The prion diseases occur following the conversion of the cellular prion protein (PrPC) into an alternatively folded, disease-associated isoform (PrPSc). However, the spread of PrPSc from cell to cell is poorly understood. In the present manuscript we report that soluble PrPSc bound to and replicated within both GT1 neuronal cells and primary cortical neurons. The capacity of PrPSc to bind and replicate within cells was significantly reduced by enzymatic modification of its GPI (glycosylphosphatidylinositol) anchor. Thus PrPSc that had been digested with phosphatidylinositol-phospholipase C bound poorly to GT1 cells or cortical neurons and did not result in PrPSc formation in recipient cells. PrPSc that had been digested with phospholipase A2 (PrPSc-G-lyso-PI) bound readily to GT1 cells and cortical neurons but replicated less efficiently than mock-treated PrPSc. Whereas the addition of PrPSc increased cellular cholesterol levels and was predominantly found within lipid raft micro-domains, PrPSc-G-lyso-PI did not alter cholesterol levels and most of it was found outside lipid rafts. We conclude that the nature of the GPI anchor attached to PrPSc affected the binding of PrPSc to neurons, its localization to lipid rafts and its ability to convert endogenous PrPC.
Transmissible spongiform encephalopathies, more commonly known as the prion diseases, are mostly fatal neurodegenerative disorders that include scrapie in sheep and goats, bovine spongiform encephalopathy in cattle and kuru and Creutzfeldt–Jakob disease in humans. A key event in prion disease is the conversion of a normal host protein (PrPC) into a disease-associated isoform (PrPSc), which represents the major component of infectious scrapie prions . The conversion process results in a portion of the α-helix and random coil structure in PrPC being refolded into a β-pleated sheet in PrPSc . The newly formed PrPSc subsequently acts as a template that facilitates conversion of PrPC into PrPSc. This conversion is accompanied by changes in biological and biochemical properties of PrPSc, including increased resistance to proteases .
Precisely how PrPSc spreads throughout the brain during infection is poorly understood. PrPSc can be released from cells in exosomes [4,5], nanometre-sized particles that contain specific proteins  which could infect neighbouring cells or by tunnelling nanotubules . The mechanisms by which PrPSc infects neuronal cells has been examined in vitro; PrPSc infected a neuronal cell line (SN56), hamster neurons  and murine neurons . Furthermore, N2a cells were used to provide a highly sensitive cell-based infectivity assay for mouse scrapie prions . In the present study we show that PrPSc bound to and replicated within both GT1 neuronal cells and murine cortical neurons.
Proteins that contain a GPI (glycosylphosphatidylinositol) anchor are rapidly incorporated into both living cells and synthetic membranes [11,12]. PrPSc molecules are linked to membranes via a GPI anchor , suggesting that the GPI anchor may have a role in the binding of PrPSc to cells. Previously, the role of GPI anchors in the biological activity of PrPSc has been difficult to determine as the aggregated forms of PrPSc were thought to be resistant to digestion with PI-PLC (phosphatidylinositol-specific phospholipase C) [14,15]. However, sonication and shaking techniques, originally developed to amplify small amounts of PrPSc, break up large prion particles into smaller soluble oligomers . In the present study, a combination of a shaking technique and soluble PrPSc released from ScGT1 cells, rather than PrPSc extracted with strong detergents from infected brains, allowed successful digestion of the PrPSc-GPI anchor with PI-PLC or PLA2 (phospholipase A2). These enzymes target the acyl chains of the GPI anchor (Figure 1). We report that PrPSc digested with PI-PLC [PrPSc-IPG (PrPSc-inositolphosphoglycan)] did not bind readily to GT1 cells or cortical neurons and did not lead to the formation of PrPSc in recipient cells over the time course of these experiments. Surprisingly, although PrPSc that had been digested with PLA2 (PrPSc-G-lyso-PI) bound readily to GT1 cells and cortical neurons, it had little capacity to convert endogenous PrPC into PrPSc.
GT1 cells were grown in Ham's F12 medium supplemented with 2 mM glutamine, 100 units/ml penicillin, 100 μg/ml streptomycin, 200 nM retinoic acid and 2% fetal calf serum (PAA). GT1 cells were plated in six-well plates (105 cells/well for PrPSc replication studies and 106 cells/well for binding studies) and were allowed to adhere overnight.
Primary cortical neurons
Cortical neuronal cultures were prepared from the brains of mouse embryos (day 15.5) derived from wild-type (PrP+/+) or PrP-knockout mice (PrP0/0) after mechanical dissociation and cell sieving. Mice were killed by cervical dislocation to provide cells. Neuronal precursors were plated at 106 cells/well in six-well plates, which had been pre-coated with 5 μg/ml poly-L-lysine, in Ham's F12 containing 5% fetal calf serum for 2 h. Cultures were shaken (600 rev./min for 5 min) and non-adherent cells were removed by 2 washes in PBS. Neurons were subsequently grown in neurobasal medium containing B27 components (PAA) for 7 days before use. Immunostaining showed that the cells were greater than 97% neurofilament positive. Less than 3% stained for GFAP (glial fibrillary acidic protein; astrocytes) or for F4/80 (microglial cells) (results not shown).
Test PrPSc preparations
Two distinct PrPSc preparations were used in these assays. The first consisted of the supernatants collected from ScGT1 cells which were centrifuged (500 g for 5 min) to remove cell debris. Supernatants were digested with 0.2 units/ml PI-PLC (Bacillus cereus) or 10 units/ml bee venom PLA2 (Sigma) with intermittent shaking at 37 °C for 24 h. The enzymatic activity was destroyed by heating at 70 °C for 5 min. This method eschewed proteinase K digestion in order to preserve protease-sensitive forms of PrPSc. The amount of PrPSc in these supernatants was determined by ELISA and diluted to 1 ng/ml for further studies. We also used partially purified PrPSc preparations isolated from ScGT1 cells homogenized in distilled water. Cell debris and nuclei were removed by centrifugation at 500 g for 5 min. Cell membranes were collected from the supernatant by centrifugation (10 min at 14000 g) and solubilized in buffer containing 10 mM Tris/HCl, 10 mM NaCl, 10 mM EDTA, 0.5% Nonidet P40 and 0.5% sodium deoxycholate. PrPC and PrPSc were immunoprecipitated by incubation with 0.5 μg/ml ICSM35, which recognizes native PrPSc (D-Gen, http://www.d-gen.co.uk) and μMACS Protein G microbeads (10 μl/ml) (Miltenyi Biotech). Protein G-bound antibody complexes were isolated using a μMACS magnetic system (Miltenyi) and digested with 0.2 unit/ml PI-PLC or 10 units/ml bee venom PLA2 with intermittent shaking at 37 °C for 24 h. Precipitates were further digested with 5 μg/ml proteinase K for 1 h at 37 °C to remove PrPC. Trifluoroacetic acid was added to achieve a 0.1% final concentration, samples were shaken in a cell disruptor (10 min) and applied to C18 columns (Waters). PrPSc was eluted under a gradient of acetonitrile in water and 0.1% trifluoroacetic acid and fractions were tested for PrP by ELISA. PrP-positive fractions were pooled and run on a second C18 column. Samples containing PrP were pooled, washed twice with PBS and the amount of PrPSc was determined by ELISA. Samples were frozen at −20 °C for storage. On the day of use, PrPSc preparations were diluted to 1 ng/ml and solubilized in culture medium.
For PrPSc replication experiments, cells were pulsed with PrPSc preparations for 24 h, washed five times in warmed PBS to remove unbound PrPSc and incubated in fresh medium for a further 6 days (GT1 cells) or 9 days (cortical neurons). For binding studies, cells were incubated with PrPSc preparations for 1 or 24 h. Cells were then washed five times in warmed PBS and cell membrane extracts collected.
