The FASEB Journal article fj.11-201772. Published online June 1, 2012.
The FASEB Journal • Research Communication
Endogenous prion protein conversion is required for prion-induced neuritic alterations and neuronal death Sabrina Cronier,*,1 Julie Carimalo,‡,§ Brigitte Schaeffer,† Emilie Jaumain,* Vincent Béringue,* Marie-Christine Miquel,‡ Hubert Laude,* and Jean-Michel Peyrin*,‡,1 *UR892, Virologie et Immunologie Moléculaires, and †UR0341, Mathématiques et Informatique Appliquées, Institut National de la Recherche Agronomique (INRA), Jouy-en-Josas, France, ‡ Neurobiologie des Processus Adaptatifs, Unité Mixte de Recherche (UMR) Centre National de la Recherche Scientifique (CNRS)-7102, Université Pierre et Marie Curie, Paris, France; and §National Reference Centre for Transmissible Spongiform Encephalopathy Surveillance, Department of Neurology, Georg August University, Göttingen, Germany ABSTRACT Prions cause fatal neurodegenerative conditions and result from the conversion of hostencoded cellular prion protein (PrPC) into abnormally folded scrapie PrP (PrPSc). Prions can propagate both in neurons and astrocytes, yet neurotoxicity mechanisms remain unclear. Recently, PrPC was proposed to mediate neurotoxic signaling of -sheet-rich PrP and non-PrP conformers independently of conversion. To investigate the role of astrocytes and neuronal PrPC in prion-induced neurodegeneration, we set up neuron and astrocyte primary cocultures derived from PrP transgenic mice. In this system, prion-infected astrocytes delivered ovine PrPSc to neurons lacking PrPC (prion-resistant), or expressing a PrPC convertible (sheep) or not (mouse, human). We show that interaction between neuronal PrPC and exogenous PrPSc was not sufficient to induce neuronal death but that efficient PrPC conversion was required for prion-associated neurotoxicity. Prion-infected astrocytes markedly accelerated neurodegeneration in homologous cocultures compared to infected single neuronal cultures, despite no detectable neurotoxin release. Finally, PrPSc accumulation in neurons led to neuritic damages and cell death, both potentiated by glutamate and reactive oxygen species. Thus, conversion of neuronal PrPC rather than PrPC-mediated neurotoxic signaling appears as the main culprit in prion-induced neurodegeneration. We suggest that active prion replication in neurons sensitizes them to environmental stress regulated by neighboring cells, including astrocytes.—
Abbreviations: CAS, cerebellar astrocyte; CGN, cerebellar granule neuron; DAPI, 4=,6-diamidino-2-phenylindole; DIV, day in vitro; DPE, days postexposure; GFAP, glial fibrillary acidic protein; GndSCN, guanidine thiocyanate; MAP2, microtubule-associated protein 2; NeuN, neuronal nuclei; PK, proteinase K; PrP, prion protein; PrPC, cellular prion protein; PrPres, proteinase K-resistant prion protein; PrPSc, scrapie prion protein; ScCAS, scrapie-exposed cerebellar astrocyte; ScCGN, scrapie-exposed CGN; TSE, transmissible spongiform encephalopathy 0892-6638/-1900/0026-0001 © FASEB
Cronier, S., Carimalo, J., Schaeffer, B., Jaumain, E., Béringue, V., Miquel, M.-C., Laude, H., Peyrin, J.-M. Endogenous prion protein conversion is required for prion-induced neuritic alterations and neuronal death. FASEB J. 26, 000 – 000 (2012). www.fasebj.org Key Words: neurodegeneration 䡠 primary culture 䡠 astrocyte 䡠 scrapie Transmissible spongiform encephalopathies (TSEs) include Creutzfeldt–Jakob disease in humans, bovine spongiform encephalopathy, sheep scrapie, and chronic wasting disease in cervids. They feature neuronal loss, spongiosis, and pronounced astrogliosis. These fatal disorders are caused by prions, a class of unconventional agents that preferentially target the central nervous system (CNS). Prions are essentially composed of aggregated misfolded scrapie prion protein (PrPSc) derived from the host-encoded cellular PrP (PrPC). Prion propagation within and between species is believed to stem from the ability of PrPSc seeds to promote the conformational transition from PrPC to PrPSc, through a nucleated polymerization process (for review, see refs. 1, 2). At the molecular level, cell-free assays have suggested 2 sequential steps: initial binding of PrPC to pathological PrP species and conversion into PrPSc (3). Although the essential role of PrPC in prion propagation is beyond any doubt (4, 5), its involvement in neurotoxicity mechanisms remains controversial, and the nature of PrP toxic species themselves are still unknown (6). Exposure of primary neuronal cultures expressing or lacking PrPC to purified PrPSc, aggre1 Correspondence: S.C., INRA, UR892, Virologie et Immunologie Moléculaires, 78352 Jouy-en-Josas, France. E-mail: sabrina.
