response to misfolded
protein in the ER (ER stress). Different inhibitors were used and
only a PKR inhibitor efficiently prevented the phosphorylation
of eIF2α (not shown). Fig. 3A shows a representative
experiment using a specific ATP-binding site directed inhibitor
of PKR that effectively inhibits PKR activation . Increasing
the concentration of the inhibitor resulted in the complete
inhibition of the phosphorylation of eIF2α (Fig. 3A, lanes 3–6).
An inactive structural analog of the inhibitor that serves as a
negative control did not prevent the phosphorylation of eIF2α
482K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
(Fig. 3A, lanes 7–10). The levels of total eIF2α were similar in
all experimental conditions (Fig. 3A).
Activation of PKR leads to critical autophosphorylation at
multiple sites, including Thr446 in the activation loop . The
activation status of PKR was determined by Western blot
analysis of whole cell lysates using specific antibodies directed
against phospho-Thr446. Fig. 3A shows that PKR is phos-
phorylated in cells expressing CyPrPEGFP(lanes 3 and 7). In
contrast, PKR is not phosphorylated in cells transfected with
empty vector (lane 1) or cells expressing EGFP (lane 2). The
reduction in levels of phospho-PKR confirmed the efficacy of
the PKR inhibitor (lanes 3–6), whilst the inactive structural
analog had no influence on the phosphorylation of PKR (lanes
7–10). Levels of total PKR were similar in all conditions. We
determined if CyPrPEGFPwas as efficient as the potent PKR
activator polyinosinic–polycytidylic acid or poly(I:C), a
synthetic double-stranded polyribonucleotide used to mimic a
viral infection . Levels of phospho-PKR were similar in
cells expressing CyPrPEGFPand cells treated with poly(I:C)
(Fig. 3B). These results demonstrate that PrP aggresomes
efficiently activate the eIF2α kinase PKR by inducing its
autophosphorylation at Thr446, and confirm our initial hypoth-
esis that PrP aggresomes induce a stress response.
3.3. PrP aggregsomes prevent the formation of stress granules
Stress-induced phosphorylation of eIF2α is necessary and
sufficient for SGs assembly . Therefore, we determined if
SGs are present in cells producing PrP aggresomes. The
formation of SGs was monitored with antibodies directed
against TIAR, a protein that co-aggregates with poly(A)+RNA
at mammalian SGs . Fig. 4A shows that TIAR is localized
in the cytoplasm and the nucleus of N2a cells. Unexpectedly,
TIAR did not relocalize to SGs in cells expressing CyPrPEGFP.
In addition to TIAR, other protein markers of SGs including
HuR , and eIF4E , did not accumulate in cytoplasmic
granules (Fig. 4A). HuR remained mainly located in the
nucleus, and eIF4E in the cytoplasm. In control experiments, a
sublethal dose of arsenite induced the relocalization of TIAR,
HuR, and eIF4E to SGs (Fig. 4A).
These observations indicate that phospho-eIF2α is unable to
induce the assembly of SGs in N2a cells with PrP aggresomes.
Alternatively, the phosphorylation of eIF2α may not be
sufficient for SGs assembly, in contradiction with previous
studies . To address this question, cells were treated with
stresses that activate the four eIF2α kinases. HRI was activated
with arsenite ; PKR was activated with poly(I:C) ;
PERK was activated with thapsigargin ; and GCN2 was
activated in the absence of leucine and in the presence of the
proteasome inhibitor MG132 . All stress conditions resulted
in the formation of SGs in untransfected cells (Fig. 4B). In
contrast, cells expressing CyPrPEGFPdid not assemble SGs after
these treatments. EGFP, used as a control cytoplasmic protein,
did not prevent the assembly of SGs (Fig. 4B). Expression of
wild-type PrP did not prevent the assembly of SGs (Fig. 4C). It
was of interest to test if the inhibition of SGs is specific to PrP
aggresomes. To address this point, we expressed a cytosolic
chimera termed GFP-250. This chimeric polypeptide composed
of the entire soluble protein GFP fused at its COOH terminus to
a 250-amino acid fragment of the cytosolic protein, p115, has
been used to determine the dynamics of aggresomes formation
. GFP-250 aggresomes did not prevent the formation of SGs
(Fig 4D). We tested if PrP aggresomes could inhibit the
formation of SGs in neuronal cells different from N2a. Indeed,
PrP aggresomes prevented the formation of SGs in neuroblas-
toma BE(2)M17 and SK-N-SH cells (Fig. 4E). All together,
these results indicate that the inability to assemble SGs is a
robust and specific characteristic of cells producing cytoplasmic
In contrast to Hela and Cos-7 cells generally used to in-
vestigate the assembly of SGs , N2a cells have never been
used for such studies. Thus, in order to determine if PrP
Fig. 3. PKR mediates the phosphorylation of eIF2α in cells producing
cytoplasmic PrP aggregates. (A) N2a cells were transiently transfected with
empty vector, or vector encoding EGFP or CyPrPEGFPin the presence of
different concentrations of PKR inhibitor (PKR-I), or an inactive analog of PKR
inhibitor (inactive PKR-I). After 24 h, cells were lysed and processed for
immunoblotting using an antibody recognizing eIF2α phosphorylated at Ser51
(P-eIF2α), an antibody recognizing both phosphorylated and non-phosphory-
lated eIF2α, an antibody recognizing PKR phosphorylated at Thr446 (P-PKR),
and an antibody recognizing both phosphorylated and non-phosphorylated
PKR, respectively. (B) Cells were either transiently transfected with empty
vector, EGFP, and CyPrPEGFP, or were non-transfected and treated with the PKR
activator poly(I:C). After cell lysis, proteins were separated in a 10% SDS-
PAGE and proteins analyzed by immunoblotting using an antibody recognizing
phosphorylated PKR, or an antibody recognizing both phosphorylated and non-
phosphorylated PKR. The position of the molecular mass markers is indicated
on the right.
483K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
Fig. 4. PrP aggresomes inhibit the assembly of SGs. (A) N2a cells were transiently transfected with CyPrPEGFP(top panels), or treated with a sublethal dose of arsenite
(bottom panels). After 24 h, cells were fixed and processed for immunofluorescence using anti-TIAR, anti-eIF4E, and anti-HuR antibodies (red). (B) N2a cells
transiently transfected with CyPrPEGFPor EGFP were treated with arsenite, thapsigargin, poly(I:C), or incubated in the absence of leucine and in the presence of
MG132 as described under Experimental procedures. The cells were fixed and processed for immunofluorescence using anti-TIAR antibodies (red). (C–D) N2a cells
were transiently transfected with PrPEGFP(C) or GFP-250 (D). After 24 h, cells were untreated or treated with a sublethal dose of arsenite, fixed and processed for
immunofluorescence using anti-TIAR antibodies (red). (E) BE(2)M17 or SK-N-SH cells were transiently transfected with CyPrPEGFP. After 24 h, cells were untreated
or treated with a sublethal dose of arsenite, fixed and processed for immunofluorescence using anti-TIAR antibodies (red). (F) Hela cells transiently transfected with
CyPrPEGFPor EGFP were treated with arsenite, dithiotreitol, poly(I:C), and processed for immunofluorescence using anti-TIAR antibodies (red). In all panels, red and
green channels are shown merged. Nuclei of transfected cells are labelled with asterisks. For cells transfected with EGFP, red, green, and merged channels are shown
from top to bottom. Scale bar: 10 μm. Original magnification ×100.
484K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
cellular models of SGs formation, we also tested the effect of
CyPrPEGFPexpression in Hela cells (Fig. 4E). As expected,
treatment of cells with arsenite, dithiotreitol (ER stress), and poly
(I:C) resulted in the assembly of SGs in untransfected cells.
Similarly to N2a cells, expression of CyPrPEGFPprevented the
formation of SGs in Hela cells. Interestingly, Hela cells did not
produce typical juxtanuclear aggresomes, and PrP aggregates
We can conclude that coalescence of dispersed PrP aggregates
into a typical aggresomal structure at the centrosome is not
necessary for the inhibition of SGs assembly.
The observation that all stress conditions do not lead to the
formation of SGs strongly suggest that PrP aggresomes interfere
validity of this hypothesis was tested by expressing the S51D
phosphomimetic eIF2α mutant. Transfection of this mutant
normally results in the constitutive assembly of SGs . N2a
cells were co-transfected with either vector alone and mutant
mutant eIF2α formed TIAR-positive SGs, whereas cells co-
expressing mutant eIF2α and CyPrPEGFPdid not assemble SGs.
As expected, cells transfected with wild-type eIF2α did not
assemble SGs, and the distribution of TIAR remained diffuse in
CyPrPEGFPprevents the assembly of SGs downstream of the
phosphorylation of eIF2α.
3.4. mRNA clustering with PrP aggresomes
In mammalian cells, the majority of all poly(A)+mRNA is
recruited to SGs . We reasoned that the absence of SGs
assembly might result from the co-aggregation of mRNAs with
PrP aggresomes. Indeed, clustering of mRNAs with PrP aggre-
somes would physically prevent their recruitment to SGs, thus
preventing the formation of these RNA granules. Furthermore,
such mechanism would explain the incapacity of phosphomi-
metic eIF2α to trigger the assembly of SGs as observed in
Fig. 5. We used in situ hybridization with an oligo-dT probe and
confocal microscopy to determine the subcellular distribution of
poly(A)+RNA. In untransfected cells, poly(A)+RNA displayed
a punctuate and diffuse staining and was distributed between the
nucleus and the cytoplasm (Fig 6A). In contrast, poly(A)+RNA
co-aggregated with PrP aggresomes in transfected cells. Merged
confocal images show a high degree of co-localization between
PrP aggresomes and poly(A)+. Higher magnification views
confirmed that PrP aggresomes may represent poly(A)+ribonu-
cleoprotein complexes (Fig. 6B). Alkaline hydrolysis treatment
of cells to eliminate RNAs prior to in situ hybridization con-
firmed the specificity of the oligo-dT probe (Fig 6A). To
validate that co-aggregation of mRNA with PrP aggresomes
prevents the coalescence of mRNAs into SGs, cells were
exposed to a sublethal concentration of arsenite. Fig. 6C shows
that poly(A)+RNA accumulates in SGs in untransfected cells.
However, poly(A)+RNA remained co-aggregated within PrP
aggresomes following arsenite treatment. In control experi-
ments, EGFP did not prevent the relocalization of poly(A)+
RNA to SGs (Fig. 6C). Furthermore, the aggresome-forming
protein GFP-250 did not induce the co-aggregation of mRNAs
and the formation of poly(A)+-containing SGs (Fig. 6D). The
mRNA clustering activity of PrP aggresomes was not restricted
to N2a cells, and was also observed in BE(2)M17 and SK-N-SH
cells (Fig. 6E).
In poly(A)+RNA pull-down experiments designed to confirm
the interaction between PrP aggresomes and poly(A)+, we
examined the binding of CyPrPEGFP–poly(A)+RNA complexes
to an oligo(dT)-cellulose resin. Cell lysates from transfectants
expressing CyPrPEGFPor the control cytoplasmic protein EGFP
were incubated with oligo(dT)-cellulose. In contrast to EGFP,
CyPrPEGFPwas recovered in the oligo(dT)-cellulose eluate
fraction (Fig. 6F, compare lanes 6 and 12). Three types of control
experiments were performed to verify that CyPrPEGFPhad bound
to the oligo(dT)-cellulose in association with poly(A)+RNA.
First, soluble poly(A) efficiently competed for the binding of
CyPrPEGFPto the resin (Fig. 6F, compare lanes 3 and 7). Second,
addition of soluble poly(U) to cell lysates prior to pull-down
prevented the binding of CyPrPEGFPto oligo(dT)-cellulose
(Fig. 6F, compare lanes 4 and 8). Third, in the absence of KCl
which is required for poly(A)+RNA binding to oligo(dT)-
cellulose, CyPrPEGFPdid not bind to the resin (Fig. 6F, compare
PrP aggregates purify on the oligo(dT)-cellulose column as
We next asked if poly(A)+RNA clustered within PrP
aggresomes is still associated with mRNA-binding complexes,
including the CAP-binding complex and ribosomes. To address
Fig. 5. AphosphomimeticmutantofeIF2αisunabletoinducetheformationofSGs
using anti-TIAR antibodies (red). Red and green channels are shown merged. Scale
485K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
this question, we used immunofluorescence microscopy with
antibodies directed against eIF4E. eIF4E interacts with the 7
methyl GTP cap structure of mRNAs, and allows the assembly
of the multisubunit eIF4F complex . The eIF4F complex
then regulates the binding of mRNA to the ribosomal 43S pre-
initiation complex. The distribution of eIF4E was diffuse in the
Fig. 6. Co-aggregationofpoly(A)+RNAwithPrP aggresomes. (A) N2acells were transiently transfectedwith CyPrPEGFP. After 24 h, cells were fixed, permeabilized, and
incubated in the absence (upper panels) or in the presence (bottom panels) of NaOH. Cells were processed for in situ hybridization to detect poly(A)+RNA (red), and
analyzed by confocal microscopy. The co-localized signals of CyPrPEGFPand red-labelled poly(A)+appear yellow. (B) A magnification of poly(A)+RNA and CyPrPEGFP
co-aggregates from the merge upper right panel of panel A. Scale bar: 5 μm. (C) N2a cells were transiently transfected with CyPrPEGFP(upper panels) or EGFP (bottom
confocal microscopy. (D) N2acellswere transiently transfected withGFP-250. After 24 h, cellswere treatedwith a sublethal dose of arsenite(0.5mM, 30min), fixed, and
processed for in situ immunofluorescence to detect poly(A)+by fluorescence microscopy. (E) BE(2)M17 and SK-N-SH cells were transiently transfected with CyPrPEGFP.
by fluorescence microscopy. (F) Lysate from cells expressing CyPrPEGFPor EGFP was incubated with oligo(dT)-cellulose beads. Following a 5-min incubation, the flow-
transiently transfected with CyPrPEGFPfor 24 h, or treated with a sublethal dose of arsenite. Cells were processed for immunofluorescence using eIF4E or protein S6
antibodies. (A), (C–E), (G), asterisks indicate the nucleus of transfected cells; scale bar: 10 μm; original magnification ×60.
486K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
cytoplasm and was not perturbed by the presence of PrP
aggresomes (Fig. 6G). The absence of 43S pre-initiation
complex within PrP aggresomes was confirmed with antibodies
directed against 43S ribosomal protein S6. Like eIF4E, protein
S6 did not co-aggregate with CyPrPEGFP(Fig. 6G). The
specificity of anti-eIF4E and anti-S6 antibodies was verified in
control experiments in which relocation of both proteins into
SGs was induced by treating cells with arsenite (Fig. 6G). These
results indicate that poly(A)+RNA co-aggregated with PrP
aggresomes is not associated with the translational machinery.
3.5. PrP aggresomes inhibit the synthesis of Hsp70
The three major components of the cell stress response are the
inhibition of protein translation, assembly of poly(A)+RNA in
SGs, and synthesis of heat shock proteins (Hsps). Hsps are
protein aggregation, participate in refolding or elimination of
misfolded proteins in their capacity as chaperones, and display
anti-apoptotic activity. The results presented above raised the
possibility that PrP aggresomes impair a full cell stress response.
In order to examine the impact of PrP aggresomes on the in-
duction of Hsps, we determined the expression of Hsp70. In
contrast to many Hsps that are expressed constitutively in cells
mammalian cells only after heat shock and other forms of cell
stress . Hsp70 could not be detected by immunoblot analysis
immediately following a treatment with a sublethal dose of
arsenite in either mock-transfected cells, or cells transiently
transfected with EGFP or CyPrPEGFP(Fig. 7A, lanes 1–3,
respectively). In contrast, Hsp70 was largely induced as soon as
4 h post-stress in mock-transfected cells and EGFP-expressing
cells (lanes 4 and5,respectively),butwas barelydetectedincells
expressing CyPrPEGFP(lane 6). The inhibition of Hsp70
induction was also observed 8 h post-stress, indicating a lasting
inhibition of the synthesis of Hsp70. This inhibition was also
confirmed by immunofluorescence with antibodies specific for
inducible Hsp70 (Fig. 7B).
3.6. Cells producing PrP aggresomes are more sensitive to
The assembly of SGs and the synthesis of Hsps are
mechanisms designed to help cells to recover from various
Fig. 6 (continued).
487K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
environmental stresses. The preceding results indicate that PrP
aggresomes inhibit a normalcell stress response, suggesting that
cells with aggresomes might be more susceptible to stress. To
test this hypothesis, N2a transfectants expressing EGFP or
CyPrPEGFPwere pulsed with increasing doses of arsenite for
30 min, and then allowed to recover in media without arsenite.
The cells were fixed after 8 h, and the percentage of cell death
was measured by condensed chromatin and fragmented nuclei
with Hoechst staining. Fig. 8 shows that basal level of cell death
in mock-transfected cells and cells expressing EGFP is between
3 and 5%. In agreement with a previous study , cells
expressing CyPrPEGFPdisplayed higher levels of cell death
(Fig. 8). The level of cell death in CyPrPEGFP-expressing cells
compared to mock-transfected cells and EGFP-expressing cells
was higher at any concentration of arsenite, except at 4 mM
which was lethal in all types of cells.
In this study, we have tested the hypothesis that PrP
aggresomes induce a cell stress response. Our results demon-
strate that PrP aggresomes provoke a PKR-mediated cellular
stress response characterized by the phosphorylation of eIF2α
and a translational arrest. However, the stress response is only
partial since cells do not assemble SGs, and do not induce the
synthesis of Hsp70. In addition, cytoplasmic PrP aggregates
interact with poly(A)+RNA to form ribonucleoprotein com-
plexes. These mechanisms may not be obligatory lethal per se,
but would likely result in premature cell death in the context of
an acute environmental stress that would be otherwise dealt with
an adequate stress response.
4.1. PrP aggresomes are poly(A)+ribonucleoprotein complexes
A major finding of this study resides in the co-aggregation of
poly(A)+RNA with PrP aggresomes. The nucleic-acid binding
activity of recombinant PrP has already been shown in vitro
[36–41]. Our results indicate for the first time that formation of
complexes between PrP and poly(A)+RNA can also occur in
Fig. 7. PrP aggresomes prevent the induction of Hsp70. (A) Cells were
transiently transfected with empty vector, EGFP, or CyPrPEGFP. At 24 h after
transfection, cells were treated with a sublethal dose of arsenite (0.5 mM,
30 min), and allowed to recover for the indicated time at 37 °C. Cells were lysed
and the expression of Hsp70 was analyzed by Western blot. Expression of
CyPrPEGFPand EGFP was determined with antibodies directed against EGFP.
Equal loading was checked with anti-actin antibodies. The position of the
molecular mass markers is indicated on the right. (B) Analysis of Hsp70
induction by confocal microscopy. N2a cells were transiently transfected with
CyPrPEGFP, EGFP, or empty vector. At 24 h after transfection, cells were treated
as in (A) and allowed to recover for 4 h. Expression of Hsp70 was analyzed by
immunofluorescence using an anti-Hsp70 antibody (red). Green (left panels),
red (middle panels), and merged (right panels) channels are shown. Asterisks
indicate the nucleus of cells expressing CyPrPEGFPor EGFP. Scale bar: 10 μm.
Original magnification ×60.
Fig. 8. Cells with PrP aggresomes are more sensitive to arsenite. N2a cells were
transiently transfected with empty vector, EGFP, or CyPrPEGFP. Twenty four
hours after transfection, cells were treated with arsenite at the indicated
concentrations for 30 min. After 8 h of recovery, the percentageof cell death was
measuredby condensedchromatinandfragmentednuclei withHoechststaining.
Data represent the mean and SD of three independent experiments. More than
200 cells were counted for each condition.
488K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
vivo, in the cytoplasm of cells. Since the CAP-binding complex
protein eIF4E and the 43S ribosomal protein S6 are not present
in PrP ribonucleoprotein complexes, poly(A)+RNA cannot be
translated. Phosphorylation of eIF2α further ensures that
protein translation is arrested. Taken together, these observa-
tions show the remarkable efficacy of PrP aggresomes as
To date, the only protein known to induce the aggregation of
poly(A)+RNA in the cytoplasm is TIA-1 . A glutamine-rich
prion-related domain of TIA-1 is linked to RNA recognition
motifs and mediates the assembly of poly(A)+RNA-containing
SGs . At present, the mechanism by which PrP aggresomes
induce the aggregation of poly(A)+RNA is not known.
However, two pieces of evidence suggest that this mechanism
is different from TIA-1. First, in contrast to TIA-1, PrP does not
possess typical RNA-binding domains including ribonucleo-
protein consensus octapeptide sequences that are particularly
conserved between RNA-binding proteins . Second, whilst
eIF4E and small ribosomal subunits accumulate with poly(A)+
RNA in SGs , eIF4E and ribosomal protein S6 do not
accumulate in poly(A)+-PrP ribonucleoprotein complexes.
If co-aggregation of PrP with poly(A)+RNA has a clear
consequence on protein translation, it may also have structural
implications for PrP. Binding of recombinant PrP to either DNA
or RNA in vitro induces profound conformational rearrange-
ments and results in a partially protease-K resistant (PrPRes)
isoform [38,44,45]. Furthermore, RNA molecules stimulate
transconformation of PrPCinto PrPResin vitro by protein-
misfolding cyclic amplification . Based on these studies and
the present results, it is tempting to propose that interactions
between poly(A)+RNA and PrP may be a facilitating if not an
essential factor in the conversion of PrP into PrPResoccurring in
the cytosol [8,47,48]. However, the possibility that PrP
conversion into PrPResprecedes binding to poly(A)+RNA
cannot be excluded.
4.2. PrP aggresomes and the cellular stress response
The second major finding of the present work relates to
critical alteration of the stress response in cells with PrP
aggresomes. These cells can neither assemble SGs, nor induce
the synthesis of Hsp70. The direct functional consequence of
this is an increased sensitivity to environmental stress, including
arsenite-mediated oxidative stress. Hsp70 is thought to prevent
aggregation or assist in the refolding of misfolded proteins .
There is an abundant literature describing the benefit of Hsp70
on the protection from cell death induced by various noxious
stimuli in cultured neurons and in animal models of nervous
system injury [50,51]. Consequently, induction of Hsp70 has
been envisaged as a potential strategy in order to treat
neurodegenerative diseases . The inhibition of expression
of Hsp70 in cells with PrP aggresomes would likely result in
enhanced cell vulnerability and/or premature neurodegeneration
under environmental stress conditions.
Interestingly, Tatzelt et al.  reported that scrapie-infected
N2a cells do not induce the synthesis of Hsp70 following
arsenite or heat shock treatment. This common characteristic
between prion-infected cells and cells with PrP aggresomes is
remarkable and supports the suggestion that PrP aggresomes
may be present in scrapie-infected animals and participate in
neurotoxic mechanisms in prion diseases .
4.3. Cytoplasmic PrP aggregates and PKR-mediated
phosphorylation of eIF2α
Our results reveal that cytoplasmic PrP induces a PKR-
mediated phosphorylation of eIF2α. The increase in phospho-
eIF2α levels attenuates protein synthesis, and may be
considered as a protective mechanism aiming at diminishing
the load of newly synthesized unfolded PrP. As such, PKR
would act as an important modulator of protein synthesis in
response to the presence of cytoplasmic PrP aggregates. How-
ever, constitutive activation of PKR and phosphorylation of
eIF2α are likely to result in cell death [54,55]. It will be
important to address the mechanism of activation of PKR.
Several studies suggest a relationship between activation of
PKR and neurodegeneration in Alzheimer's disease [56–58],
Huntington's disease , and Parkinson's disease .
Recently, it was shown that in contrast to Alzheimer's disease,
there is no significant phosphorylation of eIF2α in the brain of
patients deceased from prion diseases . However, it is
possible that phosphorylation of eIF2α occurs in the early
stages of the disease only. Time-course studies are possible in
animal models of prion diseases and may answer this question.
In the same study, the authors did perform a time-course
investigation; however, phosphorylation of eIF2α could not be
determined because a mouse-specific antibody against phos-
pho-eIF2α was not available.
4.4. Cytosolic PrP in the pathogenesis of prion disease
The role of intracellular PrP in the pathogenesis of prion
diseases is controversial. Some data argue that accumulation of
PrPCin the cytoplasm is associated with neuronal cell death
[8,12,62], whereas other reports do not favour the hypothesis
that cytoplasmic accumulation of misfolded PrP has neurotoxic
consequences [11,21,63]. One explanation resides in the use of
non-specific or high concentration of proteasome inhibitors. In
addition, many studies rely on apoptosis measurements. The
toxicity of cytosolic PrP may involve in part apoptotic-
independent pathways, including cell necrosis which is rarely
investigated. Recently, it was reported that cytoplasmic prion
protein aggresomes are present in prion-infected cells and mice,
adding new evidence supporting a role of cytoplasmic PrP
aggregates in prion diseases . Cytoplasmic PrP aggregates
may induce neurotoxic mechanisms through inhibition of the
26S proteasome .