Cell membrane extracts
Treated cells were homogenized in an extraction buffer (10 mM Tris/HCl, pH 7.4, 100 mM NaCl, 10 mM EDTA, 0.5% Nonidet P40, 0.5% sodium deoxycholate and 0.2% SDS) at 106 cells/ml. Membranes were homogenized and nuclei and large fragments were removed by centrifugation (300 g for 5 min). To determine the amount of PrPSc in cells, these extracts were digested with 5 μg/ml proteinase K for 1 h at 37 °C; digestion was stopped with mixed protease inhibitors. The soluble material was heated to 95 °C for 5 min to denature PrPSc and tested in a PrP ELISA. Some samples were mixed with an equal volume of Laemmli buffer (Bio-Rad) and subjected to electrophoresis on a 15% polyacrylamide gel. Proteins were transferred on to Hybond-P PVDF or nitrocellulose membranes (Amersham Biotech) by semi-dry blotting. Membranes were blocked using 10% non-fat dried skimmed milk powder in Tris-buffered saline, pH 7.2, containing 0.2% Tween 20. PrP was detected by incubation with mAb (monoclonal antibody) ICSM35 (D-gen), followed by a secondary anti-mouse IgG conjugated to peroxidase. Bound antibody was visualized using an enhanced chemiluminescence kit (Amersham Biotech).
Isolation of GPI anchors
PrPSc samples were further digested with 100 μg/ml proteinase K at 37 °C for 24 h to release GPI anchors. The released GPIs were extracted with water-saturated butan-1-ol, washed three times with ice-cold water, freeze-dried and dissolved in ethanol. GPIs were applied to silica gel 60 HPTLC (high-performance TLC) plates (Whatman) and developed using a mixture of choloroform/methanol/water (4:4:1, by vol.). GPI anchors were detected by immunoblotting as described previously . Briefly, plates were soaked in 0.1% poly(isobutyl methacrylate) in hexane, dried, and blocked with PBS containing 5% non-fat dried skimmed milk powder. They were probed with 1 μg/ml of a mouse mAb that binds to phosphatidylinositol, washed with PBS-Tween (PBS plus 0.2 % Tween 20), and incubated with goat anti-mouse IgG conjugated to peroxidase (Sigma) for 1 h. The bound antibody was washed and visualized using enhanced chemiluminescence.
Isolation of detergent-resistant membranes (lipid rafts)
Treated cells were lysed in an ice-cold buffer containing 1% Triton X-100, 10 mM Tris/HCl, pH 7.2, 100 mM NaCl and 10 mM EDTA at 106 cells/ml. Membranes were prepared by homogenization followed by mechanical agitation on a cell disruptor (10 min full power), and nuclei and large fragments were removed by centrifugation (300 g for 5 min). The supernatant was incubated on ice for 60 min prior to further centrifugation (16000 g for 30 min at 4 °C). The soluble material was reserved as the normal cell membrane. Insoluble proteins were solubilized in extraction buffer [10 mM Tris/HCl, pH 7.4, 10 mM NaCl, 10 mM EDTA, 0.5% Nonidet P40, 0.5% sodium deoxycholate and 0.2% SDS and mixed protease inhibitors (Sigma)], centrifuged (10 min at 16000 g), and the supernatant retained as the lipid raft fraction. Both fractions were digested with proteinase K (5 μg/ml for 1 h at 37 °C) to remove PrPC, heated to 95 °C for 5 min to denature PrPSc and tested in a PrP ELISA.