[email protected]; J.-M.P., Neurobiologie des Processus Adaptatifs, UMR CNRS-7102, Université Pierre et Marie Curie–Paris 6, 75005 Paris, France. E-mail:
[email protected] doi: 10.1096/fj.11-201772
1
gated full-length recombinant PrP or PrP-derived synthetic fragments provided a number of elements toward this end (7– 8). They were sometimes contradictory (9 –11), possibly because supraphysiological doses used in these acute paradigms did not reflect the “slow” natural infection process, or because endogenous PrP conversion is needed to produce toxic species (as we demonstrate here). In vivo, transmission studies in transgenic mice have shown that PrP expression at the neuronal cell surface was necessary to the development of typical TSE neuropathological changes (4, 12–13). However, mice expressing PrP specifically in astrocytes (14) appeared to propagate prion and develop TSEspecific neuronal lesions, suggesting that astrocytic prion replication might be sufficient to induce some neuronal damage (15). Finally, if PrPC plays a critical role in prion-mediated neuronal death following interaction with PrPSc, it is still unclear whether the conversion step is involved. Indeed, PrPC was recently proposed to mediate toxic signaling of -sheet-rich conformers, including amyloid  (16 –17), and following antibody cross-linking (18 – 19), therefore, pointing at a subversion of its putative neuroprotective functions (20), although these observations were not consistently reproduced (21–23). Sheep scrapie prions could efficiently propagate in primary cultures of either neurons or astrocytes and induce a late apoptotic neuronal death (24). Here, we have established a coculture system in which sheep scrapie-infected astrocytes are in contact with neurons devoid of PrPC or expressing homologous (i.e., convertible) or heterologous (i.e., nonconvertible) PrPC species. We report that prion neurotoxicity dramatically depends on expression and efficient conversion of neuronal PrPC into PrPSc. We further show that a dysfunction of prion-infected astrocytes is unlikely to be a major determinant of neuronal death initiation but rather that infected neurons become more susceptible to various subtoxic stimuli, such as oxidation and glutamate.
24-well plates coated with poly-d-lysine (10 g/ml, Sigma; St. Louis, MO, USA). CGN cultures were maintained at 37°C with 6% CO2 in DMEM containing glutamax I (Life Technologies, Paisley, UK), 10% FCS (BioWhittaker, Walkersville, MD, USA), 20 mM KCl, penicillin, streptomycin (Life Technologies) and complemented with N-2 and antioxidant-depleted B27 supplements (Life Technologies, Grand Island, NY, USA). Glucose was maintained at 1 mg/ml by weekly supplementation, along with antimitotics uridine and fluorodeoxyuridine (10 M; Sigma). Pure cerebellar astrocytes from ovine PrPC-expressing tg338 mice (CASOv cells) were dissociated similarly to CGNs, seeded on poly-d-lysine-coated plates (1 g/ml) and cultured in DMEM-glutamax I containing 10% FCS with antibiotics. The cell population was fully astrocytic within a week. Astrocytes were grown to 70% confluence before use, and the medium was changed weekly. Prion infection of cultured cells Infectious 10% (w/v) brain homogenates were prepared in PBS from terminally ill tg338 mice inoculated with the 127S sheep scrapie strain (24, 30). Confluent astrocytes or day in vitro 2 (DIV2) CGNs were exposed to a final concentration of 0.01% (w/v) infected brain homogenate unless specified, or not infected for mock infections. Infectivity of astrocyteconditioned medium was tested by replacing half of the culture medium by medium conditioned for 7 d on 21-d prion-infected astrocytes, supplemented with KCl and N-2. Following infection, the medium was either left unchanged for CGN cultures during the whole duration of the experiment, or changed after 7 d and then 1⫻/wk for CAS cultures. PrP immunoblot Cell lysates or brain homogenates were treated with proteinase K (PK; 7.5 g/mg or 50 g/ml, respectively; Euromedex, Mundolsheim, France), as described previously (29). After methanol precipitation (1 h, ⫺20°C) and centrifugation (16,000 g, 10 min), pellets were resuspended in sample buffer and boiled, and proteins were subjected to SDS/PAGE and electrotransferred onto nitrocellulose membranes. PK-resistant PrP (PrPres) was detected with anti-PrP monoclonal antibodies ICSM18 or biotinylated Sha31. Immunofluorescence
MATERIALS AND METHODS Transgenic mouse lines Care of mice was performed according to Institut National de la Recherche Agronomique and French animal care committee guidelines. Primary cultures were derived from the following homozygous transgenic mouse lines: PrP0/0 (PrP-knockout mice; Zurich I; ref. 25), tg338 (ovine PrPVRQ; ref. 26), tga20 (mouse Prnp-a allele; ref. 27) and tg650 (human Met129 PrP; ref. 28). These PrP-expressing lines were all established on the same Zurich I mouse PrP0/0 background, thus ensuring a consistent genetic background in the cultures. Primary cell cultures Primary cultures of cerebellar granule neurons (CGNs) were established as described previously (24, 29). Briefly, CGNs extracted from cerebellum of 6-d-old mice by enzymatic and mechanical dissociation were plated (400,000 cells/well) in 2
Vol. 26
September 2012
Cells were fixed for 10 min at room temperature in PBS containing 4% paraformaldehyde and 4% sucrose, and subsequently permeabilized for 5 min with PBS containing 0.1% Triton X-100. Following treatment for 5 min with 3 M guanidine thiocyanate (GdnSCN), PrPSc was immunodetected with ICSM35 or ICSM33 anti-PrP monoclonal antibodies (31). Neurons were labeled with monoclonal anti-neuronal nuclei (NeuN; 1:100; Chemicon; Temecula, CA, USA) antibody and astrocytes with polyclonal anti-glial fibrillary acidic protein (GFAP; 1:400; Dako, Glostrup, Denmark) antibody. Neuronal processes were stained with either monoclonal (1:250; Sigma) or polyclonal (1:500; Chemicon) antimicrotubule-associated protein 2 (MAP2) antibodies. Cells were then incubated with appropriate FITC- or Alexa Fluorconjugated secondary antibodies and nuclear marker 4=,6diamidino-2-phenylindole (DAPI; 2 g/ml; Sigma) and mounted in Fluoromount (Sigma). Cells were observed under an Axiovert 200 M epifluorescence or an AxioObserver Z1 microscope (Zeiss, Oberkochen, Germany), images were acquired with Metaview (Universal Imaging, Downingtown, PA, USA) and AxioVision (Zeiss) software, respectively, and
The FASEB Journal 䡠 www.fasebj.org
CRONIER ET AL.
analysis was performed with ImageJ (U.S. National Institutes of Health; http://rsbweb.nih.gov/ij/). Neuron/astrocyte cocultures Mock- or 127S-infected CASOv cultures were maintained for 21 d. Culture medium was then removed, and freshly dissociated CGNs from tg338, tga20, tg650 or PrP0/0 mice suspended in complete neuronal medium were seeded at a density of 2000 cells/mm2 on top of CAS cultures. After 7 d, cells were processed for immunofluorescence. Neuronal survival rate was determined by counting NeuN-positive cells with nonpyknotic nuclei. For each experiment, PrPSc immunostaining was performed to confirm that neurons were in contact with heavily infected astrocytes. Quantification of astrocyte-conditioned medium toxicity Fresh culture medium was conditioned for 7 d on mock- or scrapie-infected astrocytes at 21 days postexposure (DPE) to brain homogenate, collected, and supplemented with KCl and N-2 to obtain neuronal medium. At d 7, 25, 50, or 75% of the culture volume of tg338 CGN (CGNOv) cultures was replaced by astrocyte-conditioned medium. Cells were monitored daily and processed for immunofluorescence after 4 d. Neuronal survival rate was determined by counting NeuNpositive cells with nonpyknotic nuclei. Glutamate and free-radical treatments After 14 d exposure to mock- or 127S-infected tg338 brain homogenate, CGNOv cultures were treated with glutamate (1–100 M) or hydrogen peroxide (1–10 M) (Sigma) for 48 h. Cells were then processed for immunofluorescence. Neuronal survival was estimated by counting NeuN-positive cells with nonpyknotic nuclei, and dendritic area was quantified by measuring MAP2 labeling. Statistical analysis Each data set corresponding to glutamate or hydrogen peroxide treatment was analyzed using the MIXED procedure of SAS 1999 (SAS Institute, Cary, NC, USA). The ANOVA model included a random factor “day” and fixed factors correspond-
ing to the cell state (mock- or scrapie-infected), the treatment dose, and their interaction. To satisfy homoscedasticity assumptions, logarithm of neuronal survival and square root of dendritic area were considered. LS means were computed, and all pair-wise differences were evaluated using TukeyKramer adjustment for P values. For the other sets of experiments, data were analyzed by Student’s t test (2-tailed distribution; 2-sample equal variance).
RESULTS Neurons cocultured with prion-infected astrocytes rapidly degenerate First, we investigated the contribution of prion-infected astrocytes in neurodegeneration. Exposure of CASOv primary cultures to brain homogenate infected with 127S sheep scrapie strain led to the detection of neosynthesized PrPres from 7 DPE onward by immunoblot (Fig. 1A). PrPres amounts then increased 4-fold by 28 DPE. These data were consistent with those obtained using cell culture medium instead of brain homogenate as an infectious source (24). At 21 DPE, a large majority (⬃80%) of astrocytes exposed to 127S scrapie inoculum (ScCASOv) showed a GdnSCN-dependent, bright punctated PrP immunostaining specific for prion infection (Fig. 1B, C), therefore confirming that astrocyte cultures were heavily infected and laden with PrPSc. At this time point, we seeded freshly dissociated CGNs expressing ovine PrPC (CGNOv) on ScCASOv cells or on mock-infected CASOv cells in a medium depleted of antioxidants. After 7 d of CGNOv/ScCASOv coculture, neuronal survival significantly decreased by ⬃40% as compared to CGNOv/CASOv cocultures (n⫽4 independent experiments, Fig. 2A). Colocalization of terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL)-positive cells and pyknotic nuclei indicated that apoptosis was a prominent feature in dying neurons (data not shown). Surviving neurons showed massive dendritic fragmentation (Fig. 2D, top
Figure 1. Efficient prion infection of cerebellar astrocyte cultures. Cerebellar astrocyte cultures from tg338 mice (CASOv) were exposed to 127S scrapie-infected tg338 brain homogenate. A) Immunoblot of PK-treated lysates probed with anti-PrP mAb ICSM18, showing PrPres accumulation between 7 and 28 DPE in prion-infected CASOv cultures (ScCASOv). As a control, nonpermissive CAS0/0 cells were exposed in parallel (ScCAS0/0). B–D) CASOv cells were either mock-infected (B) or prion-infected (scrapie; C, D), and immunofluorescence was performed 3 wk after infection. PrPSc was detected with mAb ICSM33 (green) after permeabilization (B, C) or not (D) and GdnSCN denaturation. Red, GFAP; blue, DAPI. In infected cultures, ⬃80% of cells show a punctate fluorescence assumed to reflect PrPSc accumulation. Scale bars ⫽ 10 m. PRION NEURODEGENERATION REQUIRES PRP CONVERSION
3
To examine whether a possible alteration of such functions by astrocyte prion replication could contribute to the neuronal death observed in our 7-d coculture, the culture medium from mock- or prion-infected astrocytes was conditioned between 21 and 28 DPE and subsequently added to differentiated CGNOv cells. No significant difference in neuronal survival was observed after up to 4 d of exposure (Fig. 3A). This suggested
Figure 2. Influence of neuronal PrPC expression and primary sequence in prion-induced neuronal death. Freshly dissociated neurons expressing ovine (CGNOv), murine (CGNMo), or human PrPC (CGNHu), or devoid of PrP (CGN0/0), were seeded on top of mock- or sheep scrapie-infected astrocytes (CASOv or ScCASOv, respectively) and cocultured for a week. A, B) Neuronal survival is expressed as a percentage of total living neurons (NeuN-positive cells) in mock-infected cocultures (control). Values are means ⫾ se; n ⫽ 4 (A); n⫽3 (B). In infected cocultures, survival significantly decreased for CGNOv, but not for CGN0/0, CGNMo, and CGNHu (N.S., not significant). **P ⬍ 0.01; ***P ⬍ 0.001; Student’s t test. C) PrPSc was quantified by immunofluorescence in CGNOv, CGN0/0, and CGNMo cells cocultured with ScCASOv cells using mAb ICSM33. For each experiment, PrPSc levels in CGN0/0/ScCASOv cultures were normalized to 1. Mean ⫾ se (n⫽4). D) CGNOv or CGN0/0 cells were cocultured with CASOv or ScCASOv cells and stained with anti-MAP2 mAb (dendrites, green) and DAPI (blue). Note the intense dendritic fragmentation in CGNOv cells cocultured with prioninfected astrocytes (arrows). Scale bars ⫽ 10 m.
panels), indicating widespread ongoing neurodegeneration. Conditioned medium of prion-infected astrocytes is not neurotoxic and is poorly infectious In other proteinopathies, it has been shown that astrocytes could contribute to neurodegeneration by releasing toxic factors (32–34). Astrocytes provide trophic support to neurons and regulate neurotransmitters and toxin concentrations in the extracellular environment. 4
Vol. 26
September 2012
Figure 3. Conditioned medium of prion-infected astrocytes is not toxic to neurons and contains little infectivity. A) Differentiated CGNOv cells were cultured for 4 d with 25% (open bars), 50% (shaded bars), or 75% (solid bars) of mockinfected (CASOv) or prion-infected (ScCASOv) astrocyte-conditioned medium. Neuronal survival is expressed as a percentage of total living neurons (NeuN-positive cells) in nontreated cultures. Values are means ⫾ se (n⫽3). B) PrPres accumulation in CGNOv cells exposed to a serial dilution of scrapie-infected tg338 brain homogenate or to ScCASOvconditioned medium for 28 d. As a control, CGN0/0 cells were exposed to conditioned medium. C) Comparison of PrPres amounts between CASOv cells exposed or not to scrapieinfected tg338 brain homogenate (0.01% final concentration) at 21 DPE, and the inoculum used. For quantification purposes, a serial dilution of the inoculum, ranging from 0.5to 2-fold the amount of brain homogenate inoculated per well, was loaded along with lysates of whole CASOv culture wells. B, C) Immunoblots were probed with biotinylated mAb Sha31.
The FASEB Journal 䡠 www.fasebj.org
CRONIER ET AL.
that ScCASOv culture medium does not contain neurotoxins, including possibly neurotoxic PrPSc species. Relative efficacy of infection of CGNOv cells with ScCASOvconditioned medium in comparison with serial dilutions of infectious brain homogenate indicated a low infectivity titer equivalent to ⬍0.0001% brain homogenate or ⬃103 ID50/ml (Fig. 3B). In contrast, after 21 DPE, astrocyte cell fraction had an infectivity titer ⬃100-fold higher (data not shown), and a PrPres content equivalent to 0.01% brain homogenate (Fig. 3C). Immunostaining of nonpermeabilized GndSCN-treated ScCASOv cells showed intense PrPSc labeling, presumably at the cell surface (Fig. 1D). Together, these results indicate that prion-infected astrocytes do not exhibit major dysfunctions and rather suggest that, given the low infectivity level of the conditioned medium, the abundant cell-associated PrPSc could be responsible for neuronal death through direct astrocyte contact in cocultures.
molecules that are efficiently converted by nearby PrPSc into new PrPSc species. Prion infection of neurons increases their sensitivity to glutamate and free radical insults Finally, we investigated whether neuronal neosynthesized PrPSc alone was responsible for CGNOv cell death observed in cocultures with ScCASOv cells. Infection of “pure” CGNOv cells with 127S-infected brain homogenate led to specific accumulation of PrPres as early as 7 DPE. Amounts steadily increased 5-fold by 28 DPE (Fig. 4A). By 14 DPE, ScCGNOv cells accumulated substantial levels of PrPSc (Fig. 4A), distributed in the soma and processes (24). However, at this stage, neuronal survival was unaltered (Fig. 4B, see dose 0), and a
Expression and efficient conversion of neuronal PrPC are required for neurotoxicity mediated by prion-infected astrocytes To assess whether the rapid death observed in neurons cocultured with prion-infected astrocytes was PrP mediated, neurons derived from mice with identical genetic background but devoid of PrPC (CGN0/0) were cocultured for 1 wk with either CASOv or ScCASOv cells in the same culture conditions as CGNOv cells. No alteration in neuronal survival was observed in prioninfected cocultures (Fig. 2B), in striking contrast with the CGNOv situation (Fig. 2A). Moreover, CGN0/0 neurite immunostaining failed to reveal any ongoing neurodegeneration (Fig. 2D, bottom panels). As neuronal PrPC expression appeared necessary for neuronal cell death in our cocultures, we next examined whether its efficient conversion into PrPSc was equally crucial. Thus, mock- or scrapie-infected CASOv cells were cocultured with CGNs expressing either mouse (CGNMo) or human (CGNHu) PrPC, i.e., heterologous PrP species assumed to be inefficiently converted by ovine PrPSc (35). Indeed, in vivo, 127S prions exhibit a substantial transmission barrier on transmission to human and mouse PrP transgenic mice (unpublished results). As a result, neuronal survival slightly decreased, but not to statistically significant levels, when CGNHu (P⫽0.29) or CGNMo (P⫽0.10) cells were cocultured with ScCASOv cells for 7 d (n⫽3 independent experiments; Fig. 2B). Abnormal PrP levels in the cocultures did not vary significantly whether ScCASOv cells were cocultured with CGN0/0 or CGNMo cells, as assessed by immunostaining and immunoblot (Fig. 2C and data not shown). Remarkably, and in contrast, PrPSc levels were more than doubled on coculture with CGNOv cells, suggesting active prion replication in these cells (Fig. 2C). Altogether, our coculture system shows that neurons apposed to PrPSc-producing astrocytes undergo a rapid and significant death only when they express PrPC PRION NEURODEGENERATION REQUIRES PRP CONVERSION
Figure 4. Prion-infected neurons display an increased sensitivity to glutamate and H2O2. A) Immunoblot probed with biotinylated mAb Sha31 showing PrPres accumulation kinetics in CGNOv cells exposed to 127S scrapie-infected tg338 brain homogenate along with CGN0/0 (ScCGN0/0). B) CGNOv cells were either mock-treated (open bars) or prion-infected (solid bars). At 14 DPE, cells were exposed to increasing concentrations of glutamate or hydrogen peroxide (H2O2) for 48 h. Neuronal survival is expressed as a percentage of total living neurons in nontreated cultures. Dendritic area represents the extent of MAP2 labeling, expressed in pixels. Quantifications correspond to means ⫾ se of 3 wells in 1 experiment and are representative of n ⫽ 3– 4 independent experiments. Statistical analysis was performed using MIXED procedure in SAS (see Materials and Methods). **P ⬍ 0.01; ***P ⬍ 0.001. 5
