Introduction Top

Transmissible spongiform encephalopathies (TSE diseases) or prion diseases are fatal neurodegenerative diseases of humans and animals. These diseases include scrapie in sheep, bovine spongiform encephalopathy (BSE) in cattle, and chronic wasting disease (CWD) in cervids, as well as several human diseases including kuru, Gerstmann-Sträussler-Scheinker syndrome(GSS), and sporadic, familial and variant forms of Creutzfeldt-Jakob disease (CJD) (see [1] for review). TSE diseases are transmissible within a species, but can also cross to new species in some cases. For example, variant CJD appears to be a form of BSE transmitted to humans. In addition, experimental transmission to rodents such as mice, hamsters and, more recently, bank voles [2],[3], has provided numerous models for laboratory research.

In prion diseases brain pathology is characterized by spongiform degeneration of the gray matter together with neuronal loss and gliosis. During disease there is an accumulation in brain of an abnormal partially protease-resistant form of prion protein (PrPres) derived from host-encoded protease-sensitive prion protein (PrPsen). PrPres can be detected by immunoblot or immunohistochemistry, and this detection is often used as an important diagnostic feature of prion disease. PrPres can be deposited in brain either as large fibrillar amyloid plaques and/or as small diffuse punctate deposits of non-amyloid aggregated protein. The diffuse non-amyloid PrPres form is prevalent in many human sCJD cases and most prion disease animal models [4]–[7]. However, both amyloid and non-amyloid forms of PrPres coexist in some human and animal prion diseases [6] [8]–[13], and both forms may contribute to prion disease pathogenesis.

A variety of proteins are capable of forming amyloid deposits in nervous system tissues as well as other organs. Amyloid deposits often displace organ structure resulting in dysfunction and cell death. In cerebral amyloid angiopathy (CAA), associated with Alzheimer’s disease (AD) and several genetic CNS amyloid diseases, vascular amyloid deposits can damage the structure of blood vessel walls leading to hemorrhage or thrombosis [14],[15]. However, in AD, oligomeric pre-amyloid Aß aggregates are also thought to have important neuropathogenic effects. For therapy of diseases such as AD and prion diseases, where both amyloid and non-amyloid may be pathogenic, it will be important to understand the contribution of both types of abnormal protein aggregates to the various pathogenic processes present in these complex diseases. Therefore, we focused on the pathogenic effects of PrPres amyloid versus non-amyloid induced by prion infection in mice.

In uninfected animals PrPsen is anchored to the plasma membrane by a glycosylphosphatidylinositol (GPI) moiety [16]. In prion disease PrPres is found on plasma membranes of neurons and other brain cells, where it is associated with morphological membrane changes that are common to different animal TSEs [7] [17],[18]. Membrane attachment of PrP may have an important influence on the prion disease process. To study the role of PrP membrane linkage on pathogenesis of prion disease we previously generated 2 lines of transgenic mice (tg44+/− and tg23+/−), which express PrP lacking the GPI anchor at similar levels and do not express GPI membrane–anchored PrP. Anchorless PrP in these mice is secreted by cells and is not attached to the plasma membrane [19]. After scrapie infection, both lines of transgenic mice developed high titers of prion infectivity and extensive PrPres amyloid deposits in brain at late times after infection; however, typical scrapie clinical signs and gray matter spongiosis characteristic of prion diseases were not seen [19].

In the present paper we studied homozygous anchorless PrP transgenic mice, which expressed two-fold more anchorless PrPsen than the above mentioned heterozygous transgenic mice. In these experiments scrapie infection of homozygous mice produced a fatal clinical disease. However, this disease differed in incubation period, clinical signs, and neuropathology from typical prion disease seen in non-transgenic mice, which express anchored PrP, and thus appeared to be a distinct pathogenic process. Therefore, depending on the presence or absence of anchored PrP, scrapie infection could induce two different fatal brain diseases: PrPres amyloidosis without gray matter spongiosis in anchorless PrP transgenic mice, and diffuse non-amyloid PrPres with gray matter spongiosis in mice with anchored PrP.

Results Top

Brain PrPsen levels in anchorless PrP transgenic mice and non-transgenic mice

Since PrPsen expression is known to influence scrapie incubation period [20] [21], it is possible that low PrP expression might account in part for the lack of clinical scrapie disease in previous experiments using heterozygous tg44+/− and tg23+/− mice [19]. Therefore, in the present study we generated homozygous anchorless PrP transgenic mice from both lines 44 and 23. These mice each expressed two anchorless PrP transgene alleles and no normal mouse PrP alleles.

PrPsen levels were analyzed by immunoblotting in brain homogenates of uninfected transgenic and non-transgenic mice. Homozygous tg44+/+ mice expressed 2-fold higher levels compared to tg44+/− mice (Figure 1A). Non-transgenic C57BL/10SnJ mice homozygous for the PrP gene (Prnp+/+) were used as controls, and these mice expressed 2-fold higher PrP levels than did Prnp+/− mice (generated by crossing Prnp+/+ mice to Prnp-null mice also on the C57BL/10SnJ background (see methods)) (Figure 1A). Quantitative comparisons between transgenic and non-transgenic mice were difficult due to PrP glycosylation differences (Figure 1A). Therefore we compared these mice using PrPsen deglycosylated with PNGase F (Figure 1B). In these experiments Prnp+/− mice had approximately four-fold higher PrPsen levels than tg44+/+ mice (compare lanes 2 vs. 4 and lanes 7 vs. 10). Given the 2-fold difference between Prnp+/+ and +/− mice, tg44+/+ expressed 8-fold lower levels of brain PrPsen than did Prnp+/+ mice. Brain PrPsen levels in tg44+/+ and tg23+/+ mice were indistinguishable (data not shown).

Figure 1. Brain PrPsen expression levels.

C57BL/10 (Prnp+/+), Prnp+/−, expressing normal anchored mouse PrP, and transgenic mice (tg44+/+ and tg44+/−), expressing only anchorless mouse PrP were compared. Immunoblots were done using brain homogenates made as described in methods, and samples were serially diluted two-fold in sample buffer to give the mg brain equivalents shown on the figure. Bands were detected with monoclonal antibody D13. (A) No PNGase digestion. Lanes 1–3: Prnp+/+, Lanes 4–6: Prnp+/−, Lanes 7–9: tg44+/+ and Lanes 10–12: tg44+/−. (B) After PNGase F treatment. Lanes 1–3: Prnp+/−, Lanes 4–5: tg44+/+, Lanes 6–8: Prnp+/− and 9–11: tg44+/+ mice. Lanes 1–5 were from one experiment and 6–11 were from a separate experiment and results shown are for 4 different mice. Lower apparent molecular weight in PrP of transgenic mice is due to lack of the GPI anchor. Prnp+/− mice appeared to have 4-fold more brain PrPsen than tg44+/+ mice (compare lanes 2 and 4, also lanes 7 and 10). Data shown are for tg44+/+ and tg44+/− mice. By immunoblot PrPsen expression levels in tg23 mice were indistinguishable from those in tg44 mice (data not shown).

doi:10.1371/journal.ppat.1000800.g001

Scrapie-induced clinical disease in transgenic and non-transgenic mice

Scrapie-induced clinical disease was analyzed in transgenic and non-transgenic mice using two different scrapie strains, RML and 22L. After intracerebral (IC) inoculation of scrapie strain 22L, Prnp+/+ and +/− mice developed clinical scrapie at 150–165 dpi and 245–260 dpi respectively (Figure 2A). Similar results were also seen using the RML scrapie strain in Prnp+/+ mice (Figure 2B). Prnp+/− mice were not tested with the RML strain. In contrast to experiments with non-transgenic mice, all tg44+/+ and tg23+/+ mice infected with strains 22L or RML developed neurological signs and required euthanasia from 300 to 480 dpi (Figure 2). Homozygous transgenic mice differed from non-transgenic mice in incubation period, duration and progression of clinical signs, as well as gait and postural abnormalities (Table 1). The most obvious signs in homozygous transgenic mice were the presence of a wide-based gait, rear extremity weakness with low posture, and lack of kyphosis. The signs seen in homozygous transgenic mice differed from the narrow-based tippy-toed gait and frequent kyphosis seen in infected Prnp+/+ and +/− mice (Table 1). The gait and postural differences probably reflected different patterns of neurological damage. Overall the differences between non-transgenic mice and homozygous transgenic mice suggested that there might be different pathogenic mechanisms operating in these two scrapie-induced disease models.

Figure 2. Survival curves for scrapie-infected mice.

Transgenic mice (tg44+/+ and tg23+/+) expressing only anchorless PrP, C57BL/10 (Prnp+/+) and Prnp+/− mice, expressing normal anchored mouse PrP were compared. Mice were inoculated intracerebrally with 22L scrapie (panel A) and RML scrapie (panel B) and observed weekly for development of disease (see Table 1). Mice were euthanized when clinical signs were severe as described in Table 1. N values for each group are as follows: Panel A (22L): Prnp+/+, 11; Prnp+/−, 17; tg44+/+, 23; tg23+/+, 10. Panel B (RML): Prnp+/+, 8; tg44+/+, 21; tg23+/+, 9.

doi:10.1371/journal.ppat.1000800.g002

Table 1. Clinical aspects of disease in 22L or RML scrapie-infected C57BL/10 and tg44+/+ mice.

doi:10.1371/journal.ppat.1000800.t001

In our earlier studies, heterozygous tg44+/− and tg23+/− mice did not manifest the usual clinical signs of scrapie during 600 days of observation after infection with scrapie strains 22L or RML (Table 1) [19]. However, in the present experiments with the experience of observing the clinical signs described above in homozygous transgenic mice, we noted similar clinical signs in the infected heterozygous transgenic mice starting around 480 dpi. Clinical diagnosis in these mice was difficult due the erratic presence of signs in the initial stages, the longer duration of signs, and the possibility of confusion with signs of old age. These mice were euthanized between 480 and 700 dpi, but the indication for euthanasia was primarily debilitation (weight loss, dermatitis, bladder distention, cancer and infections), rather than neuromuscular dysfunction.

PrPres levels

PrPres deposition in brain is a major hallmark of prion diseases, and PrPres is often associated with areas of brain degeneration. Therefore brain PrPres levels in scrapie-infected mice were analyzed by immunoblot. As shown in Figure 3, tg44+/+ mice with clinical neurological disease had higher amounts of PrPres at 348–408 dpi than tg44+/− mice had at 567–594 dpi, which was when these mice had to be euthanized due to debilitating signs as described in Table 1. A similar difference was seen between tg23+/+ and tg23+/− mice (data not shown). The difference between tg44+/+ and tg44+/− mice in timing and levels of PrPres correlated with the higher PrPsen expression level seen in homozygous mice (Figure 1A), and appeared to explain the earlier onset and more prominent clinical signs seen in homozygous transgenic mice.

Figure 3. Detection of PrPres by immunoblot using monoclonal antibody D13.

Comparison of PrPres in brain of 22L scrapie-infected tg44+/+ and tg44+/− mice at the time of clinical disease. All lanes were loaded with 0.25 mg brain tissue equivalents. A clinical Prnp+/− mouse is shown for comparison. Lane 1: Prnp+/− (251 dpi), PrPres bands are seen at 21, 28, and 31 kD; lanes 2–5: tg44+/− mice (567, 589, 594, 594 dpi) and lanes 6–9 tg44+/+ (348, 365, 384, 408 dpi). PrPres bands are at 18 and 23 kD. The sizes are lower in these mice due to lack most carbohydrates and lack of GPI [19]. PrPres levels in tg44+/− mice were approximately 50% lower than in tg44+/+ mice.

doi:10.1371/journal.ppat.1000800.g003

The PrPres detected in these immunoblots had a molecular weight of 19 kD. In our earlier study this band was found to react with an anti-PrP peptide antibody (R20) directed at the C-terminal region of PrP (residues 218–232) [19]. Therefore, there was no evidence for loss of these C-terminal residues as often occurs in human GSS [6],[22].

Interestingly tg44+/+ and tg44+/− mice had higher PrPres levels than did non-transgenic Prnp+/+ or +/− mice, but the non-transgenic mice died earlier than the tg44+/+ mice (Figure 2). This suggested either that amyloid PrPres in tg44+/+ mice might be less pathogenic than the non-amyloid PrPres in non-transgenic mice, or that transgenic mice might be less susceptible to the pathogenic effects of PrPres amyloid due to the absence of membrane-anchored PrP.

Infectivity levels in tg44 mice

Previously we showed that scrapie-infected tg44+/− mice lacked signs of clinical scrapie but had infectivity titers in brain as high as 4.6×10⁸ ID50/gram brain at 120–286 dpi as measured by end-point titration in C57BL/10 mice [19]. In the present experiments we passaged brain from infected tg44+/− (passage 1) mice into other tg44+/− mice (passage 2), and at 512 and 531 dpi we found brain infectivity titers of 1.1–1.3×10¹⁰ ID50/gram brain (Table 2). These mice also had brain PrPres levels similar to those shown in other tg44+/− mice (Figure 3). Similar high titers were detected in passage 1 homozygous tg44+/+ mice at 384 dpi (Table 2). Thus in the transgenic anchorless PrP model, scrapie infectivity was present in brain at very high titers. Furthermore, the agent did not appear to develop new strain-like properties selective for tg44+/− mice as it could passage easily from tg44+/− mice to either C57BL/10 or tg44+/− mice.

Table 2. Infectivity titers of 22L scrapie from brains of Tg44 mice after 1 or 2 passages.

doi:10.1371/journal.ppat.1000800.t002

Histopathology and PrPres distribution detected by IHC

We compared the PrPres deposition and neuropathology after scrapie infection in transgenic tg44+/+ mice and non-transgenic C57BL/10 control mice (Table 3). Following scrapie infection in C57BL/10 mice typical TSE-specific diffuse deposits of PrPres were found in many brain areas (Figure 4A, 4B). This PrPres did not stain with the amyloid stain, Thioflavin S [19]. In many brain regions by H&E staining we observed gray matter spongiosis (Figure 4C), which is an important feature of TSE/prion diseases.

Figure 4. Light microscopic histopathology of scrapie-infected C57BL/10 and tg44+/+ transgenic mice.