In summary, identification of PrP aggresomes as poly(A)+
ribonucleoprotein complexes provides new experimental frame-
work to investigate the mechanism of formation of cytoplasmic
PrP aggregates. Furthermore, our findings on the activation of
PKR and the inhibition of the cell stress response should prove
to be extremely useful for better understanding the biological
significance of cytoplasmic PrP aggregates in prion diseases.
489K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
We thank Dr Leonid Volkov for his expertise in confocal
analysis, and Karine Drapeau for technical assistance. We also
thank Dr Elisabeth Sztul (University of Alabama at Birming-
ham, AL USA) for providing the construct encoding GFP-250.
This work was supported by grants from the Canadian Institutes
for Health Research and Prionet Canada to XR. XR is a Junior 2
research scholar from the Fonds de la Recherche en Santé du
 R.R. Kopito, Aggresomes, inclusion bodies and protein aggregation,
Trends Cell Biol. 10 (2000) 524–530.
 R. Garcia-Mata, Y.S. Gao, E. Sztul, Hassles with taking out the garbage:
aggravating aggresomes, Traffic 3 (2002) 388–396.
 J.A. Johnston, C.L. Ward, R.R. Kopito, Aggresomes: a cellular response to
misfolded proteins, J. Cell Biol. 143 (1998) 1883–1898.
 S. Brandner, S. Isenmann, A. Raeber, M. Fischer, A. Sailer, Y. Kobayashi,
S. Marino, C. Weissmann, A. Aguzzi, Normal host prion protein necessary
for scrapie-induced neurotoxicity, Nature 379 (1996) 339–343.
 A.F. Hill, J. Collinge, Subclinical prion infection, Trends Microbiol. 11
Depleting neuronal PrP in prion infection prevents disease and reverses
spongiosis, Science 302 (2003) 871–874.
 J. Ma, S. Lindquist, Conversion of PrP to a self-perpetuating PrPSc-like
conformation in the cytosol, Science 298 (2002) 1785–1788.
 J. Ma, R. Wollmann, S. Lindquist, Neurotoxicity and neurodegeneration
when PrP accumulates in the cytosol, Science 298 (2002) 1781–1785.
 C. Grenier, C. Bissonnette, L. Volkov, X. Roucou, Molecular morphology
and toxicity of cytoplasmic prion protein aggregates in neuronal and non-
neuronal cells, J. Neurochem. 97 (2006) 1456–1466.
 B. Drisaldi, R.S. Stewart, C. Adles, L.R. Stewart, E. Quaglio, E. Biasini, L.
Fioriti, R. Chiesa, D.A. Harris, Mutant PrP is delayed in its exit from the
endoplasmic reticulum, but neither wild-type nor mutant PrP undergoes
retrotranslocation prior to proteasomal degradation, J. Biol. Chem. 278
 N.S. Rane, J.L. Yonkovich, R.S. Hegde, Protection from cytosolic prion
 A. Orsi, L. Fioriti, R. Chiesa, R. Sitia, Conditions of endoplasmic
reticulum stress favor the accumulation of cytosolic prion protein, J. Biol.
Chem. 281 (2006) 30431–30438.
 S.W. Kang, N.S. Rane, S.J. Kim, J.L. Garrison, J. Taunton, R.S. Hegde,
Substrate-specific translocational attenuation during ER stress defines a
pre-emptive quality control pathway, Cell 127 (2006) 999–1013.
 M. Kristiansen,M.J. Messenger, P.C. Klohn,S. Brandner, J.D.Wadsworth,
J. Collinge, S.J. Tabrizi, Disease-related prion protein forms aggresomes in
neuronal cells leading to caspase activation and apoptosis, J. Biol. Chem.
280 (2005) 38851–38861.
 T.E. Dever, A.C. Dar, F. Sicheri, The eiF2α kinases, in: M.B. Mathews, N.
Sonenberg, J.W.B. Hershey (Eds.), Translational Control in Biology and
 J.W.Hershey, W.C.Merrick,Initiationofproteinsynthesis, in:N. Sonenberg
(Ed.), Translational Control of Gene Expression, Cold Spring Harbor
Laboratory, Cold Spring Harbor, N.Y., 2000, pp. 33–38.
 P. Anderson, N. Kedersha, Stressful initiations, J. Cell Sci. 115 (2002)
 F.U. Hartl, Molecular chaperones in cellular protein folding, Nature 381
 H.M. Beere, Death versus survival: functional interaction between the
apoptotic and stress-inducible heat shock protein pathways, J. Clin. Invest.
115 (2005) 2633–2639.
 X. Roucou, Q. Guo, Y. Zhang, C.G. Goodyer, A.C. LeBlanc, Cytosolic
prion protein is not toxic and protects against Bax-mediated cell death in
human primary neurons, J. Biol. Chem. 278 (2003) 40877–40881.
 R.F. Duncan, J.W. Hershey, Translationalrepression bychemical inducersof
the stress response occurs by different pathways, Arch. Biochem. Biophys.
256 (1987) 651–661.