significant increase in neuronal death (⬃40%) actually occurred at 21 DPE (data not shown). Thus, the presence of nearby astrocytes seemed to accelerate neurodegeneration in CGNOv/ScCASOv cocultures. Because astrocytes are assumed to regulate extracellular signals, such as reactive oxygen species and glutamate, we next subjected 14-d CGNOv or ScCGNOv cultures to subtoxic concentrations of glutamate or hydrogen peroxide for 48 h (Fig. 4B). Prion infection significantly decreased both viability and dendritic area when neurons were exposed to 100 M glutamate (P⬍0.001). It also significantly decreased dendritic area following treatment with 10 M H2O2 (P⬍0.01), indicating that ScCGNOv neurons underwent a degenerative process. Altogether, these data suggest that PrPSc accumulation in neurons increases their sensitivity to oxidative stress and glutamate.
DISCUSSION Our study aimed at deciphering the relative importance of infected astrocytes and neuronal PrPC expression and conversion in prion-mediated neurodegeneration, by exposing primary differentiated neurons expressing or lacking PrPC to a continuous and physiological source of PrPSc delivered by neighboring prion-infected astrocytes. As a main result, we found that neuronal PrPC must not only be present but also be efficiently converted into PrPSc for neurotoxicity to occur, therefore further extending at the cell level earlier in vivo observations (4, 13) with another prion model. We also provide evidence that prion-infected astrocytes are not detrimental per se but that they accelerate neurodegeneration. No significant increase in neuronal cell death was observed on coculture of ovine PrPSc-producing astrocytes with neurons genetically devoid of PrPC or expressing mouse or human PrPC. In striking contrast, when neurons expressed ovine PrPC, their survival was decreased by ⬃40% within a week, the surviving cells showing widespread ongoing neurodegeneration concomitantly with detectable PrPSc accumulation. As differences in primary sequence appear to have little effect on PrPC-PrPSc binding (36, 37), it is likely that ovine PrPSc bound similarly to sheep, mouse, and human PrPC, thus dismissing a subversion of PrPC neuroprotective function by infecting PrPSc as the major cause of the prominent neuronal death observed here (38). Instead, the delayed or absence of conversion of mouse and human PrPC by 127S prions in transgenic mice and the significant PrPSc increase observed only in homologous cocultures strongly support the view that the conversion step is critical in the initiation of neurodegeneration. This sharply contrasts with the emerging notion that PrP could indifferently convey toxicity of prions or other abnormally folded proteic conformers, such as those involved in Alzheimer’s disease. Notably, our results are in apparent contradiction with the recent findings of Resenberger 6
Vol. 26
September 2012
et al. (17), in which short-term (16 h) coculture with minimal cell contact between chronically prion-infected mouse neuroblastoma (ScN2a) cells and immortalized SH-SY5Y cells transiently expressing various heterologous PrP induced an increased apoptosis rate in the latter, independent of PrP sequence and with no apparent conversion. However, in these conditions, the apoptosis increase appeared marginal (⬃5–10%), and when they subjected mouse primary cortical neurons to a 5-d coculture with ScN2a cells, a 40% viability decrease was observed, similar to our findings, except that bona fide prion replication was not assessed, thus reconciling their data with our observation that PrP conversion and accumulation are the major contributors to neuronal death. Neurons devoid of PrPC have been reported to exhibit an increased sensitivity to neurotoxins (39). The absence of neurodegeneration when they were cocultured with heavily infected astrocytes indicates that the latter did not misfunction in a way that would affect cocultured neurons, nor produce labile non-PrPrelated neurotoxins that could have been diluted or lacking in astrocyte-conditioned medium. Moreover, this indicates that extraneuronal, infectious PrPSc particles, either released (at low levels) by astrocytes in the extracellular medium or membrane bound, were not significantly neurotoxic per se. Intriguingly, however, apoptosis and neuritic alterations appeared considerably earlier in ovine PrPC-expressing neurons infected on coculture with prion-infected astrocytes than by scrapie inoculum [respectively, at 7 d (this study) vs. 28 d (24) and 21 d when using antioxidant-depleted medium (unpublished results)]. As infectivity and PrPres levels of both infectious sources seemed comparable, this indicates that astrocyte-mediated infection was able to potentiate neurodegeneration. Neuron/ glia close contacts along with nearly maximal PrPSc levels—presumably at the astrocyte cell surface—are likely to favor ignition of prion infection in neurons through efficient cell-to-cell transfer of PrPSc particles, as previously shown in immortalized cell models (31). This is further supported by findings that prion infection of cells occurs very rapidly and that cell membranes probably constitute the primary site of prion conversion (40). Whether PrPSc produced by long-term infected astrocytes not only promotes conversion but also catalyzes the formation of neurotoxic species from neuronal PrPC, as recently suggested in mice (41), will be examined in the future. Interestingly, we show here that PrPSc-accumulating neurons are more sensitive to exogenous stimuli, such as glutamate and hydrogen peroxide, suggesting an altered response to excitotoxicity and oxidative stress, as reported with other misfolded proteins and peptides (42– 44). Although our results suggest that infected astrocytes do not release toxic species, we cannot exclude that they might no longer help neurons cope with stresses in their environment, including intracellular protein aggregates. In other neurodegenerative diseases, accumulation of misfolded protein aggregates
The FASEB Journal 䡠 www.fasebj.org
CRONIER ET AL.
