Inoculation of Scrapie with the Self-Assembling RADA-Peptide Disrupts Prion Accumulation and Extends Hamster Survival

Intracerebral inoculation of 263K Scrapie brain homogenate (PrPsc) with a self-assembling RADA-peptide (RADA) significantly delayed disease onset and increased hamster survival. Time of survival was dependent on the dose of RADA and pre-incubation with PrPsc prior to inoculation. RADA treatment resulted in the absence of detectable PrPsc at 40 d followed by an increased rate of PrPsc accumulation at 75 d up to sacrifice. In all PrPsc inoculated animals, clinical symptoms were observed ∼10 d prior to sacrifice and brains showed spongiform degeneration with Congo red positive plaques. A time-dependent increase in reactive gliosis was observed in both groups with more GFAP detected in RADA treated animals at all time points. The PrP protein showed dose-dependent binding to RADA and this binding was competitively inhibited by Congo Red. We conclude that RADA disrupts the efficacy of prion transmission by altering the rate of PrPsc accumulation. This is the first demonstration that a self-assembling biomolecular peptide can interact with PrPsc, disrupt the course of Scrapie disease process, and extend survival.


Introduction
Transmissible spongiform encephalopathies are incurable, fatal neurodegenerative diseases characterized by the accumulation of abnormal prion protein (PrPsc), neuronal cell death and vacuolation of brain [1]. The PrPsc protein is extractable from diseased tissue and is distinguished from endogenous PrPc by partial protease resistance and detergent insolubility [2]. The transmissible agent is the PrPsc protein and it serves as a template for the molecular conversion of endogenous PrPc into the abnormal PrPsc structural isoform [1]. Host expression of PrPc is necessary for disease transmission, as ablation of the PrPc gene prevents disease [3] and over expression of PrPc followed by PrPsc challenge accelerates disease [4]. The molecular events that mediate neuronal PrPc to PrPsc conversion, not simply accumulated PrPsc, appears to be the initiating factor mitigating the neurodegenerative disease process [5]. Yet the mechanism responsible for the conformational conversion of PrPc to an infectious prion remains enigmatic.
The use of synthetic peptides designed with intrinsic functional domains have been engineered to facilitate drug delivery, cell attachment, and tissue regeneration [6][7][8]. These biomolecular scaffolds exploit the properties of natural protein sequences to mediate targeted cellular events [9]. They offer distinct advantages over traditional pharmacotherapies because they are composed of normal biological constituents, devoid of animal contaminants, and are biodegradable [10]. Moreover, some of these substrates can mimic the dynamic structural changes observed in protein mis-folding diseases [11,12]. The use of these peptides to inhibit endogenous protein mis-folding may prove useful in the elucidation of molecular conversion events or therapeutic intervention.
In this report we identify a self-assembling synthetic peptide composed of a 16-mer RADA repeat that significantly extends hamster survival when pre-incubated with 263K Scrapie prior to intracerebral inoculation. RADA combined with Scrapie results in an initial delay in detectable PrPsc followed by increased accumulation at later time points. There is a concomitant delay in observable Scrapie symptoms despite elevated PrPsc levels at 75-days when equivalent control animals required sacrifice. Moreover, RADA treatment shows a similar time-course for the induction of reactive gliosis and GFAP as PrPsc alone, but with increased levels at all time points. Furthermore, we show dosedependent binding of PrP to RADA and demonstrate competitive binding inhibition of PrP to RADA with the amyloid-binding dye Congo red. We postulate that a physiochemical interaction of PrPsc with RADA impedes the efficacy of prion conversion and disease progression by disrupting the rate of PrPsc accumulation through altered clearance and distribution.

Increased survival time in animals inoculated with PrPsc and RADA-peptide
Hamsters inoculated with PrPsc combined with RADA survived significantly longer than those that received an equivalent dose of PrPsc alone. The mean survival time of animals inoculated with 1% PrPsc (10 22 dose) combined with 0.9% RADA was 114 d (n = 24) as compared to 78 d (n = 24) for animals that received equivalent PrPsc alone or 80 d (n = 6) for animals inoculated with PrPsc-agarose plugs ( Fig. 1A; P,0.001). A rapid toxicity was observed with this combined inoculum in ,20% of the hamsters within 24 h (Table 1). This toxicity did not occur in animals that received equivalent doses of PrPsc alone, PrPsc combined with agarose, RADA alone, or normal brain homogenate combined with RADA. With increased dilution of PrPsc+RADA there was reduced morbidity in that 24 h period. No animals died unexpectedly after this initial 24 h period.
As expected, increased dilution of PrPsc resulted in a dosedependent increase in survival time ( Fig. 1B; squares). Importantly, increased dilution of PrPsc with 0.9% RADA resulted in a significant increase in survival time as compared to the equivalent doses of PrPsc alone ( Fig. 1B; circles; Mann-Whitney P,0.01). The combined inoculum showed dose-dependence, but a plateau in total survival time (,125 d) was observed at dilutions of PrPsc below 2610 24 , suggesting that maximal survival promoted by  Fig. 1D shows a dose-dependent increase in hamster survival in days (mean6SEM) with increasing concentration of RADA (open circles) inoculated with 1% Scrapie. A synthetic 16-mer RADApeptide was hydrated combined with 1% Scrapie and inoculated. Survival in days was plotted (grey triangle) and used to estimate RADA concentration. doi:10.1371/journal.pone.0004440.g001 RADA had been achieved. Based on our survival curve for titrated 263K Scrapie ( Fig. 1C; open circles; n = 6 each data point) the initial PrPsc inoculum of 10 22 when combined with RADA was equivalent to a starting inoculum of 10 29 PrPsc alone, a 7-log reduction in prion titer ( Fig. 1C; dashed line). In all groups inoculated with PrPsc, clinical Scrapie symptoms (startle response followed by increased righting reflex and ataxia) began 10-15 d prior to sacrifice (data not shown). No animals injected with RADA alone or combined with normal brain homogenate exhibited Scrapie symptoms and those not sacrificed as agematched controls survived .200 days.

Increased survival is RADA-peptide dose-dependent
Increased survival time of hamsters inoculated with PrPsc was RADA dose-dependent. When RADA concentration was varied 0-1% with a constant 10 22 dose of PrPsc there was a dosedependent increase in survival time with maximal survival achieved at the highest concentration of RADA used ( Fig. 1D; circles). The survival curve was sigmoid suggesting that no additional increase in survival would be achieved with RADA concentrations above 1% when combined with the 10 22 PrPsc dose. We synthesized our own 16-mer RADA-peptide to 95% purity and estimated the concentration of our preparation to be ,0.2% w/v. This material when combined with PrPsc increased survival to ,87 d from 80 d compared to PrPsc alone (Mann-Whitney; P = 0.02) and RADA concentration calculated from our survival curve was ,0.14% ( Fig. 1D; triangle).
Total PrP, Prion and GFAP proteins are elevated in brain of animals inoculated with PrPsc -RADA Western blot showed increased detection of total PrP protein in brain homogenate of animals that received combined PrPsc with RADA ( Fig. 2, top left panel) at the time of sacrifice. Both PrPsc alone and PrPsc combined with RADA had increased PrP protein compared to animals treated with normal brain homogenate (PrPc). This increase in total PrP is a measure of both PrPc and accumulated PrPsc. The highest level of PrP was detected in brain homogenates from animals treated with PrPsc combined with RADA. Protein load was normalized by BCA and confirmed by detection of flotillin-1 which remained unchanged ( