Correspondence Ken'ichi Hagiwara, Department of Biochemistry and Cell Biology, National Institute of Infectious Diseases, 1-23-1 Toyama, Shinjuku-ku, Tokyo 162-8640, Japan. Tel: +81 3 5285 1111; fax: +81 3 5285 1157; email: firstname.lastname@example.org
* Present address National Institute of Health Sciences, 1-18-1 Kamiyoga, Setagaya-ku, Tokyo 158-8501, Japan.
Transmissible spongiform encephalopathies are infectious and neurodegenerative disorders that cause neural deposition of aggregates of the disease-associated form of PrPSc. PrPSc reproduces by recruiting and converting the cellular PrPC, and ScN2a cells support PrPSc propagation. We found that incubation of ScN2a cells with a fibril peptide named P9, which comprises an intrinsic sequence of residues 167–184 of mouse PrPC, significantly reduced the amount of PrPSc in 24 hr. P9 did not affect the rates of synthesis and degradation of PrPC. Interestingly, immunofluorescence analysis showed that the incubation of ScN2a cells with P9 induced colocalization of the accumulation of PrP with cathepsin D-positive compartments, whereas the accumulation of PrP in the cells without P9 colocalized mainly with lysosomal associated membrane proteins (LAMP)-1-positive compartments but rarely with cathepsin D-positive compartments in perinuclear regions. Lysosomal enzyme inhibitors attenuated the anti-PrPSc activity; however, a proteasome inhibitor did not impair P9 activity. In addition, P9 neither promoted the ubiquitination of cellular proteins nor caused the accumulation of LC3-II, a biochemical marker of autophagy. These results indicate that P9 promotes PrPSc redistribution from late endosomes to lysosomes, thereby attaining PrPSc degradation.
Transmissible spongiform encephalopathies (or prion diseases) are fatal neurodegenerative disorders which are usually accompanied by vacuolation and accumulation of the disease-associated form of the prion protein (PrPSc) in the brain (1). PrPSc is identical to the normal cellular form of the prion protein (PrPC) with respect to its amino acid sequence, but differs in its conformation (2). According to the ‘protein-only’ hypothesis (1, 3), a major component of the infectious agents associated with prion diseases is PrPSc (4). Although PrPC folds into three α-helices and a set of anti-parallel β-sheets (Fig. 1) (5, 6), infrared spectroscopy shows that PrPSc is rich in β-sheets (7). These physicochemical properties cause the formation of amyloid aggregates of PrPSc, and confer partial resistance to proteolytic digestion by PK (1). PrPC distributes in the lipid raft domains of the plasma membrane as a glycophosphatidyl inositol-anchored protein, and it is believed that the direct interaction of PrPC with PrPSc in lipid raft domains and/or early endosomes triggers the conformational conversion of PrPC to PrPSc (8). Similar to other glycophosphatidyl inositol-anchored proteins, PrPC in the lipid raft domains is internalized by endocytosis, delivered to late endosomes, and rapidly degraded (9). Kinetic analyses have shown that the turnover of PrPC takes 4 to 6 hr (10). In contrast, PrPSc, which is converted from PrPC along the endocytic pathway from the cell surface, accumulates in late endosome-like organelles (11–13), and has been reported to have a half-life of more than 24 hr (14).
To date, a variety of approaches preventing the conversion of PrPC to PrPSc have been proposed. For example, some polyanions and anti-PrPC monoclonal antibodies were reported to inhibit the formation of PrPSc by binding to PrPC, and consequently interfere with the interaction between PrPC and PrPSc (15–17). The pivotal importance of the direct interaction between PrPC and PrPSc was also indicated in cell-free experimental settings, in which some peptides derived from PrPC prevented the conversion of PrPC to PrPSc possibly through competitive interference with their interaction (18). In addition, the in vitro incubation of PrPSc with iPrP13, a short peptide possessing a conserved sequence motif of PrPC with artificial insertions of proline residues, rendered PrPSc sensitive to PK-digestion (19). The peptide iPrP13 also inhibited the propagation of PrPSc in Chinese hamster ovary cells that expressed mPrP with additional octapeptide repeats (19), although iPrP13 or its dendrimers were ineffective in ScN2a cells (20).
Elimination of PrPSc could be accomplished not only by inhibiting the conversion of PrPC to PrPSc but also by enhancing the degradation of PrPSc which accumulates in late endosome-like organelles in prion-infected cells (11–13). However, few attempts have been made to activate the mechanisms responsible for the inherent and endogenous degradation of PrPSc, although a previous study implied that lysosomal proteases would, to some extent, degrade PrPSc in prion-infected cells in culture (21). A recent study showing that an inhibitor of tyrosine kinases enhanced PrPSc lysosomal degradation (22) is intriguing in this regard, although precisely how this inhibitor causes PrPSc lysosomal degradation remains unclear.
In the present study, we screened PrPC-derived peptides for the ability to inhibit PrPSc propagation in living cells. We found that an intrinsic sequence of PrPC bears a fibril peptide that does not act as an amyloid seed but rather eliminates pre-existing PrPSc in prion-infected neuroblastoma cells. We further show that the clearance of pre-existing PrPSc by this peptide can be ascribed to the promotion of endosomal-lysosomal degradation.
MATERIALS AND METHODS
Mouse neuroblastoma N2a cells were purchased from American Type Culture Collection (ATCC: CCL131). ScN2a cells were established as described previously (23). N2a and ScN2a cells were maintained in a normal growth medium, DMEM, containing 4 mM l-glutamine, 50 U/mL penicillin G, 50 μg/mL streptomycin, and 10% heat-inactivated fetal bovine serum (FBS), at 37 °C in 5% CO2/95% air. All procedures were carried out according to the biosafety guidelines of the National Institute of Infectious Diseases.
Fifteen peptides derived from the amino acid sequences of bovine PrP (bPrP) or mPrP were synthesized by Fmoc solid-phase chemistry (Operon Biotechnologies, Tokyo, Japan). The amino acid sequences of these peptides are summarized in Table 1 and Figure 1. Purity of the peptides was verified by reversed-phase high-performance liquid chromatography and matrix-assisted laser desorption time-of-flight mass spectrometry (purity >90%). Stock solutions of peptides were prepared in distilled water at a final concentration of 5 mM. Aliquots of the stock solutions were sonicated briefly using a water bath-type sonicator, and stored at −80 °C until use. The stock solutions of the peptides were added to the culture medium immediately after thawing.
Table 1. Sequences of synthetic peptides derived from the amino acid sequences of bovine PrP or mPrP
Amino acid residues†
†Residues are numbered in relation to mPrP.
Sample preparation, PK digestion, and western blot analysis
The cells were lysed in T-DOC buffer (0.5% Triton X-100, 0.5% sodium deoxycholate, 20 mM Tris-HCl, pH 7.4, and 100 mM NaCl) on ice for 10 min. After centrifugation of the lysate at 1000 ×g for 5 min, the supernatant was collected as PNF. The protein concentration of PNF was determined by the bicinchoninic acid protein assay (Pierce, Rockford, IL, USA). To detect PrPSc, PNF (1 mg of protein/mL) was digested with 20 μg/mL PK (Sigma, St Louis, MO, USA) at 37 °C for 30 min. PK digestion was stopped by the addition of 0.5 mM phenylmethylsulfonyl fluoride (Sigma). The proteins were precipitated in 10 volumes of ice-cold methanol by centrifugation at 20 000 ×g for 30 min. The resultant pellet was dissolved in a sodium dodecyl sulfate (SDS) loading buffer (5% SDS, 4 M urea, 5% 2-mercaptoethanol, 5% glycerol, 0.04% bromophenol blue, 3 mM ethylenediaminetetraacetic acid (EDTA), and 62.5 mM Tris-HCl, pH 6.8) and boiled for 10 min. The PK-digested PNF (equivalent to 50 μg total protein) was subjected to SDS-polyacrylamide gel electrophoresis (PAGE) using 12% NuPAGE Bis-Tris gels (Invitrogen, Carlsbad, CA, USA), and proteins were transferred to polyvinylidene difluoride membranes (Invitrogen) for western blot analysis using the anti-PrP monoclonal antibody 44B1 as the primary antibody (24). Lysates undigested by PK were also subjected to western blotting using 44B1 and anti-actin polyclonal antibody (Sigma). The membranes were probed with the following secondary antibodies: horseradish peroxidase-conjugated anti-mouse immunoglobulin (IgG) antibody (GE Healthcare, Buckinghamshire, UK) and anti-rabbit IgG horseradish peroxidase-linked antibody (Jackson Immunoresearch Laboratories, West Grove, PA, USA). Proteins were detected using an ECL Plus western blotting detection system (GE Healthcare) and a LAS-3000 mini chemiluminescence imaging system (Fuji Photo Film, Tokyo, Japan). image gauge software (Fuji Photo Film) was used for quantification. To examine the effect of P9 on PrPSc in a cell-free system, ScN2a cells were homogenized in PBS and PNF was analyzed as described above.