and/or changes in lipid raft content or interactions have both been shown to be detrimental to neurons, by inducing massive dendritic degeneration and axonal damage due to microtubule disruption (45– 46). Whether similar alterations are responsible for the toxicity observed in our study remains to be determined. Overall the massive neurodegeneration observed in our cocultures would, therefore, result from both a particularly efficient initiation of prion infection in neurons by infected astrocytes and a subsequent increased neuronal vulnerability. In neurons, the cellular form of the prion protein PrPC appears to be involved in a number of possibly independent neurodegenerative pathways, some being acute following exposure to extraneuronal PrPSc or other -sheet-rich conformers (17, 47) and some necessitating efficient PrPC conversion, acting putatively on a downstream cascade or physiological process (48). In this regard, our ex vivo model may help to further dissect these neurodegenerative mechanisms that could be relevant to other brain disorders and development of rational therapies. This project was supported by grants from the French government (GIS-Infections a` Prion) and from the European Union (Neurodegeneration–QLG3CT2001). S.C. was a recipient of a French Ministry of Research and Education (MRE) fellowship. J.C. was funded by an MRE fellowship and the French Foundation France Alzheimer. The authors thank S. Hawke (Imperial College, London, UK; now at University of Sydney, Sydney, NSW, Australia) and G. S. Jackson (Medical Research Council Prion Unit, London, UK) for kindly providing the antibodies ICSM18 and ICSM35, and ICSM33, respectively; J. Grassi and S. Simon (Commissariat a` l’énergie Atomique et aux énergies Alternatives, Saclay, France) for Sha31 antibody; C. Weissmann (Scripps Research Institute, Jupiter, FL, USA) for authorizing the inclusion of tga20 and PrP0/0 mice in this study. The authors thank T. Szeto for careful reading of the manuscript and R. Young for preparation of the figures.
9. 10. 11.
12.
13.
14.
15. 16.
17.
18.
19.
20.
REFERENCES 1. 2. 3. 4.
5. 6.
7. 8.
Collinge, J. (2001) Prion diseases of humans and animals: their causes and molecular basis. Annu. Rev. Neurosci. 24, 519 –550 Prusiner, S. B. (1998) Prions. Proc. Natl. Acad. Sci. U. S. A. 95, 13363–13383 Silveira, J. R., Caughey, B., and Baron, G. S. (2004) Prion protein and the molecular features of transmissible spongiform encephalopathy agents. Curr. Top. Microbiol. Immunol. 284, 1–50 Brandner, S., Isenmann, S., Raeber, A., Fischer, M., Sailer, A., Kobayashi, Y., Marino, S., Weissmann, C., and Aguzzi, A. (1996) Normal host prion protein necessary for scrapie-induced neurotoxicity. Nature 379, 339 –343 Bueler, H., Aguzzi, A., Sailer, A., Greiner, R. A., Autenried, P., Aguet, M., and Weissmann, C. (1993) Mice devoid of PrP are resistant to scrapie. Cell 73, 1339 –1347 Caughey, B., and Lansbury, P. T. (2003) Protofibrils, pores, fibrils, and neurodegeneration: separating the responsible protein aggregates from the innocent bystanders. Annu. Rev. Neurosci. 26, 267–298 Brown, D. R. (2002) Mayhem of the multiple mechanisms: modelling neurodegeneration in prion disease. J. Neurochem. 82, 209 –215 Gavin, R., Braun, N., Nicolas, O., Parra, B., Urena, J. M., Mingorance, A., Soriano, E., Torres, J. M., Aguzzi, A., and del
PRION NEURODEGENERATION REQUIRES PRP CONVERSION
21.
22.
23.
24. 25.
26.