(A) Whole brain sagittal section of C57BL/10 mouse infected IV with RML scrapie at 180 dpi showing wide distribution of diffuse PrPres stained with monoclonal antibody D13. (B) High power view of C57BL/10 mouse infected IC with RML scrapie at 157 dpi showing diffuse punctuate pattern of PrPres in hippocampus. (C) H&E stain of forebrain of C57BL/10 mouse infected IP with 22L scrapie at 375 dpi. Anterior commissure white matter (WM) is seen in lower right. Numerous scrapie vacuoles (arrows) are visible in surrounding gray matter(GM). (D) Whole brain saggital section of tg44+/+ mouse D554 infected IC with RML scrapie showing dense plaque-like PrPres stained with monoclonal antibody D13 at 341 dpi. PrPres was found in most CNS areas including cerebral cortex, corpus callosum, forebrain, hippocampus, thalamus, hypothalamus, midbrain, colliculi, brainstem, and spinal cord. Cerebellar involvement was minimal after RML infection, as shown in panel 4D, but was strong in cerebellar molecular layer, granular layer and meninges after 22L infection (not shown). (E) Higher power of panel D shows large dense PrP plaques surrounding dentate gyrus of hippocampus often in a perivascular distribution (arrows). Note difference compared to diffuse PrPres staining in panel B. (F) H&E stain of mouse D554 showing vacuoles in the white matter (WM) near the anterior commissure and no vacuoles in surrounding gray matter, i.e. opposite distribution of vacuoles compared to C57BL/10 mouse in panel C. (G) Astrogliosis seen by staining with anti-GFAP in dentate gyrus of mouse D554. (H) H&E stain of dentate gyrus of mouse D554 shows marked neuronal loss in lower arm of gyrus (box and arrow). Boxed outlines region shown in panel I. (I) High power view of lower arm of dentate gyrus outlined in panel H, shows multiple areas of neuronal loss (arrows). Plaques surround this area and one plaque is indicated with the arrowhead at the left. (J) High power view of area shown in panel I shows D13 staining of PrPres plaques impinging on damaged neurons of the dentate gyrus (arrow). Arrowhead shows plaque around blood vessel in upper left corner. (K) Deposition of amyloid precursor protein, APP, (red-brown stain) adjacent to area of neuronal loss (arrows) in dentate gyrus of same area shown in panels I and J. (L) Abnormal axonal proliferation shown by staining with anti-neurofilament protein in area of neuronal loss (arrows) in dentate gyrus of mouse D554. (M) Anti- neurofilament protein staining of dentate gyrus of uninfected control mouse shows no neuronal damage or abnormal axonal staining within the gyrus. Scale bars in panels B, E, G, and H are 100 microns; all other scale bars are 50 microns. Similar pathological changes were seen at the time of clinical disease in both tg44+/+ and tg23+/+ mice infected with either 22L or RML strains of scrapie. D13 staining of PrPres in panels E and J was similar to results described previously in tg44+/− and tg23+/− mice at ≥498 dpi [19].

doi:10.1371/journal.ppat.1000800.g004

Table 3. Comparison of histopathological features in scrapie-infected C57BL/10 mice and homozygous anchorless PrP transgenic mice (tg44+/+).

doi:10.1371/journal.ppat.1000800.t003

In scrapie-infected tg44+/+ mice PrPres accumulated as large dense plaque-like deposits, usually in a perivascular location around capillaries, veins and arteries in numerous brain regions, including leptomeninges, cerebral cortex, corpus callosum, forebrain, hippocampus, thalamus, hypothalamus, midbrain, colliculi, brainstem, and spinal cord (Figure 4D, 4E, 5A, 5B) After infection with the 22L scrapie strain, the cerebellar molecular layer and granular layer were also involved [19], but this was not seen after infection with the RML strain (Figure 4D). These deposits were Thioflavin S-positive [19], and no areas of diffuse non-amyloid PrPres were observed. The most distinguishing histopathological feature in tg44+/+ mice at the time of clinical signs was distortion of brain structures adjacent to large amyloid plaques in many areas (Figures 4D, E, H–L). These areas had intense micro- and astrogliosis (Figures 4G). Small blood vessels showed occasional micro-hemorrhages, or perivascular haemosiderin accumulation, but no lymphocyte infiltration of blood vessel walls was detected. Marked neuronal loss was seen around edges of some gray matter plaques (Figure 4H, I); however, no gray matter spongiosis typical of prion diseases was seen (Figure 4C vs 4F). In addition, tg44+/+ mice had focal areas of abnormal staining of amyloid precursor protein (APP) (Figure 4K), non-phosphorylated neurofilament protein (NFP) (Figure 4L), and phosphorylated NFP (not shown), all of which indicated a process of severe axonal dystrophy. These latter effects were rarely seen in scrapie-infected C57BL/10 mice (Table 3). These results suggested that scrapie-infected anchorless PrP transgenic mice had a different pathogenic process compared to non-transgenic C57BL/10 mice.

Ultrastructural analysis of brain from infected tg44+/+ mice

For higher resolution of details, scrapie-infected tg44+/+ mice were also studied using immunohistochemistry on 1 micron thick plastic-embedded sections as well as immunogold labeling at the ultrastructural level. Light microscopy on thin sections showed both perivascular and vascular PrPres labeling (Figure 5A), as well as occlusion of vessels in some cases (Figure 5B). By electron microscopy abundant PrPres labeling of blood vessels was seen most predominantly at basement membranes. In some larger vessels smooth muscle cells of the media were atrophied and replaced by extensive PrPres accumulation (Figure 5C). Vascular and plaque PrPres accumulation could be seen to be of a fibrillar amyloid nature at high magnification (Figure 5D). In smaller vessels PrPres was seen at both endothelial and pericyte basement membranes (Figure 5E). PrPres was also observed within the extracellular space along the borders of swollen astroglial and neurite processes in the absence of visible fibrillar amyloid (Figure 5F and G). No PrPres labeling was seen in uninfected control mice.

Figure 5. Immunological detection of PrPres in brain at both light and electron microscopic levels.

The 22L scrapie-infected anchorless PrP tg44+/+ mouse shown was clinically positive at 377 dpi. (A–B) Light microscopy of 1 µm thick plastic-embedded tissue labelled with monoclonal antibody 1A8. (A) shows intravascular and perivascular PrPres. In (B), the marked vascular amyloid infiltration is associated with occlusion of the vascular lumen (boxes indicate occluded lumens of two vessels). When these vessels were visualised in the electron microscope the smooth muscle of the vascular media was totally replaced by amyloid and an amorphous electron dense material filled the lumen (not shown). (C–F) Electron microscopy. (C) Low power view of large PrPres amyloid plaque adjacent to a small artery. Vessel lumen is in upper left corner and an endothelial cell with a prominent nucleus is to the right of the lumen. Silver enhanced gold-labeled PrPres is seen within the basement membrane (BM) and within the heavy amyloid accumulation which partially replaces the smooth muscle media layer. Amyloid bundles radiate away from the vessel and through the neuropil at the bottom right. (D) A high magnification illustration of (C) showing PrPres labelling on small bundles of amyloid fibrils at the periphery of the plaque. (E) Marked PrPres accumulation at the endothelial and pericyte basement membranes (arrows) and extending into narrow extracellular spces between nearby neurites and perivascular glial processes. Asterix (*) shows area of astrocytic cytoplasmic swelling. Lu; lumen. (F) Neuropil of cerebrum showing immunogold label for PrPres present over the extracellular spaces between neurites bounded by pairs of adjacent plasmalemmae. No visible amyloid fibrils were visible and the spaces between cellular processes were regular and even. This non-fibrillar PrPres labelling which dissects between neurite and glial cell profiles may extend over large area of neuropil as shown in (G). Asterix (*) on left side shows enlarged glial process with loss of normal cytoplasmic organelles. Similar findings were observed in tg44+/+, tg44+/− and tg23+/− mice. Tg23+/+ mice were not examined by electron microscopy. Scale bars: A and B, 20 µm; C and D, 1 µm; E and G, 2 µm; F, 500 nm.

doi:10.1371/journal.ppat.1000800.g005

Using staining with uranyl acetate/lead citrate large areas of distended swollen processes could be seen (Figure 6A), which were similar to the areas of immunogold-labeled PrPres shown above (Figure 5G). At higher magnification swollen perivascular glial processes were often seen (Figure 6B), and fibrils were visible in the endothelial basement membrane (Figure 6C) and/or pericyte basement membrane (Figure 6D). The initial site of aggregation into fibrils was in the ablumenal basement membranes (Figure 5D). Dystrophic neurites were also frequently noted in gray matter (Figure 6E). These were most conspicuous surrounding perivascular amyloid plaques and corresponded to sites of APP labeling. In white matter we observed degeneration of axons, including empty distended myelin sheaths (Figure 6F) (Table 3) which could be seen as white matter vacuoles by light microscopy (Figure 4F).

Figure 6. Ultrastructure of cerebral cortex and cerebellum of a 22L scrapie-infected tg44+/+ mouse.

Sections were from same mouse as shown in Figure 5, and were stained with uranyl acetate/lead citrate. (A) Area of neuropil with severe vacuolation where most vacuoles originate within processes and are separated from each other by intact membranes. (B) Several distended astrocytic processes (asterisks) of the perivascular glial limitans are present around a blood vessel Lu: lumen. (C) On higher magnification of boxed area from (B), the endothelial basement is shown to be filled with irregularly orientated amyloid fibrils (arrows). (D) The earliest stage of vascular amyloid is shown. Here the endothelial basement membrane (black arrowheads) is intact, but the pericyte basement membrane (black arrows) is thickened and heavily infiltrated with short amyloid fibrils (white arrowheads). (E) Severe neuritic dystrophy in which several processes show an excessive accumulation of organelles and abnormal electron dense bodies. (F) White matter of the cerebellum showing an empty myelin sheath (seen as vacuoles by light microscopy) and also a dark degenerate axon (asterisk) within an intact myelin sheath. Similar findings were observed in RML scrapie-infected tg44+/+ mice. Scale bars: A, 2 µm; B, E, F, 1 µm; C, 500 nm; D, 500 nm.

doi:10.1371/journal.ppat.1000800.g006

In contrast to transgenic mice, infected C57BL/10 mice at the time of clinical disease had numerous TSE vacuoles with broken or “hanging” membranes (not shown) [23] [24]. Such vacuoles were never seen in infected transgenic mice (Table 3). In C57BL/10 mice other ultrastructural hallmarks specific for classical prion diseases including membrane accumulation of disease-specific PrPres and TSE-specific membrane alterations were also seen, as reported previously in other prion disease models [17] [7],[18],[24] (Table 3). However, none of these prion disease-specific features was seen in infected tg44+/+ mice. The ultrastructural differences between scrapie-infected C57BL/10 and tg44+/+ mice supported the conclusion that the pathogenesis of disease in these transgenic mice was not typical TSE/prion disease.

Brain graft experiments

The reasons for the different types of scrapie-induced pathogenesis in C57BL/10 mice and anchorless PrP transgenic mice are not known. Two possibilities include: first, PrPres amyloid and diffuse non-amyloid PrPres might have different neurotoxic effects; second, PrPsen anchoring might influence neurotoxicity induced by infection.

To test whether PrPres derived from GPI-anchored PrPsen could induce gray matter vacuoles in tissue expressing anchorless PrPsen, brain tissue from C57BL/6 mice at embryonic day E12–E14, which expressed green fluorescent protein constitutively in all tissues [25], was grafted into the brain of adult tg44+/− mice or PrP null mice as controls [26]. One month after grafting, mice were infected IC with scrapie, and at 132–511 dpi the brain tissue was examined by histopathology. Recipients had from 1–6 detectable grafts per mouse (Table 4). Representative grafts are shown in Figure 7.

Figure 7. Detection of PrPres and vacuolation in brain tissue of PrPnull and tg44+/− mice with C57BL/6 brain grafts.

Grafts expressed green fluorescent protein (GFP), and mice were infected IC with 22L scrapie approximately 5 weeks after grafting. Panels A–C, PrPnull recipient at 261 dpi; Panels D–F, tg44+/− recipient at 261 dpi; Panels G–I, tg44+/− recipient at 200 dpi. Panels A, D, G show staining with anti- GFP which detects constitutive GFP expression in the C57BL/6 donor tissue. Panels B, E, H show D13 staining of PrPres. Panels C, F, I show H&E staining to detect scrapie-induced vacuoles indicated by arrows.

doi:10.1371/journal.ppat.1000800.g007

Table 4. Studies of PrPres and gray matter vacuolation in scrapie-infected anchorless PrP tg44+/− mice with C57BL/6 embryonic brain grafts.

doi:10.1371/journal.ppat.1000800.t004

At 261 dpi in the control PrPnull recipient, C57BL/6 graft tissue, identified by presence of green fluorescent protein (GFP) (Figure 7A), had easily detectable PrPres (Figure 7B) present both within the graft and at the interface between the graft and the host tissue, but PrPres did not appear to spread extensively into the PrPnull tissue. TSE gray matter vacuolation was seen only within the graft tissue (Figure 7B and 7C). This was similar to a previous report [26].

In tg44+/− recipient mice receiving C57BL/6 grafts, PrPres and gray matter vacuolation was also seen in the graft (Figure 7F and 7I). In the adjacent host tissue expressing anchorless PrP, amyloid PrPres and white matter vacuoles were noted; however, the C57BL/6 PrPres present at the edges of the graft appeared to be unable to induce gray matter vacuoles in the adjacent transgenic tissue (Figure 7F). In some cases the graft cells were not well-demarcated from the host (Figure 7G), and it was not clear whether the PrPres and vacuoles were in the graft or the host (Figure 7H and 7I). These results were representative of observations in 25 grafts in tg44+/− recipients where PrPres was detected in the graft (Table 4). In summary, we found no grafts where expression of anchored PrPres from the C57BL/6 graft could be associated with gray matter spongiosis in adjacent transgenic host tissue. This result suggested that expression of anchored PrPsen in gray matter might be a fundamental requirement for the induction of the typical TSE/prion disease pathogenic process.

Discussion Top

In the present experiments scrapie infection of transgenic mice expressing anchorless PrP resulted in a slow fatal brain disease. These results demonstrated new mechanisms of prion-induced pathogenesis associated with the presence of PrPres amyloid and the absence of GPI-anchored PrP. This disease lacked gray matter spongiosis and differed in this respect from scrapie infection in non-transgenic mice, where the disease is characterized by extensive gray matter spongiosis and non-amyloid PrPres deposition.

The current results raised the question of how lack of GPI-linked membrane anchoring of PrP might facilitate formation of PrPres amyloid. GPI anchorless PrP has a longer biological half-life [27] and is secreted by the cell. Both of these attributes might allow more effective and extensive interactions between soluble PrP molecules. In addition, the minimal amount of carbohydrates and the absence of the GPI group on anchorless PrP might favor amyloidogenic hydrophobic protein-protein interactions, particularly at a time of partial protein unfolding during PrP conversion. These features of anchorless PrP are likely to contribute to its enhanced tendency to form amyloid during conversion to PrPres. Anchorless protease-resistant PrP, cleaved at residue 228, comprises 15% of the PrPres in hamster scrapie brain extracts [28], but it is unclear whether this material contributes to the amyloid PrP seen in this model.

Our results differed from those of two interesting mouse prion disease models where PrPres was also found almost entirely in an amyloid form. In the GSS PrP-8kd model [29] and the G3-ME7 model [30], which both used PrP mutant mice, PrP amyloid was seen primarily in the corpus callosum, but did not spread significantly to other brain regions. There was no clinical disease in these models, and transmission experiments suggested very low infectivity titers in the GSS PrP-8kd model. Compared to these two models, the three main distinguishing features of the anchorless PrP model are the ability of the PrPres amyloid to accumulate widely throughout the brain (Figure 4), the resulting fatal brain disease (Figure 2), and the high titer of transmissible agent (Table 2) [31].

The association of amyloid deposition without gray matter spongiosis in our system is reminiscent of the neuropathology seen in certain human familial prion diseases. For example, GSS patients with PrP mutations Y145Stop and Y163Stop had both CAA and parenchymal perivascular amyloid without gray matter spongiosis [32] [33]. Both these mutations result in C-terminally truncated PrP lacking the GPI anchor. Parenchymal amyloid deposition without gray matter spongiosis has also been seen in GSS patients with several other PrP mutations including P102L, P105L, A117V and F198S [6]. Recently two human GSS patients with new PrP mutations producing nonsense codons at positions 226 and 227 were described [34]. Both patients had widespread PrPres amyloid deposition in the absence of gray matter spongiosis, and one had CAA. These patients expressed a nearly full-length form of PrP lacking 6–7 C-terminal residues and the GPI anchor, which was quite similar to the PrP expressed in our anchorless PrP tg mice.

In many GSS patients, amyloid PrPres purified from brain was truncated resulting in a 7–11 kDa protease-resistant fragment from the central region of PrP (approximately residues 81–150)[6],[22],[34]. Interestingly, presence of this truncation has been correlated with the lack of gray matter spongiosis [8],[9]. In contrast, based on previous immunoblot studies, the proteinase K-resistant PrPres amyloid in our model appeared to contain residues 88–231 [19], which was similar to the PrPres found in human and animal prion diseases with extensive gray matter spongiosis. Furthermore, PrPres in tissue sections could be stained with anti-PrP serum R24, specific to residues 23–37 (data not shown) suggesting that there was no significant truncation at the N-terminus beyond the signal peptide. Thus, lack of spongiosis in our model appeared dependent on the absence of GPI-anchoring rather than truncation of the PrPres.

Two possibilities might explain the correlation between lack of GPI- anchored PrP and lack of gray matter spongiosis in our infected transgenic mice: (1) anchorless amyloid PrPres might be less neurotoxic than diffuse PrPres, and/or (2) anchored PrPsen might be required for PrPres-mediated neurotoxic membrane interactions. The former explanation could not be proven or excluded by our results. However, the latter interpretation was supported by data from brain graft experiments. After scrapie infection of tg44+/− mice grafted with C57BL/6 brain expressing normal anchored PrPsen, we observed gray matter spongiosis and non-amyloid PrPres deposition in C57BL/6 grafts, but not in adjacent host tissue expressing only anchorless PrPsen. Tissue expressing only anchorless PrPsen appeared to be unable to respond to the presence of GPI-anchored PrPres produced in the nearby grafts, and no gray matter spongiosis was produced. Therefore, lack of anchored PrPsen might by itself explain the lack of gray matter spongiosis in transgenic mice.

However, even in the absence of anchored PrPsen, the amyloid PrPres was able to induce additional pathogenic processes capable of causing fatal neurological disease. By both light and electron microscopy we observed evidence for three distinct pathogenic processes not seen in typical prion disease in C57BL/10 mice (Box 1):

(1) Brain damage caused by tissue distortion by large amyloid plaques. These plaques were associated with neuronal loss, axonal pathology and gliosis (Figure 4E, H–L Figure 6A,E,F). The more rapid accumulation of PrPres in tg44+/+ mice compared to tg44+/− mice (Figure 3) suggested a faster growth of large space-occupying plaques which might explain in part the clinical neurological signs leading to death of tg44+/+ and tg23+/+ mice (Figure 2).

(2) A second pathogenic process in scrapie-infected transgenic mice was suggested by ultrastructural studies finding that the early aggregation of PrPres into fibrillar amyloid was located at or within vascular basement membranes (Figures 5C, 5D, 6C, 6D). This was associated with vascular damage including occlusion (Figure 5B), amyloid replacement of basement membrane and tunica media, and occasional micro-hemorrhages. This pathology was similar to that observed in CAA seen in Alzheimer’s disease and several familial amyloid diseases including two prion diseases [14],[15],[32],[33].

(3) Evidence of a third pathogenic process in the transgenic mice was suggested by finding of small deposits of immunogold-labeled PrPres at the ultrastructural level in the extracellular spaces between glial and neuritic processes in gray matter (Figures 5D, 5E, 5F). These PrPres deposits were small, and there was no distortion of the extracellular space or visible aggregation into amyloid fibrils. However, the adjacent processes were often highly dystrophic (Figure 6E) or swollen and devoid of organelles, and they appeared to coalesce to form empty spaces larger than the original processes (Figures 5G, 6A, 6B). These abnormal areas, which were also noted in heterozygous tg44+/− and tg23+/− mice, appeared to represent a form of damage related to small, rather than large, PrPres deposits, and they did not require the presence of anchored PrPsen for their formation.

The early localization of PrPres at basement membranes (Figures 5A, C, D), suggested that the PrP conversion process might initiate at these sites, and implied that basement membrane molecules might facilitate PrP conversion. For example, basement membrane might filter or trap soluble PrPsen molecules or small PrPres oligomers from the extracellular interstitial fluid of brain increasing their local concentration, thus favoring conversion to larger PrPres amyloid aggregates. Serum amyloid P-component which binds to all amyloids and is a constituent of basement membranes might also contribute to local PrP conversion [35]. In addition, collagen, laminin and heparin sulfate-containing proteoglycans are major components of basement membranes, and PrP can bind to both the laminin receptor and heparan sulfate which can associate directly or indirectly with PrP [36]–[38]. Heparan sulfate and other glycosaminoglycan (GAG) moieties can delay scrapie disease in vivo [39]–[42] [43],[44], and some GAG molecules can alter PrP conversion in vitro [45]. A scaffolding mechanism might account for this effect. For example, soluble anchorless PrPsen monomers might be held in place by GAG polymers to increase local concentration and facilitate conversion by PrPres, analogous to the tethering of anchored PrPsen on cell membranes [46],[47]. In addition, attachment of small mobile PrPres oligomers to GAG polymers might assist conversion at the basement membrane. Subsequently newly formed larger less mobile PrPres could serve as an efficient scaffold for further conversion allowing the process to extend out into the brain parenchyma. Eventually this process might form very large PrPres amyloid plaques with blood vessels at the center as we observed (Figure 4E and 5A).

The vascular amyloid pathology seen in our scrapie-infected transgenic mice (Figures 4E, 5A–D, 6B–D) was similar to CAA seen in Alzheimer’s disease as well as several familial amyloid diseases [14], including two forms of familial prion disease [32] [14]. In Alzheimer’s disease, amyloid fibrils within vascular basement membranes are thought to impede interstitial fluid drainage leading to an increase in Aβ concentrations within the extracellular space. Such increased soluble Aβ and oligomeric proto-amyloid fragments are considered a likely contributory factor in the cognitive decline of Alzheimer’s disease patients [15]. Similar processes might contribute to the clinical disease seen in the anchorless PrP scrapie model. Since all these diseases with CAA show amyloid localization with basement membranes, drugs capable of blocking amyloid-basement membrane interactions might be effective treatments for some of these diseases. In the case of prion diseases, one such compound, pentosan polysulfate, a small GAG oligomer, was effective in blocking PrPres generation in an infected cell line [48] and delayed onset of clinical scrapie in vivo [41] [42] [44]. Similarly a decoy molecule preventing PrP interaction with the laminin receptor (LRP/LR) reduced PrPres levels and delayed disease in vivo [37]. Determining the precise glycans and proteins involved in the protein interactions leading to amyloid deposition in all the CAA diseases might be important in designing new therapeutic approaches.

Methods Top

Mice

Ethics statement: All mice were housed at the Rocky Mountain Laboratories (RML) in an AAALAC-accredited facility, and research protocols and experimentation were approved by the NIH RML Animal Care and Use Committee.

C57BL/10SnJ mice (Prnp+/+) were obtained from Jackson Laboratories (Bar Harbor, Maine). C57BL/10SnJ PrP−/− mice were created at RML by crossing 129/Ola PrP−/− mice [49] with C57BL/10SnJ mice, followed by nine serial backcrosses to C57BL/10SnJ with selection for the Prnp+/− genotype using previously described PCR reactions to detect both the Prnp+ and Prnp null alleles [19]. One intercross was then done, and C57BL/10SnJ Prnp−/− (PrP−/−) mice were selected and interbred. Heterozygous Prnp+/− mice were obtained by intercrossing C57BL/10 (Prnp+/+) mice with C57BL/10 Prnp−/− mice.

Transgenic GPI anchorless PrP mice (tg44+/− and tg23+/−) were made as described previously [19] and then backcrossed to C57BL/10SnJ-Prnp−/− mice for six to nine generations with selection for the Prnp−/− genotype and the tg44 or 23+/− genotype. Thus these mice contained one anchorless PrP transgene allele and did not express any normal anchored mouse PrP allele. Heterozygous transgene lines tg23+/− and tg44+/− were each interbred to create homozygous lines (tg44+/+ and tg23+/+). Offspring were tested for transgene zygosity using real-time DNA PCR on an ABI Prism 7900 HT Sequence detection system and SDS 2.2.2 software. The following probes and primers were designed to amplify the mouse Prnp sequence: probe (moPrPlower418T): (5′-CGGTCCTCCCAGTCGTTGCCAAA), forward primer (moPrP-396F): (5′-CGTGAGCAGGCCCATGATC), reverse primer (moPrP-465R): (5′GCGGTACATGTTTTCACGGTAGT). Individual mice identified by rtPCR as transgene homozygous were then bred to Prnp−/− mice to confirm homozygosity. Homozygous mice were then interbred to create additional mice for experimentation. Both tg44 and tg23 lines were used in the present experiments to demonstrate that the observed findings were consistent with transgene expression rather than a result of an integration site artifact.

Scrapie infections

Four to six week old mice were inoculated intracerebrally with 50 µl of a 1% brain homogenate of 22L or RML scrapie containing 0.7–1.0×10⁶ ID₅₀. One ID₅₀ is the dose causing infection in 50% of C57BL/10 mice. Animals were observed daily for onset and progression of scrapie. Mice were euthanized when clinical signs were consistent and progressive. Signs differed somewhat in C57BL/10 and tg44+/+ and tg23+/+ mice (Table 1). In heterozygous tg44+/− and tg23+/− mice many mice developed signs of debilitation such as weight loss, dermatitis and infections requiring euthanasia prior to severe neurological signs.

Immunoblotting

For detection of PrPsen from uninfected brains, tissues were homogenized (20% w/v) using a bead beater in ice-cold 0.01 M Tris-HCl pH 7.6 containing protease inhibitors (10 µM leupeptin, 1 µM pepstatin, and 1 µM aprotinin). Each sample was vortexed for 1 minute followed by sonication for 1 minute. Insoluble debris was removed by centrifugation at 2700 g for 10 minutes at 4°C. Samples were mixed 1:1 with 2X SDS-PAGE sample buffer and boiled for 3–5 minutes. PNGase F reactions were done using 4.4 mg tissue equivalents in a total volume of 20 µl SDS-PAGE sample buffer [50]. Samples were serially diluted two-fold in sample buffer to give the amount of brain tissue (mg brain equivalents) indicated for each lane. Immunoblots were probed by using monoclonal anti-PrP D13 at a dilution of 1:5000 (InPro Biotechnology, South San Francisco, CA), followed by secondary antibody sheep anti-human Ig (dilution 1:5000) (GE Healthcare, formerly Amersham Biosciences, Piscataway, NJ) and enhanced chemiluminescence according to the manufacturers instructions (Amersham-Pharmacia, Uppsala, Sweden).

For detection of PrPres either with or without PNGase F, samples were prepared as described [51]. Blots were probed as described above.

Brain grafting

Embryonic brain tissue was obtained from E12–E14 C57BL/6 embryos which expressed green fluorescent protein (GFP) in all tissues [25]. Mice were purchased originally from Jackson laboratories and were bred at Rocky Mountain Laboratories by Dr. Kim Hasenkrug. Pregnant mothers were euthanized and embryos dissected with forceps in media under a dissecting microscope to obtain the mesencephalon and telencephalon. Tissue was partially disrupted by pipetting to generate small fragments. This suspension (30 µl) was inoculated intracerebrally through the skull into the parietal brain region of 3–4 week old PrPnull mice or tg44+/− mice. One month later recipient mice were infected intracerebrally with 22L scrapie as described above. At various times thereafter mice were euthanized and brain tissue was examined histologically for GFP and PrPres by specific immunohistochemistry and for typical scrapie-induced gray matter spongiosis by H&E staining.

Histopathology and immunohistochemistry

Mice were euthanized and brains were placed in 3.7% phosphate-buffered formalin for 3 to 5 days before dehydration and embedding in paraffin. Serial 4 µm sections were cut using a standard Leica microtome, placed on positively charged glass slides and dried overnight at 56°C. Slides were stained with a standard protocol of hematoxylin and eosin (H&E) for observation of overall pathology. For PrPres detection, slides were rehydrated in 0.1 M citrate buffer, pH 6.0 and then heated at 120°C, 20 psi for 20 minutes in a decloaking chamber (Biocare, Walnut Creek, CA). Immunohistochemical staining was performed using the Ventana automated Nexus stainer (Ventana, Tucson, AZ). Staining for PrP used a standard avidin-biotin complex immunoperoxidase protocol using anti-PrP antibody D13 (In-Pro Biotechnology, South San Francisco, CA) at a dilution of 1:500 and incubated at 4°C for 16 hours. Biotinylated goat anti-human IgG (Jackson Immuno Research, West Grove, PA) was used at a 1:500 dilution as the secondary antibody. Detection was performed with Ventana streptavidin-alkaline phosphatase with Fast Red chromogen. Tissue sections for microglia staining were pretreated and stained with anti-Iba1 as described [52] except that detection was done using the Ventana Fast Red chromagen as above. Astroglia were stained with anti-GFAP as described [52], and detection was completed with Ventana streptavidin-alkaline phosphatase using Fast Red. Tissue sections for staining with anti-amyloid precursor protein (APP) were pretreated as described for anti-PrP antibody D13. Anti-APP (Zymed Laboratories, San Francisco, CA) was used at a 1:500 dilution followed by a 1:250 dilution of biotinylated-goat anti-rabbit IgG (Vector Laboratories, Burlington, CA), and detection with Ventana streptavidin-horseradish peroxidase plus amino ethyl carbazol (AEC) chromagen. Staining of phosphorylated neurofilament proteins was performed using a monoclonal antibody cocktail pan-axonal neurofilament marker SMI-312 (Covance, Princeton, NJ) at a 1:250 dilution. Monoclonal antibody to nonphosphorylated neurofilament proteins was also used (SMI-311). Primary antibodies were followed by biotinylated horse anti-mouse IgG secondary antibody at a 1:250 dilution. Ventana AEC reagent was used for detection. Green fluorescent protein (GFP) was detected using a mixture of two mouse anti-GFP monoclonal antibodies (clones 7.1 and 13.1) at dilution of 1:200 (Roche Applied Science, Indianapolis, IN), followed by biotinylated horse anti-mouse IgG (Vector Laboratories, Burlington, CA) at a dilution of 1:250 and detected with AEC chromogen (Ventana) as described above. All histopathology slides were read using an Olympus BX51 microscope and images were obtained using Microsuite FIVE software.

Perfusion/Processing for electron microscopy

Mice were perfused with fixative containing 3% paraformaldehyde and 1% glutaraldehyde in PBS. Excised tissues were then immersed in this fixative and held overnight at 4 degrees C. Tissue pieces were processed further using a Lynx® automated tissue processor with agitation as follows: one wash in PBS for 3 hr at 20 degrees, one wash in 0.1 M sodium phosphate buffer pH 7.2 at 20 degrees for 4 hr, post-fix in 2% osmium tetroxide in phosphate buffer at 20 degrees for 6 hr, one wash in phosphate buffer at 20 degrees for 3 hr, three washes in water at 20 degrees for 3 hr each, in-block staining with 1% uranyl acetate in water at 20 degrees for 6 hr, 3 washes in water at 20 degrees for 3 hr each, dehydration in 70%, 100%, and 100% acetone at 10 degrees for 3 hr each, and infiltration at 20 degrees in Araldite resin (Structure Probe, Inc., West Chester, PA) at 50% for 8 hr, 75% for 12 hr, and two changes of 100% for 20 hr each. Further tissue blocks were processed using a Leica EM TP processor using the procedure above with the omission of the uranyl acetate. Tissue blocks were then transferred to fresh resin in molds and polymerized at 65 degrees for 24 to 48 hr.

Immunolabelling of resin block sections for light microscopy

Thick (1 µm) sections were stained by toluidine blue or were immunolabelled using the avidin-biotin technique. Sections were deplasticized with saturated sodium ethoxide for up to 30 minutes. Endogenous peroxidase was blocked and sections were de-osmicated with 6% hydrogen peroxide for 10 minutes, followed by pre-treatment with neat formic acid for 5 minutes. Normal serum was then applied for 1 hour to block non-specific labeling. 1A8 anti-PrP serum [53] at a dilution of 1:6000, or pre-immune serum were then applied for 15 hours and reaction product developed using 3-3′ diaminobenzidine.

Immunolabeling for electron microscopy

For routine electron microscopy areas were selected from 1 µm thick toluidine blue stained sections and counterstained with uranyl acetate and lead citrate. For ultrastructural immunohistochemistry, serial 65 nm sections were taken from blocks previously identified from immuno-labeled 1 µm thick sections as described above. The 65 nm sections were placed on 600 mesh gold grids and etched in sodium periodate for 60 minutes. Endogenous peroxidase was blocked and sections de-osmicated with 6% hydrogen peroxide in water for 10 minutes followed by enhancement of antigen expression with formic acid for 10 minutes. Residual aldehyde groups were quenched with 0.2 M glycine in PBS, pH 7.4 for 3 minutes. Preimmune serum or anti-PrP primary antibody 1A8 [53] or R30 [54] at a 1:500 or 1:1500 dilution respectively in incubation buffer were then applied for 15 hours. After rinsing extensively, sections were incubated with Auroprobe 1 nm colloidal gold diluted 1:50 in incubation buffer for 2 hours. Sections were then post-fixed with 2.5% glutaraldehyde in PBS and labeling enhanced with Goldenhance (Universal Biologicals, Cambridge, UK) for 10 minutes. Grids were counterstained with uranyl acetate and lead citrate.

Acknowledgments Top

The authors thank Drs. Karin Peterson, Suzette Priola and Byron Caughey for helpful suggestions concerning the manuscript, Lynne Raymond for assistance in breeding the transgenic mice, and Ed Schreckengust for animal husbandry.

References Top

1. Aguzzi A, Polymenidou M (2004) Mammalian prion biology: one century of evolving concepts. Cell 116: 313–327. Find this article online
   
2. Di Bari MA, Chianini F, Vaccari G, Esposito E, Conte M, et al. (2008) The bank vole (Myodes glareolus) as a sensitive bioassay for sheep scrapie. J Gen Virol 89: 2975–2985. Find this article online
   
3. Nonno R, Di Bari MA, Cardone F, Vaccari G, Fazzi P, et al. (2006) Efficient transmission and characterization of Creutzfeldt-Jakob disease strains in bank voles. PLoS Pathog 2: e12. doi:10.1371/journal.ppat.0020012.
   
4. Jeffrey M, Goodsir CM, Bruce ME, McBride PA, Fraser JR (1997) In vivo toxicity of prion protein in murine scrapie: ultrastructural and immunogold studies. Neuropathol Appl Neurobiol 23: 93–101. Find this article online
   
5. Gonzalez L, Martin S, Begara-McGorum I, Hunter N, Houston F, et al. (2002) Effects of agent strain and host genotype on PrP accumulation in the brain of sheep naturally and experimentally affected with scrapie. J Comp Pathol 126: 17–29. Find this article online
   
6. Ghetti B, Piccardo P, Frangione B, Bugiani O, Giaccone G, et al. (1996) Prion protein amyloidosis. Brain Pathol 6: 127–145. Find this article online
   
7. Ersdal C, Goodsir CM, Simmons MM, McGovern G, Jeffrey M (2009) Abnormal prion protein is associated with changes of plasma membranes and endocytosis in bovine spongiform encephalopathy (BSE)-affected cattle brains. Neuropathol Appl Neurobiol 35: 259–271. Find this article online
   
8. Parchi P, Chen SG, Brown P, Zou W, Capellari S, et al. (1998) Different patterns of truncated prion protein fragments correlate with distinct phenotypes in P102L Gerstmann-Straussler-Scheinker disease. Proc Natl Acad Sci U S A 95: 8322–8327. Find this article online
   
9. Piccardo P, Dlouhy SR, Lievens PM, Young K, Bird TD, et al. (1998) Phenotypic variability of Gerstmann-Straussler-Scheinker disease is associated with prion protein heterogeneity. J Neuropathol Exp Neurol 57: 979–988. Find this article online
   
10. Piccardo P, Seiler C, Dlouhy SR, Young K, Farlow MR, et al. (1996) Proteinase-K-resistant prion protein isoforms in Gerstmann-Straussler-Scheinker disease (Indiana kindred). J Neuropathol Exp Neurol 55: 1157–1163. Find this article online
    
11. Giaccone G, Verga L, Bugiani O, Frangione B, Serban D, et al. (1992) Prion protein preamyloid and amyloid deposits in Gerstmann- Straussler- Scheinker disease, Indiana kindred [published erratum appears in Proc Natl Acad Sci U S A 1993 Jan 1;90(1):302]. Proc Natl Acad Sci U S A 89: 9349–9353. Find this article online
    
12. Bruce ME, Dickinson AG (1985) Genetic control of amyloid plaque production and incubation period in scrapie-infected mice. J Neuropathol Exp Neurol 44: 285–294. Find this article online
    
13. Jeffrey M, Goodsir CM, Bruce ME, McBride PA, Scott JR, et al. (1992) Infection specific prion protein (PrP) accumulates on neuronal plasmalemma in scrapie infected mice. Neurosci Lett 147: 106–109. Find this article online
    
14. Revesz T, Ghiso J, Lashley T, Plant G, Rostagno A, et al. (2003) Cerebral amyloid angiopathies: a pathologic, biochemical, and genetic view. J Neuropathol Exp Neurol 62: 885–898. Find this article online
    
15. Weller RO, Subash M, Preston SD, Mazanti I, Carare RO (2008) Perivascular drainage of amyloid-beta peptides from the brain and its failure in cerebral amyloid angiopathy and Alzheimer’s disease. Brain Pathol 18: 253–266. Find this article online
    
16. Stahl N, Borchelt DR, Prusiner SB (1990) Differential release of cellular and scrapie prion proteins from cellular membranes by phosphatidylinositol-specific phospholipase C. Biochemistry 29: 5405–5412. Find this article online
    
17. Jeffrey M, Goodsir CM, Bruce M, McBride PA, Scott JR, et al. (1994) Correlative light and electron microscopy studies of PrP localisation in 87V scrapie. Brain Res 656: 329–343. Find this article online
    
18. Jeffrey M, McGovern G, Goodsir CM, Siso S, Gonzalez L (2009) Strain-associated variations in abnormal PrP trafficking of sheep scrapie. Brain Pathol 19: 1–11. Find this article online
    
19. Chesebro B, Trifilo M, Race R, Meade-White K, Teng C, et al. (2005) Anchorless prion protein results in infectious amyloid disease without clinical scrapie. Science 308: 1435–1439. Find this article online
    
20. Prusiner SB, Scott M, Foster D, Pan KM, Groth D, et al. (1990) Transgenetic studies implicate interactions between homologous PrP isoforms in scrapie prion replication. Cell 63: 673–686. Find this article online
    
21. Manson JC, Clarke AR, McBride PA, McConnell I, Hope J (1994) PrP gene dosage determines the timing but not the final intensity or distribution of lesions in scrapie pathology. Neurodegen 3: 331–340. Find this article online
    
22. Tagliavini F, Prelli F, Ghiso J, Bugiani O, Serban D, et al. (1991) Amyloid protein of Gerstmann-Straussler-Scheinker disease (Indiana kindred) is an 11 kd fragment of prion protein with an N-terminal glycine at codon 58. EMBO J 10: 513–519. Find this article online
    
23. Jeffrey M, Scott JR, Fraser H (1991) Scrapie inoculation of mice: light and electron microscopy of the superior colliculi. Acta Neuropathol 81: 562–571. Find this article online
    
24. Jeffrey M, Goodsir CM, Race RE, Chesebro B (2004) Scrapie-specific neuronal lesions are independent of neuronal PrP expression. Ann Neurol 55: 781–792. Find this article online
    
25. Okabe M, Ikawa M, Kominami K, Nakanishi T, Nishimune Y (1997) ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett 407: 313–319. Find this article online
    
26. Brandner S, Isenmann S, Raeber A, Fischer M, Sailer A, et al. (1996) Normal host prion protein necessary for scrapie-induced neurotoxicity. Nature 379: 339–343. Find this article online
    
27. Campana V, Caputo A, Sarnataro D, Paladino S, Tivodar S, et al. (2007) Characterization of the properties and trafficking of an anchorless form of the prion protein. J Biol Chem 282: 22747–22756. Find this article online
    
28. Stahl N, Baldwin MA, Burlingame AL, Prusiner SB (1990) Identification of glycoinositol phospholipid linked and truncated forms of the scrapie prion protein. Biochemistry 29: 8879–8884. Find this article online
    
29. Piccardo P, Manson JC, King D, Ghetti B, Barron RM (2007) Accumulation of prion protein in the brain that is not associated with transmissible disease. Proc Natl Acad Sci U S A 104: 4712–4717. Find this article online
    
30. Tuzi NL, Cancellotti E, Baybutt H, Blackford L, Bradford B, et al. (2008) Host PrP glycosylation: a major factor determining the outcome of prion infection. PLoS Biol 6: e100. doi:10.1371/journal.pbio.0060100.
    
31. Race B, Meade-White K, Oldstone MB, Race R, Chesebro B (2008) Detection of prion infectivity in fat tissues of scrapie-infected mice. PLoS Pathog 4: e1000232. doi:10.1371/journal.ppat.1000232.
    
32. Ghetti B, Piccardo P, Spillantini MG, Ichimiya Y, Porro M, et al. (1996) Vascular variant of prion protein cerebral amyloidosis with tau-positive neurofibrillary tangles: the phenotype of the stop codon 145 mutation in PRNP. Proc Natl Acad Sci U S A 93: 744–748. Find this article online
    
33. Revesz T, Holton JL, Lashley T, Plant G, Frangione B, et al. (2009) Genetics and molecular pathogenesis of sporadic and hereditary cerebral amyloid angiopathies. Acta Neuropathol 118: 115–130. Find this article online
    
34. Jansen C, Parchi P, Capellari S, Vermeij AJ, Corrado P, et al. (2009) Prion protein amyloidosis with divergent phenotype associated with two novel nonsense mutations in PRNP. Acta Neuropathol. Find this article online
    
35. Dyck RF, Lockwood CM, Kershaw M, McHugh N, Duance VC, et al. (1980) Amyloid P-component is a constituent of normal human glomerular basement membrane. J Exp Med 152: 1162–1174. Find this article online
    
36. Pflanz H, Vana K, Mitteregger G, Pace C, Messow D, et al. (2009) Microinjection of lentiviral vectors expressing small interfering RNAs directed against laminin receptor precursor mRNA prolongs the pre-clinical phase in scrapie-infected mice. J Gen Virol 90: 269–274. Find this article online
    
37. Pflanz H, Vana K, Mitteregger G, Renner-Muller I, Pace C, et al. (2009) Scrapie-infected transgenic mice expressing a laminin receptor decoy mutant reveal a prolonged incubation time associated with low levels of PrPres. J Mol Biol 388: 721–729. Find this article online
    
38. Caughey B, Brown K, Raymond GJ, Katzenstien GE, Thresher W (1994) Binding of the protease-sensitive form of PrP (prion protein) to sulfated glycosaminoglycan and Congo red. J Virol 68: 2135–2141. Find this article online
    
39. Kimberlin RH, Walker CA (1986) Suppression of scrapie infection in mice by heteropolyanion 23, dextran sulfate, and some other polyanions. Antimicrob Agents Chemother 30: 409–413. Find this article online
    
40. Ladogana A, Casaccia P, Ingrosso L, Cibati M, Salvatore M, et al. (1992) Sulphate polyanions prolong the incubation period of scrapie-infected hamsters. J Gen Virol 73: 661–665. Find this article online
    
41. Diringer H, Ehlers B (1991) Chemoprophylaxis of scrapie in mice. J Gen Virol 72: 457–460. Find this article online
    
42. Doh-ura K, Ishikawa K, Murakami-Kubo I, Sasaki K, Mohri S, et al. (2004) Treatment of Transmissible Spongiform Encephalopathy by Intraventricular Drug Infusion in Animal Models. J Virol 78: 4999–5006. Find this article online
    
43. Farquhar CF, Dickinson AG (1986) Prolongation of scrapie incubation period by an injection of dextran sulphate 500 within the month before or after infection. J Gen Virol 67: 463–473. Find this article online
    
44. Larramendy-Gozalo C, Barret A, Daudigeos E, Mathieu E, Antonangeli L, et al. (2007) Comparison of CR36, a new heparan mimetic, and pentosan polysulfate in the treatment of prion diseases. J Gen Virol 88: 1062–1067. Find this article online
    
45. Wong C, Xiong L-W, Horiuchi M, Raymond LD, Wehrly K, et al. (2001) Sulfated glycans and elevated temperature stimulate PrP^Sc dependent cell-free formation of protease-resistant prion protein. EMBO J 20: 377–386. Find this article online
    
46. Baron GS, Wehrly K, Dorward DW, Chesebro B, Caughey B (2002) Conversion of raft associated prion protein to the protease-resistant state requires insertion of PrP-res (PrP(Sc)) into contiguous membranes. EMBO J 21: 1031–1040. Find this article online
    
47. Baron GS, Caughey B (2003) Effect of glycosylphosphatidylinositol anchor-dependent and – independent prion protein association with model raft membranes on conversion to the protease-resistant Isoform. J Biol Chem 278: 14883–14892. Find this article online
    
48. Caughey B, Raymond GJ (1993) Sulfated polyanion inhibition of scrapie-associated PrP accumulation in cultured cells. J Virol 67: 643–650. Find this article online
    
49. Manson JC, Clarke AR, Hooper ML, Aitchison L, McConnell I, et al. (1994) 129/Ola mice carrying a null mutation in PrP that abolishes mRNA production are developmentally normal. Mol Neurobiol 8: 121–127. Find this article online
    
50. Race BL, Meade-White KD, Ward A, Jewell J, Miller MW, et al. (2007) Levels of abnormal prion protein in deer and elk with chronic wasting disease. Emerg Infect Dis 13: 824–830. Find this article online
    
51. Meade-White K, Race B, Trifilo M, Bossers A, Favara C, et al. (2007) Resistance to chronic wasting disease in transgenic mice expressing a naturally occurring allelic variant of deer prion protein. J Virol 81: 4533–4539. Find this article online
    
52. Kercher L, Favara C, Striebel JF, LaCasse R, Chesebro B (2007) Prion protein expression differences in microglia and astroglia influence scrapie-induced neurodegeneration in the retina and brain of transgenic mice. J Virol 81: 10340–10351. Find this article online
    
53. Jeffrey M, Goodsir CM, Fowler N, Hope J, Bruce ME, et al. (1996) Ultrastructural Immuno-localization of Synthetic Prion Protein Peptide Antibodies in 87V Murine Scrapie. Neurodegen 5: 101–109. Find this article online
    
54. Caughey B, Raymond GJ, Ernst D, Race RE (1991) N-terminal truncation of the scrapie-associated form of PrP by lysosomal protease(s): implications regarding the site of conversion of PrP to the protease-resistant state. J Virol 65: 6597–6603. Find this article online

Introduction Top

Creutzfeldt-Jakob (CJD) disease of humans, bovine spongiform encephalopathy of cattle and scrapie of sheep are all examples of prion diseases. These diseases propagate through the misfolding of the normal cellular form of the prion protein (PrP^C) into the disease-associated isoform (PrP^Sc) [1]. The conversion of PrP^C to PrP^Sc is accompanied by a large increase in the β-sheet content of the protein and a propensity to aggregate into larger macromolecular structures. PrP^C is post-translationally modified with a glycosyl-phosphatidylinositol (GPI) anchor attached to the C-terminus. The GPI anchor facilitates the association of PrP^C with cholesterol- and sphingolipid-rich membrane microdomains, termed lipid rafts (reviewed in [2]). Lipid rafts are characterised biochemically by their resistance to solubilisation with detergents, such as Triton X-100, at low temperature, with the resulting detergent-resistant membranes (DRMs) enriched in raft resident proteins and lipids [3]. PrP^C also associates with lipid rafts by virtue of raft targeting determinants within its N-terminal domain [4],[5]. However, the identity of the raft interacting partner(s) for the N-terminal domain of PrP^C remains unknown.

A number of studies suggest that the formation of PrP^Sc takes place in lipid rafts. For example, PrP^Sc, like PrP^C, is present in DRMs isolated from cultured cells [6],[7]. Furthermore, when the GPI anchor of PrP^C is replaced by a transmembrane anchor, the protein redistributes to non-raft regions of the plasma membrane and no conversion occurs [7],[8]. In addition, depletion of cellular cholesterol levels to disrupt lipid rafts leads to a reduction in the PrP^Sc-load in infected cell culture models [8]–[10].

The presence of prion conversion cofactors in a given subcellular location may rationalise why a particular site is favoured for conversion [11]. Of these potential cofactors, there is evidence linking proteoglycans and their glycosaminoglycan (GAG) side chains to PrP^C metabolism [12]. Sulfated GAGs, including heparan sulfate, were identified as constituents of PrP^Sc plaques in the brains of CJD, Gerstmann-Straussler-Scheinker disease and kuru cases, as well as in hamster brains infected with scrapie [13],[14]. The N-terminal half of PrP^C has been shown to bind heparan sulfate [15],[16], with the basic residues at the extreme N-terminus of mature PrP^C constituting a particularly strong site of interaction [17]–[19]. GAGs stimulated the endocytosis of chicken PrP^C [15] and heparin reduced the level of cell surface human PrP^C by an unknown mechanism [17]. Furthermore, the incorporation of PrP^Sc into Chinese Hamster Ovary cells required endogenous GAG expression [20] and heparan sulfate acted as a cellular receptor for prion rods in neuroblastoma cells [21]. However, despite all these observations, the identity of the cellular heparan sulfate involved in the interaction with PrP^C and/or PrP^Sc, whether it is lipid raft-associated and involved in the conformational conversion of PrP^C to PrP^Sc remain to be determined.

In the current study we show that heparin promotes the endocytosis of mammalian PrP^C and displaces it from lipid rafts, suggesting that heparin competes with an endogenous raft-resident heparan sulfate proteoglycan (HSPG) for binding to PrP^C. We then utilised a transmembrane-anchored construct of PrP, PrP-TM [22], which associates with DRMs solely through the raft targeting determinant in its N-terminal domain [5], to identify the GPI-anchored HSPG glypican-1 as a molecule that targets PrP^C to detergent-resistant lipid rafts. In addition, we show that glypican-1 directly interacts with both PrP^C and PrP^Sc and that depletion of glypican-1 in scrapie-infected murine neuroblastoma N2a (ScN2a) cells inhibits the formation of PrP^Sc.

Results Top

Heparin stimulates the endocytosis of PrP^C and displaces it from lipid rafts

To investigate whether heparin promotes the endocytosis of mammalian PrP^C, and to determine the mechanism involved, we utilised human neuroblastoma SH-SY5Y cells stably expressing murine PrP^C (tagged with the 3F4 antibody epitope) and an established endocytosis assay [23],[24]. It should be noted that the SH-SY5Y cells do not express detectable levels of endogenous PrP^C (see Fig. 1E) as reported previously [19],[25],[26]. In the endocytosis assay, plasma membrane proteins are selectively labelled using a cell-impermeable biotin reagent, thus allowing their distinction from proteins either in the secretory pathway or already endocytosed. Cells were then incubated with heparin for 1 h at 37°C and subsequently treated with trypsin to remove residual cell surface PrP^C prior to lysis. Any PrP^C endocytosed during the course of the experiment is protected from trypsin digestion. In the absence of heparin, negligible (4%) cell surface PrP^C was endocytosed, whereas heparin stimulated the endocytosis of PrP^C in a dose-dependent manner, with 77% internalised by 100 µM heparin in 1 h (Fig. 1A and B).

Figure 1. Heparin stimulates the endocytosis of PrP^C in a dose-dependent manner and displaces it from detergent-resistant lipid rafts.

(A) SH-SY5Y cells expressing PrP^C were surface biotinylated and then incubated for 1 h at 37°C in the absence or presence of various concentrations of heparin diluted in OptiMEM. Prior to lysis cells were, where indicated, incubated with trypsin to digest cell surface PrP^C. Cells were then lysed and PrP^C immunoprecipitated from the sample using antibody 3F4. Samples were subjected to SDS PAGE and western blot analysis and the biotin-labelled PrP^C detected with peroxidase-conjugated streptavidin. (B) Densitometric analysis of multiple blots from four separate experiments as described in (A) is shown. (C) SH-SY5Y cells expressing PrP^C were surface biotinylated and then incubated in the absence or presence of 50 µM heparin prepared in OptiMEM for 1 h at 37°C. Cells were homogenised in the presence of 1% (v/v) Triton X-100 and subjected to buoyant sucrose density gradient centrifugation. PrP^C was immunoprecipitated from equal volumes of each gradient fraction using 3F4 and subjected to SDS-PAGE and western blotting. The gradient fractions from both the untreated and heparin treated cells were analysed on the same SDS gel and immunoblotted under identical conditions. The biotin-labelled PrP^C was detected with peroxidase-conjugated streptavidin. Flotillin-1 and transferrin receptor (TfR) were detected by immunoblotting as markers for DRM and detergent-soluble fractions, respectively. (D) Densitometric analysis of the proportion of total PrP^C in the detergent soluble fractions of the plasma membrane. (E) Untransfected SH-SY5Y cells and SH-SY5Y cells expressing either PrP^C or PrP-TM were grown to confluence and then incubated for 1 h in the presence or absence of 50 µM heparin prepared in OptiMEM. Media samples were collected and concentrated and cells harvested and lysed. Cell lysate samples were immunoblotted for PrP^C using antibody 3F4, with β-actin used as a loading control. (F) Quantification of PrP^C and PrP-TM levels after treatment of cells with heparin as in (E). Experiments were performed in triplicate and repeated on three occasions. * P<0.05.

doi:10.1371/journal.ppat.1000666.g001

Cu²⁺ ions stimulate the clathrin-dependent endocytosis of PrP^C by a mechanism which is initiated by the protein dissociating from lipid rafts [24]. Therefore, we determined whether heparin could likewise displace PrP^C from lipid rafts. Cells were first surface biotinylated and then incubated with heparin before homogenisation in the presence of Triton X-100 followed by buoyant sucrose density gradient centrifugation. Consistent with previous reports [5],[24], PrP^C in cells incubated in the absence of heparin resided almost exclusively at the 5%/30% sucrose interface in the DRM fractions containing the raft-associated protein, flotillin-1 (Fig. 1C). However, in cells incubated with heparin, a significant amount (37%) of PrP^C redistributed to detergent-soluble fractions of the plasma membrane isolated at the bottom of the sucrose gradient containing the transferrin receptor (TfR, Fig. 1C and D). To determine whether the heparin treatment had any effect on other aspects of PrP^C metabolism, we investigated the effect of the 1 h heparin treatment on the expression and shedding of PrP^C in untransfected and PrP^C-transfected SH-SY5Y cells (Fig. 1E). Endogenous PrP^C expression in untransfected SH-SY5Y cells remained undetectable following heparin treatment (Fig. 1E). In the SH-SY5Y cells transfected with PrP^C, 1 h heparin treatment led to 19.5% reduction in the total cellular amount of PrP^C (Fig. 1E and F) which may be attributable to its degradation following endocytosis. No PrP^C was detected in concentrated conditioned media in the presence or absence of heparin treatment (data not shown). Together, these data indicate that the addition of heparin to cells for 1 h does not increase the expression or shedding of PrP^C but stimulates the endocytosis of mammalian PrP^C by its lateral displacement from detergent-resistant lipid rafts into detergent-soluble regions of the plasma membrane. From this, we hypothesised that the lateral movement of PrP^C from rafts may occur as exogenous heparin competes with an endogenous raft-resident HSPG for binding to PrP^C.

PrP-TM interacts with DRMs through association with the GPI-anchored HSPG glypican-1

To test this hypothesis we employed a transmembrane-anchored construct of PrP (PrP-TM) in which the GPI anchor signal sequence of murine PrP is replaced with the transmembrane and cytosolic domains of the non-raft protein angiotensin converting enzyme [22]. PrP-TM also contains the 3F4 epitope. While wild-type PrP^C is targeted to DRMs by virtue of both its GPI anchor and lipid raft targeting determinants in its N-terminal domain, PrP-TM localises to DRMs solely by means of its N-terminal domain interacting with raft resident molecule(s) [5]. Thus, PrP-TM offers a unique system to identify cellular components that interact with the N-terminal domain of PrP^C and target it to lipid rafts. SH-SY5Y cells expressing PrP-TM were surface biotinylated and then incubated with heparin prior to DRM isolation. As reported previously [5],[24], PrP-TM resided exclusively in the DRM fractions of the plasma membrane (Fig. 2A). However, in those cells that had been incubated with heparin, 44% of PrP-TM was now localised to the detergent-soluble fractions at the base of the sucrose gradient (Fig. 2A and C). As with wild type PrP^C, the heparin treatment did not increase the expression or stability of PrP-TM (Fig. 1E and F). In parallel, SH-SY5Y cells expressing PrP-TM were incubated with bacterial phosphatidylinositol-specific phospholipase C (PI-PLC) to determine whether the lipid raft targeting of PrP-TM was due to raft-resident GPI-anchored proteins (Fig. 2B). Treatment of cells with PI-PLC after cell surface biotinylation almost completely abrogated PrP-TM association with the DRMs but had no effect on the DRM distribution of flotillin-1 (Fig. 2B and C), indicating that gross disruption of the rafts had not occurred. One interpretation of the ability of both exogenous heparin and PI-PLC treatment to inhibit the association of PrP-TM with DRMs is that a GPI-anchored HSPG is responsible for the lipid raft targeting of PrP-TM.

Figure 2. The association of PrP-TM with DRMs is disrupted by treatment of cells with either heparin or bacterial PI-PLC.

SH-SY5Y cells expressing PrP-TM were surface biotinylated and then (A) incubated in the absence or presence of 50 µM heparin prepared in OptiMEM for 1 h at 37°C or (B) incubated in the absence or presence of 1 U/ml bacterial PI-PLC for 1 h at 4°C. Cells were homogenised in the presence of 1% (v/v) Triton X-100 and subjected to buoyant sucrose density gradient centrifugation. PrP-TM was immunoprecipitated from equal volumes of each gradient fraction using 3F4 and subjected to western blotting. The biotin-labelled PrP-TM fraction was detected with peroxidase-conjugated streptavidin. Flotillin-1 and transferrin receptor (TfR) were detected by immunoblotting as markers for DRM and detergent-soluble fractions respectively. (C) Densitometric analysis of the proportion of total PrP-TM present in the detergent soluble fractions of the plasma membrane after heparin and PI-PLC treatment. Experiments were performed in triplicate and repeated on three occasions. * P<0.05.

doi:10.1371/journal.ppat.1000666.g002

The glypicans are a family of GPI-anchored HSPGs [27], of which glypican-1 is particularly abundant in neurons [28]. To assess whether glypican-1 is involved in the interaction of PrP-TM with DRMs, SH-SY5Y cells expressing PrP-TM were transfected with either a control siRNA or a siRNA pool directed against glypican-1. The specific siRNAs reduced the amount of glypican-1 in the cells by 84% (Fig. 3A). The siRNA treated cells were surface biotinylated prior to DRM isolation. In glypican-1 siRNA treated cells a large proportion (64%) of PrP-TM was displaced from the DRMs, instead localising with the detergent-soluble fractions (Fig. 3B and C), indicating that glypican-1 plays a role in the lipid raft targeting of PrP-TM.

Figure 3. Depletion of glypican-1 inhibits the association of PrP-TM with DRMs.

SH-SY5Y cells expressing PrP-TM were treated with either control siRNA or siRNA targeted to glypican-1 and then allowed to reach confluence for 48 h. Cells were subsequently surface biotinylated and incubated in OptiMEM for 1 h at 37°C in the presence of Tyrphostin A23 to block endocytosis. The media was removed and the cells washed in phosphate-buffered saline prior to homogenisation in the presence of 1% (v/v) Triton X-100 and subjected to buoyant sucrose density gradient centrifugation. (A) Quantification of glypican-1 and PrP-TM expression in cell lysates. To detect glypican-1, cell lysate samples were treated with heparinase I and heparinase III prior to electrophoresis as described in the materials and methods section. (B) PrP-TM was immunoprecipitated from equal volumes of each gradient fraction using 3F4 and then subjected to western blotting with peroxidase-conjugated streptavidin. Flotillin-1 and transferrin receptor (TfR) were detected by immunoblotting as markers for DRM and detergent-soluble fractions, respectively. (C) Densitometric analysis of the proportion of total PrP-TM present in the detergent soluble fractions of the plasma membrane after siRNA treatment from multiple blots from three independent experiments. * P<0.05.

doi:10.1371/journal.ppat.1000666.g003

Glypican-1 retains PrP^C in lipid rafts at the cell surface

Next, we sought to determine if glypican-1 was involved in the lipid raft targeting of wild type PrP^C. We reasoned that if glypican-1 is involved in the raft targeting of PrP^C, then its depletion should increase the endocytosis of PrP^C. SH-SY5Y cells expressing PrP^C were treated with either control siRNA or siRNA against glypican-1. Cells were then surface biotinylated, incubated for 1 h at 37°C and the amount of PrP^C endocytosed determined. In those cells treated with the control siRNA, incubation in serum-free medium resulted in little detectable endocytosis of PrP^C (Fig. 4A and B). However, in the cells treated with the glypican-1 siRNA, a modest but significant amount (19%) of PrP^C was endocytosed (Fig. 4A and B). In the cells treated with the glypican-1 siRNA, a significant knockdown (88%) of glypican-1 was clearly apparent, with no detectable effect on the amount of PrP^C (Fig. 4C).

Figure 4. Depletion of glypican-1 stimulates the endocytosis of PrP^C.

SH-SY5Y cells expressing wild type PrP^C were treated with either control or glypican-1 siRNA and then incubated for 60 h. Cells were surface biotinylated and incubated in OptiMEM for 1 h at 37°C. Where indicated, cells were treated with trypsin to remove remaining cell surface PrP^C. Cells were then lysed and total PrP^C immunoprecipitated from the sample using antibody 3F4. (A) Samples were subjected to western blot analysis and the biotin-labelled PrP^C fraction was detected with peroxidase-conjugated streptavidin. (B) Densitometric analysis (mean ± s.e.m.) of multiple blots from three separate experiments in (A) is shown. (C) Expression of glypican-1 (in lysate samples treated with heparinase I and heparinase III) and PrP^C in the cell lysates from (A). β-actin was used as a loading control. (D) SH-SY5Y cells expressing PrP^C were treated with either control siRNA or glypican-1 siRNA and then allowed to reach confluence for 48 h. Cells were subsequently surface biotinylated and incubated in OptiMEM for 1 h at 37°C. Cells were homogenised in the presence of 1% (v/v) Triton X-100 and subjected to buoyant sucrose density gradient centrifugation. (E) Densitometric analysis of the proportion of total PrP^C present in the detergent soluble fractions of the plasma membrane after siRNA treatment from three independent experiments. (F) SH-SY5Y cells expressing PrP^C were seeded onto glass coverslips and grown to 50% confluency. Cells were fixed, and then incubated with anti-PrP antibody 3F4 and a glypican-1 polyclonal antibody. Finally, cells were incubated with Alexa488-conjugated rabbit anti-mouse and Alexa594-conjugated goat anti-rabbit antibodies and viewed using a DeltaVision Optical Restoration Microscopy System. Images are representative of three individual experiments. Scale bars equal 10 µm. * P<0.05.

doi:10.1371/journal.ppat.1000666.g004

We also assessed the raft distribution of PrP^C in cells treated with control or glypican-1 siRNA. Cells were then surface biotinylated and incubated for 1 h at 37°C in the presence of tyrphostin A23, to block clathrin-mediated endocytosis [24]. In control siRNA treated cells, cell surface PrP^C resided exclusively in the DRMs (Fig. 4D). In contrast, in cells treated with glypican-1 siRNA a significant amount (20.6%) of cell surface PrP^C localised to the detergent-soluble fractions of the sucrose gradient (Fig. 4D and E). Immunofluorescence microscopy revealed that PrP^C and glypican-1 extensively colocalised on the surface of the cells (Fig. 4F). Collectively these data indicate that PrP^C and glypican-1 co-localise on the cell surface and that knockdown of glypican-1 allows PrP^C to translocate out of the detergent-insoluble rafts into detergent-soluble regions of the plasma membrane prior to its endocytosis.

PrP^C and PrP^Sc interact with glypican-1

We next sought to determine if glypican-1 is able to interact directly with PrP^C and PrP^Sc by co-immunoprecipitation. When cell lysates prepared using Triton X-100 from SH-SY5Y cells stably expressing PrP^C (Fig. 5A) or N2a cells endogenously expressing PrP^C (Fig. 5B) were incubated with a polyclonal antibody against glypican-1, PrP^C was co-immunoprecipitated in both cases (Fig. 5A). The interaction between glypican-1 and PrP^C was disrupted by prior treatment of the cell lysates with heparinase I and heparinase III to digest the heparan sulfate chains on glypican-1 [29],[30] or exogenous heparin (Fig. 5A and B), indicating that the interaction between PrP^C and glypican-1 involves the heparan sulfate sidechains on the latter. In order to provide further evidence that glypican-1 and PrP^C are directly associated, co-immunoprecipitation experiments were performed using cells lysed with sarkosyl or octylglucoside, detergents known to completely solubilise lipid rafts [31]–[33]. PrP^C co-immunoprecipitated with glypican-1 using lysates prepared with either detergent (Fig. 5A and B).

Figure 5. PrP^C and PrP^Sc immunoprecipitate with glypican-1.

(A) SH-SY5Y cells expressing PrP^C, (B) N2a cells or (C and D) ScN2a cells were lysed in the indicated detergents and then immunoprecipitated with a polyclonal glypican-1 antibody and where indicated, co-incubated with 50 µM heparin. Those samples pretreated with heparinase I and heparinase III were lysed with Triton X-100 followed by immunoprecipitation with glypican-1 antibody. In (D), immunopreciptiates from ScN2a cells were digested with PK. All immunoprecipitates were subsequently blotted for PrP. TX, Triton X-100; OG, octylglucoside; SK, sarkosyl.

doi:10.1371/journal.ppat.1000666.g005

To address whether glypican-1 was also able to interact with PrP^Sc we performed co-immunoprecipitation experiments using cell extracts prepared from mouse neuroblastoma cells persistently infected with PrP^Sc (ScN2a cells). Immunoprecipitation of glypican-1 led to the co-immunoprecipitation of PrP from cells lysed with either Triton X-100 or octylglucoside (Fig. 5C). Sarkosyl is unsuitable for use in co-immunoprecipitation experiments in ScN2a cells owing to its ability to aggregate PrP^Sc [34]. Co-immunoprecipitation of PrP with glypican-1 in Triton-extracted cells was inhibited by co-incubation with exogenous heparin or removal of the heparan sulfate sidechains on glypican-1 with heparinase I and heparinase III treatment (Fig. 5C). When the protein A-sepharose pellets were treated with proteinase K (PK) prior to subsequent analysis, PrP^Sc was found to have precipitated with glypican-1 in both the Triton X-100 and octylglucoside extracted cells (Fig. 5D). Similarly, the interaction between glypican-1 and PrP^Sc was inhibited by addition of exogenous heparin to the cell lysates or pre-treatment of samples with heparinase I and heparinase III (Fig. 5D). These data indicate that glypican-1 is capable of interacting via its heparan sulfate chains with both normal and protease-resistant forms of PrP.

Depletion of glypican-1 inhibits the formation of PrP^Sc

Next we investigated if the interaction with glypican-1 could modulate the conversion of PrP^C to PrP^Sc. ScN2a cells were treated with the control siRNA or one of four siRNA duplexes targeted to mouse glypican-1. After a total incubation period of 96 h, cells were harvested, lysed, and then the cell lysates digested with PK prior to SDS-PAGE and subsequent immunoblotting (Fig. 6A). Treatment with the control siRNA did not alter the level of PK-resistant PrP^Sc. However, in cells treated with each of the four glypican-1 siRNA duplexes an average 50% reduction in PK-resistant PrP^Sc was consistently observed (Fig. 6A and B). Knockdown (78–87%) of glypican-1 by the siRNA duplexes in the ScN2a cells was confirmed by immunoblotting (Fig. 6C). The reduction in PK-resistant PrP^Sc seen in the glypican-1 depleted cells was not a consequence of a more general reduction in PrP levels, as when non-PK digested samples were immunoblotted the amount of total PrP was unaltered relative to the control siRNA treated cells (Fig. 6C).

Figure 6. Depletion of glypican-1 by siRNA reduces PrP^Sc formation.

ScN2a cells were either untreated or incubated with either control siRNA or one of four siRNAs targeted to glypican-1. After 48 h incubation the treatments were repeated. After a total incubation period of 96 h cells were harvested, lysed and protein concentration determined. (A) For detection of PK-resistant PrP^Sc, samples containing 200 µg protein were digested with 4 µg PK for 30 min at 37°C. Protein was then recovered by methanol precipitation and immunoblotted for PrP using antibody 6D11. (B) Densitometric analysis (mean ± s.e.m.) of PK-resistant PrP^Sc levels for each treatment, relative to those of mock-treated cells, from multiple blots from four independent experiments. (C) To confirm that glypican-1 depletion had been achieved in the ScN2a cells, cell lysate samples were immunoblotted for glypican-1, as well as PrP and β-actin. (D) To confirm the specificity of glypican-1 in modulating PrP^Sc formation, ScN2a cells were treated with control, syndecan-1 or glypican-1 siRNA reagents. Samples were processed as described in (A). (E) Densitometric analysis (mean ± s.e.m.) of PK-resistant PrP^Sc levels for each treatment in (D), from multiple blots from three independent experiments. (F) Confirmation of syndecan-1 and glypican-1 knockdown by immunblotting.

doi:10.1371/journal.ppat.1000666.g006

Next we sought to confirm the specificity of glypican-1 in modulating the conversion of PrP^C to PrP^Sc by testing whether another HSPG, syndecan-1, could influence PrP^Sc accumulation in ScN2a cells. When cells were treated with syndecan-1 siRNA over an incubation period of 96 h, the level of PK-resistant PrP^Sc remained comparable to control siRNA treated cells (Fig. 6D and E). In contrast, treatment of ScN2a cells with glypican-1 siRNA once more led to 50% reduction in PK-resistant PrP^Sc (Fig. 6D and E). Knockdown of glypican-1 (73.3±5.3%) and syndecan-1 (60.7±6.3%) was confirmed by immunoblotting (Fig. 6F).

Cell division can modulate the accumulation of prions in ScN2a cells [35] and glypican-1 depletion has been reported to inhibit cell proliferation in endothelial cells by arresting cell cycle progression [36]. Therefore, we investigated whether the effect of glypican-1 knockdown was altering the amount of PrP^Sc in the ScN2a cells due to modulation of cell division. However, glypican-1 depletion did not significantly affect cell proliferation compared to mock-treated and control siRNA-treated cells (Fig. 7A). As the cell surface is the probable site for the initial interaction between the two isoforms of PrP [11], we assessed whether the reduction in PrP^Sc associated with glypican-1 depletion was due to an alteration in the steady state level of PrP at the cell surface. ScN2a cells were surface biotinylated following treatment with either control siRNA or glypican-1 siRNA. Although there was a reduction in the amount of PrP^Sc in the glypican-1 siRNA treated cells, there was no difference in the amount of biotinylated cell surface PrP (Fig. 7B and C).

Figure 7. Depletion of glypican-1 does not affect cell division or surface levels of PrP^C.

(A) ScN2a cells were seeded into 96 well plates and treated with transfection reagent only or incubated with either control siRNA or one of the four siRNAs targeted to glypican-1. Those experiments exceeding 48 h were dosed with a second treatment of the indicated siRNAs. Cells were then rinsed with PBS and fixed with 70% (v/v) ethanol. Plates were allowed to dry, stained with Hoescht 33342 and the fluorescence measured. (B) ScN2a cells were treated with control or glypican-1 siRNA. After 96 h, cell monolayers were labelled with a membrane impermeable biotin reagent. Biotin-labelled cell surface PrP was detected by immunoprecipitation using 6D11 and subsequent immunoblotting using HRP-conjugated streptavidin. Total PrP and PK-resistant PrP (PrP^Sc) were detected by immunoblotting using antibody 6D11. (C) Densitometric analysis of the proportion of the relative amount of biotinylated cell surface PrP in the absence or presence of glypican-1 siRNA from three independent experiments.

doi:10.1371/journal.ppat.1000666.g007

Glypican-1 is not involved in the inhibitory effect of PrP^C on the BACE1 cleavage of APP

We have recently reported that PrP^C inhibits the cleavage of the amyloid precursor protein (APP) by the β-secretase BACE1 [19]. This effect was dependent on the lipid raft localisation of PrP^C and was mediated by the N-terminal polybasic region of PrP^C through interaction with GAGs [19]. Therefore, we hypothesised that glypican-1 may be involved in the mechanism by which PrP^C regulates the BACE1 cleavage of APP. To address this, control siRNA or siRNA directed against glypican-1 were transfected into SH-SY5Y cells expressing PrP^C. Parallel experiments were performed in SH-SY5Y cells transfected with an empty pIRESneo vector. In the media from control siRNA-treated cells expressing PrP^C the level of sAPPβ, the product of BACE1 cleavage of APP, was dramatically reduced relative to cells lacking PrP^C (Fig. 8A and B), as seen previously [19]. In media from the empty vector cells treated with glypican-1 siRNA there was no alteration in the level of sAPPβ relative to control siRNA-treated cells (Fig. 8A and B). Significantly, levels of sAPPβ in the cells expressing PrP that had been treated with the glypican-1 siRNA were not altered relative to the control siRNA treated cells (Fig. 8A and B), implying that the GAG sidechains of glypican-1 are not critical in mediating the inhibition of BACE1 by PrP^C.

Figure 8. Depletion of glypican-1 does not affect the inhibition of BACE1 by PrP^C.

SH-SY5Y cells co-expressing APP695 and PrP or cells expressing APP695 and an empty pIRESneo vector were treated with control siRNA or siRNA targeted to glypican-1. After 30 h the medium was replaced and cells incubated with reduced-serum medium for a further 24 h. Conditioned medium was harvested and cell lysates prepared. (A) Expression of glypican-1, full-length APP and PrP^C in cell lysates, with β-actin as a loading control. Immunodetection of sAPPβ in conditioned medium. (B) Densitometric analysis of sAPPβ levels in glypican-1 depleted cells relative to control siRNA cells, calculated from multiple blots from three independent experiments.

doi:10.1371/journal.ppat.1000666.g008

Discussion Top

By utilising a transmembrane anchored form of PrP, PrP-TM, which associates with DRMs solely by means of its N-terminal domain and not also via the GPI anchor as in wild type PrP^C, we have identified the GPI-anchored HSPG, glypican-1, as involved in targeting PrP^C to detergent-resistant lipid rafts in the plasma membrane of neuronal cells. Furthermore, we have shown that depletion of glypican-1 by siRNA knock down significantly inhibits the formation of PrP^Sc in a scrapie-infected cell line. Thus, we have identified glypican-1 as a novel cofactor involved in the cellular conversion of PrP^C to PrP^Sc. Glypican-1 appeared to be interacting via its heparan sulfate chains with both PrP^C and PrP^Sc, as the interaction was inhibited by digestion of its GAG chains with heparinase I and heparinase III. This would be consistent with a previous study, where treatment of ScN2a cells with heparinase III caused a profound reduction in the level of PrP^Sc [37].

We propose a model where glypican-1 acts as a scaffold for prion propagation by binding to both PrP^C and PrP^Sc, thereby bringing them into close enough proximity within lipid rafts to facilitate prion conversion (Fig. 9A). This is supported by the observation that both PrP^C and PrP^Sc co-immunoprecipitated with glypican-1 and that both isoforms of PrP are localised in lipid rafts [6],[7]. Glypican-1 may be acting as a catalyst, increasing the rate of conversion of PrP^C to PrP^Sc. In this respect, a previous study has demonstrated that pentosan polysulfate destabilised protease-sensitive PrP with a loss of α-helical structure [38]. Thus, a potential mechanism by which glypican-1 may be acting is through its heparan sulfate side chains exerting a similar destabilising effect on the structural stability of PrP^C, thereby facilitating its refolding in the presence of PrP^Sc. Although depletion of glypican-1 did not alter the cell surface level of PrP, we cannot exclude the possibility that glypican-1 may also have an effect on the conversion of PrP^C to PrP^Sc by altering the trafficking or clearance of PrP^Sc.

Figure 9. Proposed model for glypican-1 in prion conversion.

Under normal circumstances, glypican-1 (light grey) with its heparan sulfate side chains (black lines) constitutes a lipid raft targeting determinant for PrP^C. (A) For PrP^C (dark grey) to exit lipid rafts and undergo endocytosis through its interaction with low density lipoprotein receptor-related protein 1 (LRP1, thick black line [59]), its interaction with glypican-1 must be disrupted (arrow 1). In prion disease, we hypothesise that glypican-1 provides a scaffold to facilitate interaction between PrP^C and PrP^Sc (chequered) and allow misfolding of PrP^C to proceed (arrow 2). (B) Prion therapeutic strategies may act, in part, by disrupting the interaction of PrP^C and PrP^Sc that is facilitated through the heparan sulfate sidechains of glypican-1.

doi:10.1371/journal.ppat.1000666.g009

Glypican-1 itself is internalised from the plasma membrane by a mechanism involving caveolin-1 [39]. During its intracellular trafficking the heparin sulfate chains of glypican-1 are removed and degraded, either by heparanase cleavage or a non-enzymatic deaminative cleavage which is nitric oxide-catalysed and Cu²⁺-/Zn²⁺-dependent [40]. The nitric oxide is derived from nitrosylated cysteine residues within the glypican-1 core protein, formed via a Cu²⁺-dependent redox reaction [41]. In both cell-free experiments and cell models, Cu²⁺-loaded PrP^C was shown to support the nitrosylation of glypican-1 [40]. Recently, it was proposed that glypican-1 autoprocessing is involved in the cellular clearance of PrP^Sc [42]. These authors used scrapie-infected hypothalamic (ScGT1) cells and immunofluorescence microscopy to show that PrP^Sc-associated immunofluorescence was increased in cells treated with a glypican-1 specific siRNA, though no direct quantification was provided [42]. In addition, when ScGT1 cells were treated with reagents to inhibit glypican-1 autoprocessing, western blot analysis revealed increased levels of PrP^Sc [42]. From this, these authors argued that glypican-1 autoprocessing may contribute to the cellular clearance of PrP^Sc, thus inhibiting this process would lead to an increase in PrP^Sc accumulation [42]. However, prevention of the autoprocessing of glypican-1 may increase the half-life of the protein, and thus lead to the observed rise in PrP^Sc if glypican-1 is a cofactor in prion conversion as our data indicate.

Interestingly, PrP^Sc still propagated following the substantial knock down of glypican-1 by siRNA; PrP^Sc levels were reduced by at most 55% despite glypican-1 protein levels being reduced by over 80%. One possibility is that the remaining glypican-1 is sufficient to support prion conversion. Alternatively, these data may be interpreted with respect to the protein-only hypothesis in that PrP^Sc is sufficient on its own to convert PrP^C, with glypican-1 acting merely as a catalyst [1]. Another possibility is the existence of other conversion-favouring cellular cofactors, which may include other HSPGs although the lack of effect of syndecan-1 depletion on PrP^Sc formation argues against the involvement of this family of proteoglycans. In this context it is also interesting to note that siRNA knock down of glypican-1 did not completely abolish the association of PrP-TM with DRMs, suggesting the existence of additional lipid raft targeting molecules. Such components may be proteins, such as other members of the glypican family [27] or lipids. In support of the latter, PrP lacking its GPI anchor was shown to interact with cholesterol and sphingomyelin containing raft-like lipid vesicles [4]. Whether the same additional molecules involved in the raft targeting of PrP^C are also involved in facilitating the conversion of PrP^C to PrP^Sc awaits their identification and subsequent determination of their role in these processes.

To our knowledge the only other protein to date shown to be capable of influencing prion accumulation in cultured cells is the 37 kDa/67 kDa laminin receptor [43]. When infected cells were depleted of the 37 kDa/67 kDa laminin receptor, PrP^Sc levels were significantly reduced, although this may be the result of reduced PrP^C expression seen after these treatments rather than the loss of a direct role for the 37 kDa/67 kDa laminin receptor in prion conversion [43]. Interestingly, PrP^C interacted with the 37 kDa/67 kDa laminin receptor via both a direct interaction involving residues 144–179 and an indirect interaction via an unidentified HSPG involving residues 53–93 of PrP^C [44]. Whether glypican-1 is this intermediate HSPG awaits further study.

Like heparan sulfate, many of the anti-prion compounds identified to date are, or have the capacity to stack into, large polyanionic chains [45]. This suggests a common mechanism of inhibition of PrP^Sc accumulation by these compounds, perhaps by binding to the same or overlapping sites on PrP. In support of this, sulfated glycans have been shown to compete with both Congo red and phosphorothioated oligonucleotides for binding to PrP^C [46],[47]. It is possible that some of the anti-prion compounds identified to date exert their effects by antagonising the binding of glypican-1 to PrP^C and PrP^Sc. The ability of exogenous heparin to inhibit the co-immunoprecipitation of glypican-1 with both PrP^C and PrP^Sc supports such a hypothesis. Thus, such anti-prion compounds may directly disrupt the interaction between PrP^C/PrP^Sc and glypican-1 which is required for optimal prion conversion. Alternatively, the observation that heparin promotes the endocytosis of PrP^C, may suggest that some anti-prion compounds could act by a similar mechanism, promoting the endocytosis of PrP^C, thereby directing the protein to late endosomes and/or lysosomes where conversion to PrP^Sc is inefficient [15] and/or removing PrP^C from the cell surface lipid rafts where glypican-1 facilitates conversion. Following from these results, the design of specific small molecule inhibitors to antagonise the binding of glypican-1 and PrP^C/PrP^Sc may represent a viable therapeutic avenue for the treatment of prion diseases. With any potential treatment for prion diseases that targets PrP^C, there is the concern that the normal functions of the protein may be adversely affected. In this respect, the lack of effect of glypican-1 depletion on the ability of PrP^C to inhibit the cleavage of APP by BACE1 [19] suggests that at least one of the proposed physiological functions of PrP^C [48] is independent of its interaction with glypican-1.

Although heparan and other GAGs disrupt prion conversion in both cell and animal models [49]–[54], such compounds stimulate the conversion of PrP^C to PrP^Sc in cell free systems [38],[55]. From our data this apparent paradox can be explained by exogenous GAGs competitively inhibiting the binding of PrP^C/PrP^Sc to glypican-1 in the cell and animal models, thus not allowing the cellular and infectious forms of PrP to come into close enough contact for conversion to occur (Fig. 9B). While in the cell-free systems the GAGs provide a scaffold to promote the interaction of PrP^Sc with PrP^C thereby facilitating conversion.

In conclusion, we have identified the cell surface HSPG, glypican-1, as a novel lipid raft targeting determinant for PrP^C. In addition, we show that depletion of glypican-1 inhibits the formation of PrP^Sc in scrapie-infected cells, implying that glypican-1 is a novel cellular cofactor in prion conversion. Furthermore, we show that glypican-1 is not required for one of the proposed physiological functions of PrP^C, inhibition of the β-secretase cleavage of APP.

Materials and Methods Top

PrP constructs and cell culture

Insertion of the coding sequence of murine PrP containing a 3F4 epitope tag into pIRESneo (Clontech-Takara Bio Europe) and generation of the PrP-TM construct have been reported previously [22]. For stable transfection of the cDNA encoding the PrP constructs, 30 µg DNA was introduced into human SH-SY5Y neuroblastoma cells by electroporation and selection was performed in normal growth medium containing G418 selection antibiotic. Mouse N2a neuroblastoma cells, N2a cells infected with the mouse-adapted 22L scrapie strain (ScN2a) [56] and SH-SY5Y cells were routinely cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% foetal bovine serum, 50 U/ml penicillin and 0.1 mg/ml streptomycin (all from Invitrogen). Cells were maintained in a humidified incubator at 37°C with 5% CO₂. For analysis of cell-associated proteins, cells were washed with phosphate-buffered saline (20 mM Na₂HPO₄, 2 mM NaH₂PO₄, 0.15 M NaCl, pH 7.4) and scraped from the flasks into phosphate-buffered saline. Cells were pelleted by centrifugation at 500 g for 5 min. Unless indicated otherwise, pelleted cells were lysed in ice cold lysis buffer (150 mM NaCl, 0.5% (v/v) Triton X-100, 0.5% (w/v) sodium deoxycholate, 50 mM Tris-HCl, pH 7.5) supplemented with a complete protease inhibitor cocktail (Roche Applied Science, Burgess Hill, U.K).

Endocytosis assay

Cells at confluency were incubated for 1 h at 4°C with 0.5 mg/ml Biotin sulfo-NHS (Sigma-Aldrich, Poole, U.K.). Cells were then incubated for 30 min at 37°C in OptiMEM. Prior to cell lysis, PrP remaining on the cell surface was removed by digestion with trypsin as described previously [23],[24].

DRM isolation

Cells at confluence were surface biotinylated and then treated as described in individual experiments in the presence of Tyrphostin A23 to prevent PrP^C endocytosis [24]. Media was then removed and cells rinsed in phosphate-buffered saline. Cells were subsequently harvested and then resuspended in Mes-buffered saline (25 mM Mes, 150 mM NaCl, pH 6.5) containing 1% (v/v) Triton X-100. Cells were then homogenised by passing through a Luer 21-gauge needle. After centrifugation at 500 g for 5 min to pellet cell debris, the supernatant was harvested and made 40% (v/v) with respect to sucrose by addition of an equal volume of 80% (v/v) sucrose. A 1 ml aliquot of the sample was then placed beneath a discontinuous sucrose gradient comprising 3 ml of 30% sucrose and 1 ml of 5% sucrose, both in Mes-buffered saline. The samples were centrifuged at 140,000 g in an SW-55 rotor (Beckman Coulter Inc., CA, U.S.A.) for 18 h at 4°C. The sucrose gradients were harvested in 0.5 ml fractions from the base of the gradient and the distribution of proteins monitored by western blot analysis of the individual fractions. Where indicated, biotin-labelled PrP was detected by subsequent immunoprecipitation of epitope-tagged PrP from the individual fractions using antibody 3F4 (Eurogentec Ltd., Southampton, U.K.) and subsequent immunoblotting using horseradish peroxidase-conjugated streptavidin (Thermo Fisher Scientific, Cramlington, U.K.).

Protein assay and enzyme treatments

Protein was quantified using bicinchoninic acid [57] in a microtitre plate-based assay using bovine serum albumin as standard. To detect PK-resistant PrP^Sc, cell lysate samples containing 200 µg of protein in a total volume of 200 µl were incubated for 30 min at 37°C with 4 µg PK. The reaction was terminated by the addition of phenylmethanesulfonyl fluoride to a final concentration of 3 mM. Protein in the samples was precipitated by the addition of 800 µl of ice cold methanol and incubated overnight at 4°C. Precipitated protein was pelleted by centrifugation and then resuspended in dissociation buffer (125 mM Tris-HCl pH 8.0, 2% (w/v) sodium dodecyl sulfate, 20% (v/v) glycerol, 100 mM dithiothreitol, bromophenol blue, pH 6.8) prior to SDS-PAGE. To remove heparan sulfate sidechains from HSPGs prior to detection of glypican-1 and syndecan-1 core proteins by immunoblotting, cell lysates (50 µg protein) were incubated for 5 h at 37°C in the presence of 1 m unit heparinase I and 1 m unit heparinase III (both Sigma). To remove cell surface GPI-anchored proteins, cell monolayers were incubated for 1 h at 4°C with 1 U/ml Bacillus thuringiensis phosphatidylinositol-specific phospholipase C (PI-PLC) (Sigma-Aldrich).

SDS-PAGE and western blot analysis

Samples were prepared in dissociation buffer and boiled for 5 min. Proteins were resolved by SDS-PAGE using either 7–17% polyacrylamide gradient gels or 14.5% polyacrylamide gels. For western blot analysis, resolved proteins were transferred to Immobilon P polyvinylidene difluoride membrane (Amersham, Little Chalfont, U.K.). The membrane was blocked by incubation for 1 h with phosphate-buffered saline containing 0.1% (v/v) Tween-20 and 5% (w/v) dried milk powder. Antibody incubations were performed in phosphate-buffered saline-Tween containing 2% (v/v) bovine serum albumin. Antibody 3F4 recognises the 3F4 epitope tag (corresponding to amino acid residues 109–112 of human PrP) at residues 108–111 of the chimeric murine PrP and was used at a dilution of 1:4000; antibody 6D11 (Eurogentec Ltd.) recognises an epitope within amino acids 93–109 of PrP and was used at 1:10000; antibody 22C11 (Millipore (UK) Ltd, Livingston, U.K.) recognises amino acid residues 66–81 at the N-terminus of APP and was used at 1:2500; antibody 1A9 (GlaxoSmithKline, Harlow, U.K.) recognises a neoepitope on the soluble ectodomain fragment of APP derived from β-secretase cleavage (sAPPβ) and was used at 1:2500; antibody S1 against human glypican-1 was a kind gift from Professor G. David (Flanders Institute for Biotechnology, Belgium) and was used at 1:2000; polyclonal anti-glypican-1 and anti-syndecan-1 (both Abcam plc, Cambridge, U.K.) were used at 1:1000; and anti-actin antibody (Sigma) at 1:5000. Horseradish peroxidase-conjugated streptavidin was used at 1:2000. Bound antibody was detected using peroxidase-conjugated secondary antibodies in conjunction with the enhanced chemiluminescence detection method (Amersham Biosciences, Amersham, U.K.).

RNA interference studies

SH-SY5Y cells expressing PrP^C were seeded into T80 flasks at 70% confluency and incubated with 500 pmol of a 2 µM Smartpool siRNA solution against glypican-1 or a control Smartpool reagent (Dharmacon Inc., Chicago, U.S.A.) complexed with DharmaFECT-1 transfection regent (Dharmacon Inc.) in serum-free medium. After 4 h, serum was added to 10% (v/v). Cells were then incubated for a further 44 h prior to experimentation. To analyse the role of glypican-1 in the inhibitory action of PrP^C on the BACE1 cleavage of APP, SH-SY5Y cells expressing APP695 were transfected with the cDNA encoding PrP^C or an empty pIRESneo vector and treated with glypican-1 or control Smartpool reagents. After 30 h the medium was changed to OptiMEM (Invitrogen), and the cells incubated for a further 24 h. Cells were pelleted and the medium harvested and centrifuged at 1000 g for 5 min to remove residual cell debris. Medium was then concentrated 50-fold using Vivaspin centrifugal concentrators (10 000 MW cut off). To assess the role of glypcian-1 on prion conversion ScN2a cells were treated with 500 pmol of the following 2 µM siRNA solutions: Smartpool glypican-1 solution; one of four individual duplexes targeted to glypican-1; Smartpool syndecan-1 solution or control Smartpool siRNA reagent. Cells were treated twice over a 96 h period with siRNA (at 0 and 48 h). After 96 h cells were harvested and lysed.

Co-immunoprecipitation

Co-immunoprecipitation of PrP isoforms and glypican-1 were performed as described previously [58]. Briefly, SH-SY5Y cells expressing PrP^C, N2a cells or ScN2a cells were grown to confluence and then harvested. Cells were lysed using chilled immunoprecipitation lysis buffer (150 mM NaCl, 10 mM EDTA, 10 mM KH₂PO₄, pH 7.5) containing, where indicated 2% (v/v) Triton X-100, 1% n-octyl-β-D-glucopyranoside (octyl glucoside) or 1% N-laurylsarcosine (sarkosyl). Cell debris were removed by centrifugation and then lysates precleared for 30 min with 0.5% (w/v) protein A-Sepharose. The protein A-Sepharose was removed by centrifugation and the supernatant incubated overnight at 4°C with anti-PrP antibody (6H4) or anti-glypican-1 antibody. Protein A-Sepharose was added to 0.5% (w/v) to the samples and incubation proceeded at 37°C for 1 h. Immunocomplexes were pelleted and the pellet washed six times with 150 mM NaCl, 10 mM Tris-HCl, pH 7.4 containing 0.2% (v/v) Tween-20.