P9 was dissolved in distilled water to make 50 μM solutions, and sonicated for 1 min. The solution was immediately subjected to negative staining with 2% uranyl acetate. The stained peptides were observed with an H-7650 transmission electron microscope (Hitachi, Tokyo, Japan).
Plasmids and transfection
The DNA sequence of wild-type mPrP was amplified by PCR, and the PCR product was inserted into a pCI-neo vector (Promega, Madison, WI, USA) (referred to as mPrP/pCI-neo). N2a cells were incubated with a complex of mPrP/pCI-neo and Lipofectamine 2000 (Invitrogen) for 6 hr at 37 °C, then the medium was replaced with the normal growth medium.
N2a cells transfected with mPrP/pCI-neo were seeded in a 10-cm dish and pre-incubated for 12 hr in the normal growth medium. Prior to the labeling, the cells were incubated for 10 min in Met- and Cys-free DMEM (Invitrogen) supplemented with 1% dialyzed FBS. Cellular PrPC was metabolically radiolabeled by cultivating for 1 hr in Met/Cys-free DMEM containing 1% dialyzed FBS and supplemented with 7.5 MBq/ml l-[35S]-methionine and cysteine (EasyTag™ express protein labeling mix; PerkinElmer, Boston, MA, USA). Then, the medium was changed to the normal growth medium with or without 50 μM P9 for chasing [35S]-labeled PrPC. The cells were lysed with T-DOC, and PNF was incubated with the anti-PrP polyclonal antibody B103 (Fujirebio, Tokyo, Japan) (25) and Protein A affinity gel (Sigma). Immunoprecipitated PrPC was treated with PNGaseF (New England Biolabs, Ipswich, MA, USA) and subjected to SDS-PAGE. [35S]-labeled PrPC was visualized with an imaging analyzer (FLA-2000; Fuji Photo Film) and quantified using image gauge software.
ScN2a cells incubated without, with P1, or with P9 were fixed with 4% paraformaldehyde and permeabilized with 0.1% saponin. The fixed/permeabilized cells were washed with PBS and treated with 3 M guanidine hydrochloride for 1 min at room temperature to denature PrPSc and enhance immunoreactivity (13). ScN2a cells were then washed with PBS, and incubated with the anti-PrP mouse monoclonal antibody 6H4 (Roche Applied Science, Basel, Switzerland), anti-lysosomal associated membrane proteins (LAMP)-1 rat polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and anti-cathepsin D rabbit polyclonal antibody (a gift from Dr Kominami, Juntendo University, Japan) overnight at 4 °C, followed by incubation with Alexa Fluor 488-conjugated anti-mouse IgG antibody (Invitrogen), Alexa Fluor 568-conjugated anti-rat IgG antibody (Invitrogen), and Alexa Fluor 568-conjugated anti-rabbit IgG antibody (Invitrogen) for 1 hr at room temperature. The nucleus was stained with DAPI (Dojindo Laboratories, Kumamoto, Japan). N2a cells were treated similarly as described above. Immunofluorescence images were acquired using a Zeiss LSM 510 laser scanning microscope and analyzed with image-processing software installed on a computer-controlled microscope.