The amount of PrP in cell extracts/supernatants was determined by ELISA as previously described . Maxisorb Immunoplates (Nunc) were coated with 0.5 μg/ml capture mAb [ICSM18 (D-Gen)] which recognizes amino acids 146–159 of murine PrP. Samples were applied and detected with biotinylated mAb ICSM35 (D-Gen) (which recognizes an epitope between amino acids 91 and 110). Biotinylated mAb was detected using extravidin-alkaline phosphatase and 1 mg/ml 4-nitrophenyl phosphate (Sigma). Absorbance was measured at 405 nm and the amount of PrP in cell extracts was calculated by reference to a standard curve of recombinant murine PrP (Prionics).
The amount of cholesterol was measured using the Amplex Red cholesterol assay kit (Invitrogen). Cholesterol is oxidized by cholesterol oxidase to yield hydrogen peroxide, which reacts with 10-acetyl-3,7-dihydroxyphenoxazine (Amplex Red reagent) to produce highly fluorescent resorufin; measured by excitation at 550 nm and emission detection at 590 nm. The amount of cholesterol was calculated by reference to cholesterol standards.
Comparison of treatment effects was carried out using one- and two-way ANOVA techniques as appropriate. For all statistical tests, significance was set at the 1% level.
Partial purification of phospholipase-digested PrPSc on C18 columns
Digestion with PI-PLC results in the removal of diacylglycerol from GPI anchors. As shown in Figure 1, the digestion of PrPSc with PI-PLC resulted in PrPSc-IPG. Whereas mock-treated PrPSc (PrPSc-GPI) bound to C18 columns and was eluted in fractions containing between 65 and 85% acetonitrile, PrPSc-IPG did not bind to C18 columns and was eluted in the void volume (Figure 2A). PrPSc digested with PLA2 (PrPSc-G-lyso-PI) bound to C18 columns and was eluted in fractions containing between 35 and 50% acetonitrile. Positive fractions were pooled, the amount of PrPSc measured by ELISA and the pool was diluted to 1 ng/ml for future assays. The amount of PrPSc in cell preparations was not significantly altered following digestion with either PLA2 (4.8±0.4 compared with 4.6±0.5 ng/ml, n=9, P=0.6) or PI-PLC (4.5±0.3 compared with 4.6±0.5 ng/ml, n=9, P=0.8). Western blot analysis did not detect an observable difference in the apparent molecular mass of these preparations (Figure 2B). However, HPTLC analysis of GPI anchors isolated from phospholipase-digested PrPSc preparations showed that the GPI isolated from PrPSc digested with PLA2 or PI-PLC had different migration patterns to those isolated from mock-treated PrPSc (Figure 2C).
Phospholipase digestion of PrPSc reduced infectivity for GT1 cells
In the first set of experiments, GT1 cells were pulsed with different amounts of mock-treated or phospholipase-digested ScGT1 supernatants (1–0.016 ng) for 24 h. Mock-treated supernatants infected GT1 cells, and the PrPSc content of recipient GT1 cells was dependent upon the dose of the original inoculum (Figure 3A). In contrast, recipient GT1 cells pulsed with 1 ng of PI-PLC-digested ScGT1 supernatants did not contain detectable levels of PrPSc (<0.05 ng). GT1 cells pulsed with PLA2-digested ScGT1 supernatants contained less PrPSc than GT1 cells pulsed with mock-treated supernatants.
GT1 cells were also pulsed with partially purified PrPSc preparations. The addition of PrPSc-GPI resulted in a dose-dependent increase in the amount of PrPSc in recipient GT1 cells (Figure 3B). The addition of 1 ng of PrPSc-GPI did not significantly affect cell survival, as measured by thiazyl blue tetrazolium. We did not detect PrPSc in GT1 cells pulsed with 1 ng PrPSc-IPG (<0.05 ng of PrPSc). GT1 cells pulsed with 1 ng of PrPSc-G-lyso-PI contained significantly less PrPSc than GT1 cells pulsed with 1 ng of PrPSc-GPI (0.28±0.25 ng of PrPSc compared with 6.1±1.12 ng of PrPSc, n=27, P<0.01). Western blots confirmed that after 7 days GT1 cells pulsed with PrPSc-GPI contained greater amounts of PrPSc than did cells pulsed with PrPSc-G-lyso-PI or PrPSc-IPG (Figure 3C).