Immunofluorescence microscopy

Cells were seeded onto coverslips and grown to 50% confluency. Cells were then fixed with 4% (v/v) paraformaldehyde/0.1% (v/v) glutaraldehyde in PBS for 15 min, and then blocked for 1 h in PBS containing 5% (v/v) fish skin gelatin (Sigma-Aldrich). Coverslips were then incubated with anti-PrP antibody 3F4 and anti-glypican-1 antibody. Finally, coverslips were incubated with the appropriate fluorescent probe-conjugated secondary antibodies (Molecular Probes, Eugene, U.S.A.) for 1 h and mounted on slides using fluoromount G mounting medium (SouthernBiotech, Alabama, U.S.A). Cells were visualised using a DeltaVision Optical Restoration Microscopy System (Applied Precision Inc., Washington, USA). Data were collected from 30–40 0.1 µm thick optical sections, and 3-D datasets were deconvolved using the softWoRx programme (Applied Precision Inc.). The presented images represent individual Z-slices taken from the middle of the cell.

Assessment of cell number by Hoescht 33342 staining

ScN2a cells (1×10⁴ per well) in 96-well tissue culture plates were cultured overnight in complete medium. After 24 h, cells were treated with the indicated siRNA duplexes. Cells were fixed in 70% ethanol at room temperature for 5 min, and the adherent cell monolayers were stained with the DNA-binding fluorochrome Hoescht 33342 (8.8 µM). Once dry, the fluorescence of each well was measured on a Synergy HT (Bio-Tek) fluorescent plate reader (350 nm excitation and 450 nm emission wavelengths) in order to determine the cell number in each well.

Statistical analysis

Data are expressed as means (± s.e.m.). Experiments were performed in triplicate and repeated on at least three occasions. Statistical analysis was performed using student’s t-test. P-values <0.05 were taken as statistically significant.

References Top

1. Prusiner SB (1998) Prions. Proc Natl Acad Sci USA 95: 13363–13383. Find this article online
   
2. Taylor DR, Hooper NM (2006) The prion protein and lipid rafts. Mol Membr Biol 23: 89–99. Find this article online
   
3. Hooper NM (1999) Detergent-insoluble glycosphingolipid/cholesterol-rich membrane domains, lipid rafts and caveolae. Mol Membr Biol 16: 145–156. Find this article online
   
4. Baron GS, Caughey B (2003) Effect of glycosylphosphatidylinositol anchor-dependent and -independent prion protein association with model raft membranes on conversion to the protease-resistant isoform. J Biol Chem 278: 14883–14892. Find this article online
   
5. Walmsley AR, Zeng F, Hooper NM (2003) The N-terminal region of the prion protein ectodomain contains a lipid raft targeting determinant. J Biol Chem 278: 37241–37248. Find this article online
   
6. Vey M, Pilkuhn S, Wille H, Nixon R, DeArmond SJ, et al. (1996) Subcellular colocalization of the cellular and scrapie prion proteins in caveolae-like membranous domains. Proc Natl Acad Sci USA 93: 14945–14949. Find this article online
   
7. Naslavsky N, Stein R, Yanai A, Friedlander G, Taraboulos A (1997) Characterization of detergent-insoluble complexes containing the cellular prion protein and its scrapie isoform. J Biol Chem 272: 6324–6331. Find this article online
   
8. Taraboulos A, Scott M, Semenov A, Avraham D, Laszlo L, et al. (1995) Cholesterol depletion and modification of COOH-terminal targeting sequence of the prion protein inhibit formation of the scrapie isoform. J Cell Biol 129: 121–132. Find this article online
   
9. Marella M, Lehmann S, Grassi J, Chabry J (2002) Filipin prevents pathological prion protein accumulation by reducing endocytosis and inducing cellular PrP release. J Biol Chem 277: 25457–25464. Find this article online
   
10. Bate C, Salmona M, Diomede L, Williams A (2004) Squalestatin cures prion-infected neurons and protects against prion neurotoxicity. J Biol Chem 279: 14983–14990. Find this article online
    
11. Campana V, Sarnataro D, Zurzolo C (2005) The highways and byways of prion protein trafficking. Trends Cell Biol 15: 102–111. Find this article online
    
12. Diaz-Nido J, Wandosell F, Avila J (2002) Glycosaminoglycans and beta-amyloid, prion and tau peptides in neurodegenerative diseases. Peptides 23: 1323–1332. Find this article online
    
13. Snow AD, Kisilevsky R, Willmer J, Prusiner SB, DeArmond SJ (1989) Sulfated glycosaminoglycans in amyloid plaques of prion diseases. Acta Neuropathol (Berl) 77: 337–342. Find this article online
    
14. Snow AD, Wight TN, Nochlin D, Koike Y, Kimata K, et al. (1990) Immunolocalization of heparan sulfate proteoglycans to the prion protein amyloid plaques of Gerstmann-Straussler syndrome, Creutzfeldt-Jakob disease and scrapie. Lab Invest 63: 601–611. Find this article online
    
15. Shyng SL, Lehmann S, Moulder KL, Harris DA (1995) Sulfated glycans stimulate endocytosis of the cellular isoform of the prion protein, PrPC, in cultured cells. J Biol Chem 270: 30221–30229. Find this article online
    
16. Chen SG, Teplow DB, Parchi P, Teller JK, Gambetti P, et al. (1995) Truncated forms of the human prion protein in normal brain and in prion diseases. J Biol Chem 270: 19173–19180. Find this article online
    
17. Pan T, Wong BS, Liu T, Li R, Petersen RB, et al. (2002) Cell-surface prion protein interacts with glycosaminoglycans. Biochem J 368: 81–90. Find this article online
    
18. Warner RG, Hundt C, Weiss S, Turnbull JE (2002) Identification of the heparan sulfate binding sites in the cellular prion protein. J Biol Chem 277: 18421–18430. Find this article online
    
19. Parkin ET, Watt NT, Hussain I, Eckman EA, Eckman CB, et al. (2007) Cellular prion protein regulates beta-secretase cleavage of the Alzheimer’s amyloid precursor protein. Proc Natl Acad Sci USA 104: 11062–11067. Find this article online
    
20. Hijazi N, Kariv-Inbal Z, Gasset M, Gabizon R (2005) PrPSc incorporation to cells requires endogenous glycosaminoglycan expression. J Biol Chem 280: 17057–17061. Find this article online
    
21. Horonchik L, Tzaban S, Ben-Zaken O, Yedidia Y, Rouvinski A, et al. (2005) Heparan sulfate is a cellular receptor for purified infectious prions. J Biol Chem 280: 17062–17067. Find this article online
    
22. Walmsley AR, Zeng F, Hooper NM (2001) Membrane topology influences N-glycosylation of the prion protein. EMBO J 20: 703–712. Find this article online
    
23. Perera WSS, Hooper NM (2001) Ablation of the metal ion-induced endocytosis of the prion protein by disease-associated mutation of the octarepeat region. Curr Biol 11: 519–523. Find this article online
    
24. Taylor DR, Watt NT, Perera WS, Hooper NM (2005) Assigning functions to distinct regions of the N-terminus of the prion protein that are involved in its copper-stimulated, clathrin-dependent endocytosis. J Cell Sci 118: 5141–5153. Find this article online
    
25. Zeng F, Watt NT, Walmsley AR, Hooper NM (2003) Tethering the N-terminus of the prion protein compromises the cellular response to oxidative stress. J Neurochem 84: 480–490. Find this article online
    
26. Parkin ET, Watt NT, Turner AJ, Hooper NM (2004) Dual mechanisms for shedding of the cellular prion protein. J Biol Chem 279: 11170–11178. Find this article online
    
27. Filmus J, Capurro M, Rast J (2008) Glypicans. Genome Biol 9: 224. Find this article online
    
28. Litwack ED, Ivins JK, Kumbasar A, Paine-Saunders S, Stipp CS, et al. (1998) Expression of the heparan sulfate proteoglycan glypican-1 in the developing rodent. Dev Dyn 211: 72–87. Find this article online
    
29. Drzeniek Z, Siebertz B, Stocker G, Just U, Ostertag W, et al. (1997) Proteoglycan synthesis in haematopoietic cells: isolation and characterization of heparan sulphate proteoglycans expressed by the bone-marrow stromal cell line MS-5. Biochem J 327: 473–480. Find this article online
    
30. Edgren G, Havsmark B, Jonsson M, Fransson LA (1997) Glypican (heparan sulfate proteoglycan) is palmitoylated, deglycanated and reglycanated during recycling in skin fibroblasts. Glycobiology 7: 103–112. Find this article online
    
31. Brown DA, Rose JK (1992) Sorting of GPI-anchored proteins to glycolipid-enriched membrane subdomains during transport to the apical cell surface. Cell 68: 533–544. Find this article online
    
32. Keshet GI, Bar-Peled O, Yaffe D, Nudel U, Gabizon R (2000) The cellular prion protein colocalizes with the dystroglycan complex in the brain. J Neurochem 75: 1889–1897. Find this article online
    
33. Garner AE, Smith DA, Hooper NM (2008) Visualization of detergent solubilization of membranes: implications for the isolation of rafts. Biophys J 94: 1326–1340. Find this article online
    
34. Meyer RK, McKinley MP, Bowman KA, Braunfeld MB, Barry RA, et al. (1986) Separation and properties of cellular and scrapie prion proteins. Proc Natl Acad Sci U S A 83: 2310–2314. Find this article online
    
35. Ghaemmaghami S, Phuan P-W, Perkins B, Ullman J, May BCH, et al. (2007) Cell division modulates prion accumulation in cultured cells. Proc Natl Acad Sci U S A 104: 17971–17976. Find this article online
    
36. Qiao D, Yang X, Meyer K, Friedl A (2008) Glypican-1 regulates anaphase promoting complex/cyclosome substrates and cell cycle progression in endothelial cells. Mol Biol Cell 19: 2789–2801. Find this article online
    
37. Ben-Zaken O, Tzaban S, Tal Y, Horonchik L, Esko JD, et al. (2003) Cellular heparan sulfate participates in the metabolism of prions. J Biol Chem 278: 40041–40049. Find this article online
    
38. Wong C, Xiong LW, Horiuchi M, Raymond L, Wehrly K, et al. (2001) Sulfated glycans and elevated temperature stimulate PrP(Sc)-dependent cell-free formation of protease-resistant prion protein. EMBO J 20: 377–386. Find this article online
    
39. Cheng F, Mani K, van den Born J, Ding K, Belting M, et al. (2002) Nitric oxide-dependent processing of heparan sulfate in recycling S-nitrosylated glypican-1 takes place in caveolin-1-containing endosomes. J Biol Chem 277: 44431–44439. Find this article online
    
40. Mani K, Cheng F, Havsmark B, Jonsson M, Belting M, et al. (2003) Prion, amyloid beta-derived Cu(II) ions, or free Zn(II) ions support S-nitroso-dependent autocleavage of glypican-1 heparan sulfate. J Biol Chem 278: 38956–38965. Find this article online
    
41. Ding K, Mani K, Cheng F, Belting M, Fransson LA (2002) Copper-dependent autocleavage of glypican-1 heparan sulfate by nitric oxide derived from intrinsic nitrosothiols. J Biol Chem 277: 33353–33360. Find this article online
    
42. Lofgren K, Cheng F, Fransson LA, Bedecs K, Mani K (2008) Involvement of glypican-1 autoprocessing in scrapie infection. Eur J Neurosci 28: 964–972. Find this article online
    
43. Leucht C, Simoneau S, Rey C, Vana K, Rieger R, et al. (2003) The 37 kDa/67 kDa laminin receptor is required for PrP(Sc) propagation in scrapie-infected neuronal cells. EMBO Rep 4: 290–295. Find this article online
    
44. Hundt C, Peyrin JM, Haik S, Gauczynski S, Leucht C, et al. (2001) Identification of interaction domains of the prion protein with its 37- kDa/67-kDa laminin receptor. EMBO J 20: 5876–5886. Find this article online
    
45. Caughey B, Baron GS (2006) Prions and their partners in crime. Nature 443: 803–810. Find this article online
    
46. Caughey B, Brown K, Raymond GJ, Katzenstein GE, Thresher W (1994) Binding of the protease-sensitive form of PrP (prion protein) to sulfated glycosaminoglycan and congo red. J Virol 68: 2135–2141. Find this article online
    
47. Kocisko DA, Vaillant A, Lee KS, Arnold KM, Bertholet N, et al. (2006) Potent antiscrapie activities of degenerate phosphorothioate oligonucleotides. Antimicrob Agents Chemother 50: 1034–1044. Find this article online
    
48. Hooper NM, Turner AJ (2008) A new take on prions: preventing Alzheimer’s disease. Trends Biochem Sci 33: 151–155. Find this article online
    
49. Adjou KT, Simoneau S, Sales N, Lamoury F, Dormont D, et al. (2003) A novel generation of heparan sulfate mimetics for the treatment of prion diseases. J Gen Virol 84: 2595–2603. Find this article online
    
50. Beringue V, Adjou KT, Lamoury F, Maignien T, Deslys JP, et al. (2000) Opposite effects of dextran sulfate 500, the polyene antibiotic MS-8209, and Congo red on accumulation of the protease-resistant isoform of PrP in the spleens of mice inoculated intraperitoneally with the scrapie agent. J Virol 74: 5432–5440. Find this article online
    
51. Caughey B, Raymond GJ (1993) Sulfated polyanion inhibition of scrapie-associated PrP accumulation in cultured cells. J Virol 67: 643–650. Find this article online
    
52. Diringer H, Ehlers B (1991) Chemoprophylaxis of scrapie in mice. J Gen Virol 72: 457–460. Find this article online
    
53. Kimberlin RH, Walker CA (1986) Suppression of scrapie infection in mice by heteropolyanion 23, dextran sulfate, and some other polyanions. Antimicrob Agents Chemother 30: 409–413. Find this article online
    
54. Schonberger O, Horonchik L, Gabizon R, Papy-Garcia D, Barritault D, et al. (2003) Novel heparan mimetics potently inhibit the scrapie prion protein and its endocytosis. Biochem Biophys Res Commun 312: 473–479. Find this article online
    
55. Deleault NR, Geoghegan JC, Nishina K, Kascsak R, Williamson RA, et al. (2005) Protease-resistant prion protein amplification reconstituted with partially purified substrates and synthetic polyanions. J Biol Chem 280: 26873–26879. Find this article online
    
56. Nishida N, Harris DA, Vilette D, Laude H, Frobert Y, et al. (2000) Successful transmission of three mouse-adapted scrapie strains to murine neuroblastoma cell lines overexpressing wild-type mouse prion protein. J Virol 74: 320–325. Find this article online
    
57. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, et al. (1985) Measurement of protein using bicinchoninic acid. Anal Biochem 150: 76–85. Find this article online
    
58. Cheng F, Lindqvist J, Haigh CL, Brown DR, Mani K (2006) Copper-dependent co-internalization of the prion protein and glypican-1. J Neurochem 98: 1445–1457. Find this article online
    
59. Taylor DR, Hooper NM (2007) The low-density lipoprotein receptor-related protein 1 (LRP1) mediates the endocytosis of the cellular prion protein. Biochem J 402: 17–23. Find this article online