Inhibition of lysosomal proteases or proteasomes
Lysosomal protease inhibitors E-64d and pepstatin A (pepA) were purchased from Peptide Institute (Osaka, Japan). For the analysis of the effects of these lysosomal proteases, ScN2a cells were pre-incubated with 50 μM E-64d, 10 μM pepA, or both E-64d and pepA for 30 min at 37 °C before 50 μM P9 was added. For the inhibition of proteasomal activity, ScN2a cells were pre-incubated with 20 μM lactacystin (Peptide Institute) for 30 min at 37 °C. Cells were cultivated with or without 50 μM P9 in the presence of lactacystin (20 μM) for 12 hr at 37 °C. After the incubation, ScN2a cells were lysed with T-DOC and PNF was subjected to western blotting using 44B1 or the anti-poly-ubiquitinylated protein monoclonal antibody FK1 (BIOMOL, Plymouth Meeting, PA, USA).
Detection of LC3
ScN2a cells were incubated with 50 μM P9, or incubated without P9, for 12 hr at 37 °C. To inhibit the degradation of LC3-II by lysosomal hydrolases (26), E-64d (50 μM) and pepA (10 μM) were added in the medium. After the incubation, ScN2a cells were lysed with T-DOC and PNF was subjected to western blotting using an anti-LC3 rabbit polyclonal antibody (26).
P9 peptide eliminates PrPSc in ScN2a cells and forms fibril
In previous studies using cell-free systems, several synthetic peptides possessing part of the amino acid sequence of PrPC blocked the conversion of PrPC to its PK-resistant form, presumably through competitive interference of the interaction between PrPC and PrPSc (18, 27). However, it has not been fully examined in living cells whether PrPC-derived peptides could inhibit PrPSc propagation or eliminate pre-existing PrPSc. Therefore, we synthesized peptides derived from PrP amino acid sequences, each consisting of nine to 21 amino acids (Table 1, Fig. 1), and explored whether the peptides inhibited PrPSc propagation in living cells. N2a cells are susceptible to prion infection, and ScN2a cells support PrPSc propagation through the continuous conversion of PrPC to PrPSc. After ScN2a cells were cultured with either of the peptides for 72 hr, PNF of the cell lysates was digested with PK. Among 14 peptides examined (named P1 to P14, Table 1, Fig. 1), only one peptide, named P9, decreased the amount of PrPSc in ScN2a cells (Fig. 2a). As the initially designed P9 was based on the amino acid sequence of bPrP, we next examined the activity of a peptide having a corresponding sequence of mPrP (mP9; Table 1, Fig. 1), and found that mP9 was as active as P9 (Fig. 2a). P9 at 20 μM induced approximately 50% inhibition of PrPSc accumulation (Fig. 2b, top panel), and no retardation of cell growth was observed up to 1 mM (results not shown). Both P9 and mP9 at 50 μM steadily decreased the amount of PrPSc in 24 hr (Fig. 2c). After three passages of ScN2a cells in the presence of 200 μM P9, PrPSc was completely eliminated, and not restored to a detectable level during a subsequent 10 passages without P9 (results not shown). Such anti-prion activity of P9 seemed to require a certain function of living cells, because P9 failed to reduce the amount of PrPSc in the lysates of ScN2a cells (Fig. 2d).
Despite the absence of hydrophobic clusters in their sequences, P9 was sparsely soluble in water and tended to yield insoluble aggregates. Electron microscopy showed that P9 formed amyloid-like fibrils of 5 to 10 nm width and >100 nm length (Fig. 2e). Notably, P14, which is known as the fibril-forming peptide PrP106-126 (28), did not inhibit PrPSc accumulation (Fig. 2a), suggesting that the elimination of PrPSc by P9 is not due to a common property of fibril peptides.
Effects of P9 on turnover rates of PrPC synthesis and degradation
Without PK digestion, the PrP-bands of ScN2a cells produced a ladder pattern in the western blot analysis (Fig. 2b, middle). The bands that appeared at 28–40 kDa corresponded to full-length forms of the di-, mono-, and non-glycosylated PrP (f-PrP) (Fig. 2b), whereas the bands at 19–28 kDa represented remnants of partially degraded PrPSc (r-PrP) (Fig. 2b) which were resistant to endogenous proteases and eventually accumulated in the cells (22). P9 preferentially weakened the intensity of the signal for r-PrP (Fig. 2b, middle), raising the possibility that P9 promotes the clearance of PrPSc by protein degradation systems in cells. In contrast, P9 did not substantially alter the signal for f-PrP in ScN2a cells (Fig. 2b) or PrPC in N2a cells (Fig. 3a). Considering the constant levels of f-PrP (Fig. 2b) and PrPC (Fig. 3a), we were curious whether P9 affected the de novo synthesis or turnover (i.e. synthesis and degradation) of PrPC. For this purpose, pulse-chase experiments were carried out using N2a cells that transiently expressed mPrP. The rate of degradation of [35S]-labeled PrPC in the presence of P9 was very similar to that in the absence of P9 (Fig. 3b, top panel). The amount of [35S]-labeled PrPC decreased to 50–60% of the initial level in a 3-hr chase period, consistent with the kinetics reported in previous studies (14, 29). Taken together, these results suggest that P9 does not affect PrPC de novo synthesis or turnover rates, but enhances PrPSc degradation in the cells.
Effects of P9 on the perinuclear accumulation of PrP
Immunofluorescence analysis with the anti-PrP monoclonal antibody 6H4 showed an accumulation of PrP in the perinuclear regions of ScN2a cells (Fig. 4a), whereas PrP signals were detected mainly in the plasma membrane but scarcely in the perinuclear regions of N2a cells (Fig. 4a). Such PrP distribution is in accordance with previous reports showing that PrPSc accumulates primarily in late endosome-like organelles, whereas PrPC is localized to the plasma membrane of ScN2a cells (11–13). The incubation of ScN2a cells with 50 μM P1, an inactive peptide, for 12 hr did not change the distribution and signal intensities of the perinuclear accumulation of PrP (Fig. 4a). In contrast, those of the accumulation of PrP in the perinuclear regions were markedly decreased by incubation with 50 μM P9 for 12 hr (Fig. 4a).
LAMP-1 is a membrane protein of late endosomes and lysosomes, and is regarded as a marker of these organelles (30). In ScN2a cells untreated with P9, the perinuclear accumulation of PrP was observed in the compartments positive for LAMP-1 (Fig. 4b, left). A large proportion (85.5%± 4.1%, mean ± SD) of the PrP signals in the perinuclear regions of ScN2a cells colocalized with the LAMP-1 signals (Fig. 4d). The perinuclear accumulation of PrP was still observed in the compartments positive for LAMP-1 even after 6 hr of incubation with 50 μM P1 or 50 μM P9 (Fig. 4b, middle and right), and populations of perinuclear PrP colocalized with LAMP-1 were unchanged (89.6%± 5.2% in the presence of P1; 90.2%± 1.9% in the presence of P9) (Fig. 4d).
On the other hand, cathepsin D is a lysosomal endopeptidase (31) and a marker of lysosomes. In the absence of P9, we observed a punctate pattern of cathepsin D signals which rarely overlapped with the perinuclear accumulation of PrP (Fig. 4c, left and middle). This indicated that in the absence of P9, PrPSc accumulated mainly in the compartments containing LAMP-1 but lacking cathepsin D. It was interesting that the incubation of ScN2a cells with P9 induced the majority of PrP/LAMP-1 signals to colocalize with anti-cathepsin D signals in the cells (Fig. 4c, right). As shown in Fig. 4d, most of the perinuclear PrP in the cells treated with P9 colocalized with cathepsin D-positive compartments (69.7%± 9.9%). In contrast, only small populations of PrPSc colocalized with cathepsin D in the untreated (14.7%± 2.4%) and P1-treated (12.9%± 6.6%) cells.