Phospholipase digestion reduced PrPSc infectivity for cortical neurons
To confirm the GT1 cell data in non-dividing cells, primary cortical neurons were pulsed with ScGT1 supernatants at different concentrations. Mock-treated ScGT1 supernatants infected cortical neurons, and there was a dose-dependent increase in the PrPSc content of recipient cortical neurons (Figure 4A). Cortical neurons pulsed with PLA2-digested ScGT1 supernatants also contained PrPSc. However, digestion with PLA2 significantly reduced the infectivity of ScGT1 supernatants. We did not detect PrPSc in cortical neurons pulsed with ScGT1 supernatants containing 1 ng of PrPSc that had been digested with PI-PLC (<0.05 ng of PrPSc).
The addition of partially purified PrPSc-GPI resulted in a dose-dependent increase in the amount of PrPSc in cortical neurons (Figure 4B). Thus the addition of 1 ng of PrPSc-GPI was amplified to 44±6.7 ng of PrPSc, n=22, after 10 days in cortical neurons. We were unable to detect PrPSc in cortical neurons pulsed with 1 ng of PrPSc-IPG. Cortical neurons pulsed with 1 ng of PrPSc-G-lyso-PI contained 3.63±2 ng of PrPSc, n=22. The PrPSc in these cells bound to C18 columns and was eluted in fractions containing greater than 70% acetonitrile, indicating that the PrPSc contained a GPI anchor with two acyl chains. Thus this PrPSc had been synthesized by the recipient cortical neurons and was not a remnant of the original inoculum. Primary cortical neurons derived from PrP-knockout mice were also pulsed with ScGT1 supernatants containing 1 ng of PrPSc or 1 ng of partially purified PrPSc-GPI for 24 h and incubated for a further 9 days. We were unable to detect PrPSc in these cells using either ELISA (<0.05 ng of PrPSc) or immmunoblot (Figure 4C).
The binding of PrPSc to cells is reduced by phospholipase digestion
To determine whether GPI anchors affected the binding of PrPSc to recipient cells, GT1 cells were incubated with 1 ng of partially purified PrPSc preparations for 1 or 24 h, washed, and the amount of cell-associated PrPSc was measured. After 1 h, there was no significant difference in the amount of PrPSc-GPI and PrPSc-G-lyso-PI that had bound (0.95±0.1 compared with 0.93±0.12 ng, n=12, P=0.7) (Figure 5A). After 1 h incubation, we were unable to detect any PrPSc-IPG that had bound to cells (<0.05 ng of PrPSc-IPG), indicating that acylation of the GPI anchor facilitates the efficient binding of PrPSc to neurons. However, increasing the incubation period to 24 h resulted in a small amount of cell-associated PrPSc-IPG (128±42 pg, n=12).
PLA2-digested PrPSc does not target lipid rafts
Next, we sought to determine why it was that, although PrPSc-G-lyso-PI bound readily to cells, only low amounts of PrPSc were produced in recipient cells. PrPSc production is thought to be dependent on the formation or function of cholesterol-sensitive lipid rafts. Although the addition of 1 ng of PrPSc-GPI significantly increased the amount of cholesterol in GT1 cell membranes (573±57 compared with 395±38 ng/106 cells, n=8, P<0.01), the addition of 1 ng of PrPSc-G-lyso-PI had no significant effect (402±47 compared with 395±38 ng/106 cells, n=8, P=0.78). PrPSc is found within lipid rafts that are defined by their resistance to solubilization by non-ionic detergents . We report that >90% of PrPSc-GPI was found in the detergent-resistant (lipid raft) fraction of GT1 cell membranes. In contrast, <10% of PrPSc-G-lyso-PI was found in the lipid raft fraction (Figure 5B). Similar results were obtained when these preparations were added to cortical neurons; PrPSc-GPI was mainly found within lipid rafts and PrPSc-G-lyso-PI within the normal cell membrane.
The molecular mechanisms by which prions spread throughout an infected brain remain poorly understood. In the present study, we demonstrate the role of acyl chains of the GPI anchor on the ability of PrPSc to bind neuronal cells and to replicate. PrPSc with a conventional GPI anchor (PrPSc-GPI) was able to bind and replicate within both GT1 neuronal cells and primary cortical neurons. As expected, PrPSc did not replicate within cortical neurons derived from PrP-knockout mice. Modification of the GPI anchor reduced the infectivity of PrPSc; PrPSc-IPG did not bind to cells and although PrPSc-G-lyso-PI did bind to cells, it did not readily convert PrPC into PrPSc.
The partial purification of phospholipase-digested PrPSc molecules was facilitated by using PrPSc isolated from ScGT1 cells rather than from infected brains. PrPSc exists as aggregates of different sizes  and large aggregates of PrPSc are thought to cause steric hindrance of PI-PLC. In the present study, vigorous shaking increased the susceptibility of PrPSc to phospholipases, presumably because it reduced the size of PrPSc aggregates. Digested products were isolated by differential binding and elution from C18 columns in a gradient of acetonitrile and water. Western blot analysis of PI-PLC or PLA2-digested preparations showed no apparent difference in molecular mass. This result was not surprising, considering digestion with PLA2 would reduce the molecular mass by only 0.2 kDa and PI-PLC digestion by 0.4 kDa. In contrast, the removal of the whole GPI anchor would be expected to reduce the molecular mass by approx. 5 kDa. However, HPTLC analysis showed that digestion with either PI-PLC or PLA2 altered the GPI anchors of these preparations.
Whereas PrPSc-GPI bound rapidly to neurons, PrPSc-IPG did not, indicating that the acylation of the GPI anchor affected PrPSc binding to cells. The binding of PrPSc to cells is complex, and although there are many reports of specific PrPSc-protein interactions , the nature and presence of specific protein receptors for PrPSc is yet to meet an agreed consensus. Our observation that the removal of both acyl chains greatly reduced the binding of PrPSc to cells shows that they contribute strongly to the binding of PrPSc to cells. Such results do not necessarily exclude prion–protein or prion–glycan interactions, which may be facilitated after PrPSc has inserted into the membrane. We conclude that in these studies PrPSc-IPG in the supernatant had limited interactions with cell-associated PrPC.
At first glance, our results appear to contradict studies from transgenic mice producing PrPC without a GPI anchor, which produced high amounts of PrPSc and infectivity . However, in these transgenic mice, anchorless PrPC was rapidly secreted from neurons. Moreover, a recent study showed that cell lines producing anchorless PrPC were resistant to scrapie infection . A direct comparison with these results is invalid considering that the removal of the entire GPI anchor in these mice affected the glycosylation of PrPC and its expression at the surface of neurons. As the glycosylation status of PrPC can affect PrPSc formation , it is not clear whether the effects of removing the entire GPI anchor was a direct effect of the loss of the GPI anchor, or an indirect effect resulting from altered glycosylation. Another study in which recombinant PrP that did not contain a GPI anchor induced prion disease in transgenic mice  suggests that the GPI anchor is a not essential for infectivity. Our studies suggest that acylation of the GPI anchor facilitates PrPSc binding, but is probably not the sole factor involved in infectivity. The results of the present study also differ from a study which reported that removal of the GPI anchor from PrPSc by cathepsin D did not affect prion infectivity . However, digestion with cathepsin D removes the whole GPI anchor as well as a fragment of the C-terminus, whereas phospholipases simply modify the GPI anchor by the removal of one or both acyl chains as shown in Figure 1. Several technical differences are also apparent in the two studies ( and ), those authors used PrPSc derived from infected brain homogenates, which was extracted using strong detergents and uninfected cells were incubated with PrPSc for 3 days. We used two PrPSc preparations, the first (ScGT1 supernatants) contained PrPSc that had been released from ScGT1 cells without the use of detergents and which had not been protease treated. Since protease-sensitive forms of PrPSc have been identified , our preparations overcame the problem of protease digestion reducing the infectivity of PrPSc. Although the second PrPSc preparation was extracted using a mild detergent, most of the detergent was removed by the subsequent passage through C18 columns prior to the addition of PrPSc to cells. The lack of detergent in our PrPSc preparations is a possible explanation of the differences between our studies and previous reports.