Rio, J. A. (2005) PrP(106-126) activates neuronal intracellular kinases and Egr1 synthesis through activation of NADPH-oxidase independently of PrPc. FEBS Lett. 579, 4099 –4106 Brown, D. R., Herms, J., and Kretzschmar, H. A. (1994) Mouse cortical cells lacking cellular PrP survive in culture with a neurotoxic PrP fragment. Neuroreport 5, 2057–2060 Kunz, B., Sandmeier, E., and Christen, P. (1999) Neurotoxicity of prion peptide 106-126 not confirmed. FEBS Lett. 458, 65–68 Simoneau, S., Rezaei, H., Sales, N., Kaiser-Schulz, G., LefebvreRoque, M., Vidal, C., Fournier, J. G., Comte, J., Wopfner, F., Grosclaude, J., Schatzl, H., and Lasmezas, C. I. (2007) In vitro and in vivo neurotoxicity of prion protein oligomers. PLoS Pathog. 3, e125 Chesebro, B., Trifilo, M., Race, R., Meade-White, K., Teng, C., LaCasse, R., Raymond, L., Favara, C., Baron, G., Priola, S., Caughey, B., Masliah, E., and Oldstone, M. (2005) Anchorless prion protein results in infectious amyloid disease without clinical scrapie. Science 308, 1435–1439 Mallucci, G., Dickinson, A., Linehan, J., Klohn, P. C., Brandner, S., and Collinge, J. (2003) Depleting neuronal PrP in prion infection prevents disease and reverses spongiosis. Science 302, 871–874 Raeber, A. J., Race, R. E., Brandner, S., Priola, S. A., Sailer, A., Bessen, R. A., Mucke, L., Manson, J., Aguzzi, A., Oldstone, M. B., Weissmann, C., and Chesebro, B. (1997) Astrocyte-specific expression of hamster prion protein (PrP) renders PrP knockout mice susceptible to hamster scrapie. EMBO J. 16, 6057–6065 Jeffrey, M., Goodsir, C. M., Race, R. E., and Chesebro, B. (2004) Scrapie-specific neuronal lesions are independent of neuronal PrP expression. Ann. Neurol. 55, 781–792 Lauren, J., Gimbel, D. A., Nygaard, H. B., Gilbert, J. W., and Strittmatter, S. M. (2009) Cellular prion protein mediates impairment of synaptic plasticity by amyloid-beta oligomers. Nature 457, 1128 –1132 Resenberger, U. K., Harmeier, A., Woerner, A. C., Goodman, J. L., Muller, V., Krishnan, R., Vabulas, R. M., Kretzschmar, H. A., Lindquist, S., Hartl, F. U., Multhaup, G., Winklhofer, K. F., and Tatzelt, J. (2011) The cellular prion protein mediates neurotoxic signalling of beta-sheet-rich conformers independent of prion replication. EMBO J. 30, 2057–2070 Lefebvre-Roque, M., Kremmer, E., Gilch, S., Zou, W. Q., Feraudet, C., Gilles, C. M., Sales, N., Grassi, J., Gambetti, P., Baron, T., Schatzl, H., and Lasmezas, C. I. (2007) Toxic effects of intracerebral PrP antibody administration during the course of BSE infection in mice. Prion 1, 198 –206 Solforosi, L., Criado, J. R., McGavern, D. B., Wirz, S., SanchezAlavez, M., Sugama, S., DeGiorgio, L. A., Volpe, B. T., Wiseman, E., Abalos, G., Masliah, E., Gilden, D., Oldstone, M. B., Conti, B., and Williamson, R. A. (2004) Cross-linking cellular prion protein triggers neuronal apoptosis in vivo. Science 303, 1514 –1516 Schneider, B., Pietri, M., Pradines, E., Loubet, D., Launay, J. M., Kellermann, O., and Mouillet-Richard, S. (2011) Understanding the neurospecificity of Prion protein signaling. Front. Biosci. 16, 169 –186 Balducci, C., Beeg, M., Stravalaci, M., Bastone, A., Sclip, A., Biasini, E., Tapella, L., Colombo, L., Manzoni, C., Borsello, T., Chiesa, R., Gobbi, M., Salmona, M., and Forloni, G. (2010) Synthetic amyloid-beta oligomers impair long-term memory independently of cellular prion protein. Proc. Natl. Acad. Sci. U. S. A. 107, 2295–2300 Calella, A. M., Farinelli, M., Nuvolone, M., Mirante, O., Moos, R., Falsig, J., Mansuy, I. M., and Aguzzi, A. (2010) Prion protein and Abeta-related synaptic toxicity impairment. EMBO Mol. Med. 2, 306 –314 Klohn, P. C., Farmer, M., Linehan, J. M., O’Malley, C., Fernandez de Marco, M., Taylor, W., Farrow, M., Khalili-Shirazi, A., Brandner, S., and Collinge, J. (2012) PrP antibodies do not trigger mouse hippocampal neuron apoptosis. Science 335, 52 Cronier, S., Laude, H., and Peyrin, J. M. (2004) Prions can infect primary cultured neurons and astrocytes and promote neuronal cell death. Proc. Natl. Acad. Sci. U. S. A. 101, 12271–12276 Bueler, H., Fischer, M., Lang, Y., Bluethmann, H., Lipp, H. P., DeArmond, S. J., Prusiner, S. B., Aguet, M., and Weissmann, C. (1992) Normal development and behaviour of mice lacking the neuronal cell-surface PrP protein. Nature 356, 577–582 Laude, H., Vilette, D., Le Dur, A., Archer, F., Soulier, S., Besnard, N., Essalmani, R., and Vilotte, J. L. (2002) New in vivo
7
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
8
and ex vivo models for the experimental study of sheep scrapie: development and perspectives. C. R. Biol. 325, 49 –57 Fischer, M., Rulicke, T., Raeber, A., Sailer, A., Moser, M., Oesch, B., Brandner, S., Aguzzi, A., and Weissmann, C. (1996) Prion protein (PrP) with amino-proximal deletions restoring susceptibility of PrP knockout mice to scrapie. EMBO J. 15, 1255–1264 Beringue, V., Le Dur, A., Tixador, P., Reine, F., Lepourry, L., Perret-Liaudet, A., Haik, S., Vilotte, J. L., Fontes, M., and Laude, H. (2008) Prominent and persistent extraneural infection in human PrP transgenic mice infected with variant CJD. PLoS One 3, e1419 Cronier, S., Beringue, V., Bellon, A., Peyrin, J. M., and Laude, H. (2007) Prion strain- and species-dependent effects of antiprion molecules in primary neuronal cultures. J. Virol. 81, 13794 – 13800 Langevin, C., Andreoletti, O., Le Dur, A., Laude, H., and Beringue, V. (2011) Marked influence of the route of infection on prion strain apparent phenotype in a scrapie transgenic mouse model. Neurobiol. Dis. 41, 219 –225 Paquet, S., Langevin, C., Chapuis, J., Jackson, G. S., Laude, H., and Vilette, D. (2007) Efficient dissemination of prions through preferential transmission to nearby cells. J. Gen. Virol. 88, 706 –713 Custer, S. K., Garden, G. A., Gill, N., Rueb, U., Libby, R. T., Schultz, C., Guyenet, S. J., Deller, T., Westrum, L. E., Sopher, B. L., and La Spada, A. R. (2006) Bergmann glia expression of polyglutamine-expanded ataxin-7 produces neurodegeneration by impairing glutamate transport. Nat. Neurosci. 9, 1302–1311 Haidet-Phillips, A. M., Hester, M. E., Miranda, C. J., Meyer, K., Braun, L., Frakes, A., Song, S., Likhite, S., Murtha, M. J., Foust, K. D., Rao, M., Eagle, A., Kammesheidt, A., Christensen, A., Mendell, J. R., Burghes, A. H., and Kaspar, B. K. (2011) Astrocytes from familial and sporadic ALS patients are toxic to motor neurons. Nat. Biotechnol. 29, 824 –828 Nagai, M., Re, D. B., Nagata, T., Chalazonitis, A., Jessell, T. M., Wichterle, H., and Przedborski, S. (2007) Astrocytes expressing ALS-linked mutated SOD1 release factors selectively toxic to motor neurons. Nat. Neurosci. 10, 615–622 Raymond, G. J., Hope, J., Kocisko, D. A., Priola, S. A., Raymond, L. D., Bossers, A., Ironside, J., Will, R. G., Chen, S. G., Petersen, R. B., Gambetti, P., Rubenstein, R., Smits, M. A., Lansbury, P. T., Jr., and Caughey, B. (1997) Molecular assessment of the potential transmissibilities of BSE and scrapie to humans. Nature 388, 285–288 Horiuchi, M., Priola, S. A., Chabry, J., and Caughey, B. (2000) Interactions between heterologous forms of prion protein:
Vol. 26
September 2012
37.
38. 39. 40.
41. 42.
43. 44.
45.
46. 47. 48.
binding, inhibition of conversion, and species barriers. Proc. Natl. Acad. Sci. U. S. A. 97, 5836 –5841 Priola, S. A., Caughey, B., Race, R. E., and Chesebro, B. (1994) Heterologous PrP molecules interfere with accumulation of protease-resistant PrP in scrapie-infected murine neuroblastoma cells. J. Virol. 68, 4873–4878 Westergard, L., Christensen, H. M., and Harris, D. A. (2007) The cellular prion protein (PrP(C)): its physiological function and role in disease. Biochim. Biophys. Acta 1772, 629 –644 Sakudo, A., and Ikuta, K. (2009) Prion protein functions and dysfunction in prion diseases. Curr. Med. Chem. 16, 380 –389 Goold, R., Rabbanian, S., Sutton, L., Andre, R., Arora, P., Moonga, J., Clarke, A. R., Schiavo, G., Jat, P., Collinge, J., and Tabrizi, S. J. (2011) Rapid cell-surface prion protein conversion revealed using a novel cell system. Nat. Commun. 2, 281 Sandberg, M. K., Al-Doujaily, H., Sharps, B., Clarke, A. R., and Collinge, J. (2011) Prion propagation and toxicity in vivo occur in two distinct mechanistic phases. Nature 470, 540 –542 Dong, X. X., Wang, Y., and Qin, Z. H. (2009) Molecular mechanisms of excitotoxicity and their relevance to pathogenesis of neurodegenerative diseases. Acta Pharmacol. Sin. 30, 379 –387 Haass, C., and Selkoe, D. J. (2007) Soluble protein oligomers in neurodegeneration: lessons from the Alzheimer’s amyloid betapeptide. Nat. Rev. Mol. Cell Biol. 8, 101–112 Milhavet, O., McMahon, H. E., Rachidi, W., Nishida, N., Katamine, S., Mange, A., Arlotto, M., Casanova, D., Riondel, J., Favier, A., and Lehmann, S. (2000) Prion infection impairs the cellular response to oxidative stress. Proc. Natl. Acad. Sci. U. S. A. 97, 13937–13942 Jin, M., Shepardson, N., Yang, T., Chen, G., Walsh, D., and Selkoe, D. J. (2011) Soluble amyloid -protein dimers isolated from Alzheimer cortex directly induce Tau hyperphosphorylation and neuritic degeneration. Proc. Natl. Acad. Sci. U. S. A. 108, 5819 –5824 Schengrund, C. L. (2010) Lipid rafts: keys to neurodegeneration. Brain Res. Bull. 82, 7–17 Nygaard, H. B., and Strittmatter, S. M. (2009) Cellular prion protein mediates the toxicity of beta-amyloid oligomers: implications for Alzheimer disease. Arch. Neurol. 66, 1325–1328 Kristiansen, M., Deriziotis, P., Dimcheff, D. E., Jackson, G. S., Ovaa, H., Naumann, H., Clarke, A. R., van Leeuwen, F. W., Menendez-Benito, V., Dantuma, N. P., Portis, J. L., Collinge, J., and Tabrizi, S. J. (2007) Disease-associated prion protein oligomers inhibit the 26S proteasome. Mol. Cell 26, 175–188
The FASEB Journal 䡠 www.fasebj.org
Received for publication February 28, 2012. Accepted for publication May 21, 2012.
CRONIER ET AL.