Degradation pathway stimulated by P9
To clarify the involvement of lysosomes in the reduction in PrPSc levels caused by P9, we examined whether lysosomal protease inhibitors such as E-64d (an inhibitor of thiol protease) and pepA (an inhibitor of cathepsin D) could impair the PrPSc-eliminating activity of P9 (Fig. 5a). The inhibitors in the absence of P9 did not reduce the amount of PrPSc in ScN2a cells (Fig. 5a, compare lane 1 with lanes 3, 5, and 7), whereas incubation with P9 without the protease inhibitors resulted in PrPSc clearance (compare lanes 1 and 2). The activity of P9 (lane 2) was attenuated by the addition of E-64d, pepA, or both (lanes 4, 6, and 8). A previous report suggested that autophagy is involved in PrPSc degradation (32). However, the conversion of LC3-I to LC3-II, a biochemical marker of autophagy (26), was not induced by P9 (Fig. 5b). The addition of the proteasome inhibitor lactacystin did not attenuate P9 activity, and P9 did not promote the poly-ubiquitination of cellular proteins (Fig. 5c,d).
In the present study, we found that a PrP-derived fibril peptide, P9, eliminates pre-existing PrPSc and prevents further persistent infection in prion-infected neuroblastoma cells. To date, a number of non-peptidic compounds have been reported to eliminate PrPSc in prion-infected cultured cells, including ScN2a cells (33), although the complete mechanisms of their anti-prion effects are still under investigation. Although many of these non-peptidic compounds require a few days of culture of the cells to eliminate PrPSc, P9 steadily decreases the amount of PrPSc in 24 hr. Considering that the half-life of PrPSc in ScN2a cells was estimated to be longer than 24 hr (14), the rapid clearance of PrPSc suggested that P9 might have induced the ablation of pre-existing PrPSc in the cells. Alternatively, one would assume that P9 reduced PrPC de novo synthesis, with the reduction in the cellular pool of PrPC eventually resulting in a decrease in the amount of PrPSc converted from PrPC. Another possibility is that P9 accelerated the turnover rate of PrPC, so that the rapid turnover prevented newly synthesized PrPC from interacting with pre-existing PrPSc. We believe that the latter two possibilities can be ruled out, because there was no significant decrease in the amount of cellular PrPC (Fig. 3a) and no acceleration of PrPC turnover (Fig. 3b) in the cells treated with P9. Furthermore, E-64d and pepA attenuated P9 activity. Western blot analysis showed that the amounts of LC3-I and LC3-II, biochemical markers of LC3-dependent autophagy (34), remained unchanged. Also, the addition of P9 did not enhance the poly-ubiquitination of cellular proteins, and lactacystin did not diminish P9 activity. All these observations support the notion that the elimination of PrPSc in the cells is due to the ablation of pre-existing PrPSc by lysosomal degradation, and not the activation of autophagy or the ubiquitin-proteasome system by P9.
Lysosomes are dynamic organelles that receive and degrade macromolecules from secretory, endocytic, autophagic and phagocytic membrane-trafficking pathways (35). Although the details of the trafficking of PrPSc remain unsolved at present, it has been shown that PrPSc accumulates mainly in late endosome-like organelles rather than in mature lysosomes in scrapie-infected mouse brain (11) or in ScN2a cells (36).
In the absence of P9, we observed dense accumulation of PrP in the perinuclear compartments positive for LAMP-1 (a marker of late endosomes/lysosomes) but negative for cathepsin D (a marker of lysosomes). In accordance with the previous reports, such accumulation of PrP was expected to be equivalent to, if not all, the aggregates of PrPSc. Upon treatment of the cells with P9, cathepsin D-positive compartments started to merge with the PrP-accumulating compartments (Fig. 4). On the basis of these results, we conclude that P9 exerts its PrPSc-eliminating activity by sorting the aggregates of PrPSc, which might have been accumulated in LAMP-1-positive late endosomes, to the cathepsin D-positive lysosomal compartments. This conclusion also gives a plausible explanation as to why P9 did not induce remarkable reduction of the signals of PrPC (Fig. 3a) or those of f-PrP (Fig. 2b, middle) in the western blot analysis. PrPC residing mainly on plasma membranes (Fig. 4a, right; references 11, 36) and f-PrP residing on plasma membranes and early endosomes (11, 36), would be out of range of the promotion of the P9-promoted sorting of endosomes to lysosomal compartments. Therefore, P9 did not cause reduction of the amounts of PrPC (Fig. 3a) or f-PrP (Fig. 2b, middle). These results indicate that the trafficking of late endosome-like vesicles containing PrPSc to lysosomes plays a crucial role in the accumulation or elimination of PrPSc.
P9 is prone to undergo self-assembly into fibril structures (Fig. 2e). It has been shown that the formation of some type of amyloid, such as amyloid β-protein, PrP, α-synuclein, and amyloid protein A, is accelerated in the presence of preformed oligomeric fibrils that serve as nucleation seeds (37–39). However, despite its self-assemblying property into fibrils, P9 did not enhance the conversion of PrPC to PrPSc. Note that the PrP-derived peptide PrP106-126, named as P14 in this study, polymerizes intrinsically into amyloid-like fibrils rich in β-sheets in vitro, and that chronic exposure of primary cultures of rat hippocampal neurons to micromolar concentrations of PrP106-126 induces neuronal cell death by apoptosis (28). However, P14 lacked the activity to eliminate PrPSc (Fig. 2a). Hence, we conclude that the elimination of pre-existing PrPSc by P9 is not a common property of PrP-derived fibrils.
Previously, Supattapone et al. (21) showed that in vitro exposure to a certain type of cationic dendrimer at a stoichiometric ratio rendered PrPSc susceptible to PK digestion, and that the dendrimer also reduced the amount of pre-existing PrPSc in ScN2a cells after culture for 4 weeks with the compound. Their plausible interpretation was that the stoichiometric binding of the dendrimer to the monomeric/oligomeric PrPSc restricted the reaggregation of monomeric/oligomeric PrPSc which were in equilibrium with the large aggregated form(s) of PrPSc. Also, they raised the possibility that the dendrimer facilitated the transport of PrPSc from the plasma membrane to lysosomes based on the observation of lysosomal accumulation of the dendrimer. Recently, Ertmer et al. (22) reported that STI571 (10 μM), an inhibitor of tyrosine kinases, eliminated PrPSc in ScN2a cells. As this STI571 activity was inhibited by the addition of 10 mM ammonium chloride to the culture medium, they ascribed the anti-prion activity of STI571 to activation of the lysosomal degradation of PrPSc. In the present study, we showed that P9 induced PrPSc sorting from late endosomes to lysosomes, showing a change in the localization of PrPSc which was not stated in previous reports describing the anti-prion activity of cationic dendrimers and STI571. Further study is needed to determine whether P9, a cationic dendrimer, and STI571 share the same mechanism for promoting the lysosomal degradation of PrPSc. Nevertheless, previous studies and ours suggest that the impairment of the recruitment of aggregated PrPSc to lysosomes is one of the causes of PrPSc accumulation in prion-infected cells, and that certain stimuli to the cells given by P9, STI571, or the cationic dendrimer can somehow restore the impairment.
Further study of the mechanism underlying the activity of P9 will provide a clue as to why prion-infected cells fail to eradicate aggregates of PrPSc, and will contribute to the development of a new strategy for clearing PrPSc without affecting the cellular level of PrPC.
We thank M. Horiuchi (Hokkaido University) for the anti-PrP antibody 44B1, E. Kominami and T. Ueno (Juntendo University) for the anti-cathepsin D antibody, and T. Muramoto (Tohoku University) for ScN2a cells. This study was supported by Grants-In-Aid for BSE Research from MHLW, Japan (17270701, Y.Y. and K.H.) and for Exploratory Research from MEXT, Japan (17659024, K.H.).