 N.V. Jammi, L.R. Whitby, P.A. Beal, Small molecule inhibitors of the
RNA-dependent protein kinase, Biochem. Biophys. Res. Commun. 308
 F. Zhang, P.R. Romano, T. Nagamura-Inoue, B. Tian, T.E. Dever, M.B.
Mathews, K. Ozato, A.G. Hinnebusch, Binding of double-stranded RNA to
protein kinase PKR is required for dimerization and promotes critical
autophosphorylationevents in the activation loop, J. Biol.Chem.276 (2001)
 M.K. Offermann, J. Zimring, K.H. Mellits, M.K. Hagan, R. Shaw, R.M.
Medford, M.B. Mathews, S. Goodbourn, R. Jagus, Activation of the
double-stranded-RNA-activated protein kinase and induction of vascular
cell adhesion molecule-1 by poly (I).poly (C) in endothelial cells, Eur. J.
Biochem. 232 (1995) 28–36.
 N.L. Kedersha, M. Gupta, W. Li, I. Miller, P. Anderson, RNA-binding
proteins TIA-1 and TIAR link the phosphorylation of eIF-2 alpha to the
 I.E. Gallouzi, C.M. Brennan, M.G. Stenberg, M.S. Swanson, A. Eversole,
N. Maizels, J.A. Steitz, HuR binding to cytoplasmic mRNA is perturbed
by heat shock, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 3073–3078.
 N. Kedersha, S. Chen, N. Gilks, W. Li, I.J. Miller, J. Stahl, P. Anderson,
Evidence that ternary complex (eIF2–GTP–tRNA(i)(Met))-deficient pre-
initiation complexes are core constituents of mammalian stress granules,
Mol. Biol. Cell 13 (2002) 195–210.
 L.Lu, A.P. Han, J.J. Chen, Translationinitiationcontrolby heme-regulated
eukaryotic initiation factor 2alpha kinase in erythroid cells under cyto-
plasmic stresses, Mol. Cell. Biol. 21 (2001) 7971–7980.
 S.R. Kimball, R.L. Horetsky, D. Ron, L.S. Jefferson, H.P. Harding,
Mammalian stress granules represent sites of accumulation of stalled
translation initiation complexes, Am. J. Physiol., Cell Physiol. 284 (2003)
 R.M. Vabulas, F.U. Hartl, Protein synthesis upon acute nutrient restriction
relies on proteasome function, Science 310 (2005) 1960–1963.
 R. Garcia-Mata, Z. Bebok, E.J. Sorscher, E.S. Sztul, Characterization and
dynamics of aggresome formation by a cytosolic GFP-chimera, J. Cell
Biol. 146 (1999) 1239–1254.
 N. Kedersha, G. Stoecklin, M. Ayodele, P. Yacono, J. Lykke-Andersen, M.J.
Fritzler, D. Scheuner, R.J. Kaufman, D.E. Golan, P. Anderson, Stress
granules and processing bodies are dynamically linked sites of mRNP
remodeling, J. Cell Biol. 169 (2005) 871–884.
 A.C. Gingras, B. Raught, N. Sonenberg, eIF4 initiation factors: effectors of
mRNA recruitment to ribosomes and regulators of translation, Annu. Rev.
Biochem. Allied Res. India 68 (1999) 913–963.
 S. Lindquist, E.A. Craig, The heat-shock proteins, Annu. Rev. Genet. 22
 P.K. Nandi, E. Leclerc, Polymerization of murine recombinant prion
protein in nucleic acid solution, Arch. Virol. 144 (1999) 1751–1763.
aggregation of linear nucleic acids to condensed globular structures, Arch.
Virol. 146 (2001) 327–345.
 C. Gabus, S. Auxilien, C. Pechoux, D. Dormont, W. Swietnicki, M.
Morillas, W. Surewicz, P. Nandi, J.L. Darlix, The prion protein has DNA
strand transfer properties similar to retroviral nucleocapsid protein, J. Mol.
Biol. 307 (2001) 1011–1021.
 E. Derrington, C. Gabus, P. Leblanc, J. Chnaidermann, L. Grave, D.
Dormont, W. Swietnicki, M. Morillas, D. Marck, P. Nandi, J.L. Darlix,
PrPC has nucleic acid chaperoning properties similar to the nucleocapsid
protein of HIV-1, C. R. Biol. 325 (2002) 17–23.
 M. Moscardini, M. Pistello, M. Bendinelli, D. Ficheux, J.T. Miller, C.
Gabus, S.F. Le Grice, W.K. Surewicz, J.L. Darlix, Functional interactions
490K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491
of nucleocapsid protein of feline immunodeficiency virus and cellular
prion protein with the viral RNA, J. Mol. Biol. 318 (2002) 149–159.
 P.K. Nandi, J.C. Nicole, Nucleic acid and prion protein interaction
produces spherical amyloids which can function in vivo as coats of
spongiform encephalopathy agent, J. Mol. Biol. 344 (2004) 827–837.
 N. Gilks, N. Kedersha, M. Ayodele, L. Shen, G. Stoecklin, L.M. Dember,
P. Anderson, Stress granule assembly is mediated by prion-like
aggregation of TIA-1, Mol. Biol. Cell 15 (2004) 5383–5398.
 R.J. Bandziulis, M.S. Swanson, G. Dreyfuss, RNA-binding proteins as
developmental regulators, Genes Dev. 3 (1989) 431–437.