The amount of PrPSc-G-lyso-PI and PrPSc-GPI that had bound to recipient cells after 1 h was not significantly different, indicating that a single acyl chain is sufficient to facilitate the binding of PrPSc to neuronal cells. Crucially, although PrPSc-G-lyso-PI bound to cells, it did not readily convert endogenous PrPC. Although these preparations contained mostly PrPSc-G-lyso-PI, we cannot exclude the possibility that they also contained a small amount of undigested PrPSc-GPI which was responsible for PrPSc formation in recipient cells. Most GPI-anchored proteins are targeted to cholesterol-dense lipid rafts and the conversion of PrPC into PrPSc is cholesterol sensitive [28,29]. GPI anchors contain predominantly unsaturated acyl chains, which encourage the solubilization of cholesterol . In the presnt study, the addition of PrPSc-GPI significantly increased the cholesterol content of cells, whereas the addition of PrPSc-G-lyso-PI had no effect. The amount of cholesterol in cell membranes affects lipid raft formation and >90% of PrPSc-GPI was targeted to lipid rafts. In contrast, <10% of PrPSc-G-lyso-PI was found within lipid rafts, indicating that these molecules were mostly in the normal cell membrane. The localization of PrPSc-G-lyso-PI to non-raft membranes may restrict its interactions with endogenous PrPC, which is localized to lipid rafts. Since lipid rafts can regulate cell signalling , the localization of PrPSc-GPI to lipid rafts may enhance cell signalling pathways involved in PrPC conversion. In contrast, the localization of PrPSc-G-lyso-PI to non-raft membranes may reduce cell signalling involved in PrPC conversion.
In summary, we showed that PrPSc was able to bind and replicate in both GT1 neuronal cells and primary cortical neurons. A combination of phospholipase digestion and shaking allowed the role of the GPI anchor on PrPSc binding and replication to be determined. PI-PLC digestion of PrPSc reduced its ability to bind to cells. Although PLA2-digested PrPSc bound to neuronal cells, it was not found within lipid rafts and infected cells contained only low amounts of PrPSc. Such studies show that the GPI anchor attached to PrPSc affects the targeting of PrPSc to specific membranes and its ability to either interact with, or convert, endogenous PrPC. We speculate that compounds that alter GPI anchors may provide a novel therapeutic approach to combat prion disease.
Clive Bate was responsible for planning and carrying out the experiments and writing the manuscript. Alun Williams contributed to the planning the experiments and writing the manuscript. Mourad Tayebi helped with the PrP ELISA and contributed to the discussion.
This work was supported by the European Commission FP6 “Neuroprion” - Network of Excellence.
Abbreviations: GPI, glycosylphosphatidylinositol; HPTLC, high-performance TLC; IPG, inositolphosphoglycan; mAb, monoclonal antibody; PI-PLC, phosphatidylinositol-specific phospholipase C; PLA2, phospholipase A2, PrP, prion protein; PrPC, normal host prion protein, PrPSc, disease-associated prion isoform; PrPSc-G-lyso-PI, PrPSc digested with PLA2
- © The Authors Journal compilation © 2010 Biochemical Society