 Y. Cordeiro, F. Machado, L. Juliano, M.A. Juliano, R.R. Brentani, D.
Foguel, J.L. Silva, DNA converts cellular prion protein into the beta-sheet
conformation and inhibits prion peptide aggregation, J. Biol. Chem. 276
 V. Adler, B. Zeiler, V. Kryukov, R. Kascsak, R. Rubenstein, A. Grossman,
Small, highly structured RNAs participate in the conversion of human
recombinantPrP(Sen)toPrP(Res)invitro,J.Mol. Biol.332 (2003)47–57.
 N.R. Deleault, R.W. Lucassen, S. Supattapone, RNA molecules stimulate
prion protein conversion, Nature 425 (2003) 717–720.
and ubiquitin are involved in the turnover of the wild-type prion protein,
EMBO J. 20 (2001) 5383–5391.
 E. Cohen, A. Taraboulos, Scrapie-like prion protein accumulates in
aggresomes of cyclosporin A-treated cells, EMBO J. 22 (2003) 404–417.
 U. Hartl, Highlight: protein folding in vivo, Biol. Chem. 379 (1998) 235.
 M.A. Yenari, Heat shock proteins and neuroprotection, Adv. Exp. Med.
Biol. 513 (2002) 281–299.
 T.B. Franklin, A.M. Krueger-Naug, D.B. Clarke, A.P. Arrigo, R.W. Currie,
The role of heat shock proteins Hsp70 and Hsp27 in cellular protection of
the central nervous system, Int. J. Hyperther. 21 (2005) 379–392.
 A. Klettner, The induction of heat shock proteins as a potential strategy to
treat neurodegenerative disorders, Drug News Perspect. 17 (2004) 299–306.
 J. Tatzelt, J. Zuo, R. Voellmy, M. Scott, U. Hartl, S.B. Prusiner, W.J.
Welch, Scrapie prions selectively modify the stress response in
neuroblastoma cells, Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 2944–2948.
 S.P. Srivastava, K.U. Kumar, R.J. Kaufman,Phosphorylation of eukaryotic
translation initiation factor 2 mediates apoptosis in response to activation
of the double-stranded RNA-dependent protein kinase, J. Biol. Chem. 273
 D. Scheuner, R. Patel, F. Wang, K. Lee, K. Kumar, J. Wu, A. Nilsson, M.
Karin, R.J. Kaufman, Double-stranded RNA-dependent protein kinase
phosphorylation of the alpha-subunit of eukaryotic translation initiation
factor 2 mediates apoptosis, J. Biol. Chem. 281 (2006) 21458–21468.
 A.L. Peel, D.E. Bredesen, Activation of the cell stress kinase PKR in
Alzheimer's disease and human amyloid precursor protein transgenic
mice, Neurobiol. Dis. 14 (2003) 52–62.
 K.C. Suen, M.S. Yu, K.F. So, R.C. Chang, J. Hugon, Upstream signaling
pathways leading to the activation of double-stranded RNA-dependent
Chem. 278 (2003) 49819–49827.
 R. Onuki, Y. Bando, E. Suyama, T. Katayama, H. Kawasaki, T. Baba, M.
Tohyama, K. Taira, An RNA-dependent protein kinase is involved in
tunicamycin-induced apoptosis and Alzheimer's disease, EMBO J. 23
 A.L. Peel, R.V. Rao, B.A. Cottrell, M.R. Hayden, L.M. Ellerby, D.E.
Bredesen, Double-stranded RNA-dependent protein kinase, PKR, binds
preferentially to Huntington's disease (HD) transcripts and is activated in
HD tissue, Hum. Mol. Genet. 10 (2001) 1531–1538.
 Y. Bando, R. Onuki, T. Katayama, T. Manabe, T. Kudo, K. Taira, M.
Tohyama, Double-strand RNA dependent protein kinase (PKR) is involved
in the extrastriatal degeneration in Parkinson's disease and Huntington's
disease, Neurochem. Int. 46 (2005) 11–18.
not in prion diseases in vivo, J. Neuropathol. Exp. Neurol. 65 (2006)
 X. Wang, F. Wang, M.S. Sy, J. Ma, Calpain and other cytosolic proteases
can contribute to the degradation of retro-translocated prion protein in the
cytosol, J. Biol. Chem. 280 (2004) 317–325.
 L. Fioriti, S. Dossena, L.R. Stewart, R.S. Stewart, D.A. Harris, G. Forloni,
R. Chiesa, Cytosolic prion protein (PrP) is not toxic in N2a cells and
primary neurons expressing pathogenic PrP Mutations, J. Biol. Chem. 280
 M. Kristiansen, P. Deriziotis, D.E. Dimcheff, G.S. Jackson, H. Ovaa, H.
Naumann, A.R. Clarke, F.W. van Leeuwen, V. Menendez-Benito, N.P.
Dantuma, J.L. Portis, J. Collinge, S.J. Tabrizi, Disease-associated prion
protein oligomers inhibit the 26S proteasome, Mol. Cell 26 (2007)
491K. Goggin et al. / Biochimica et Biophysica Acta 1783 (2008) 479–491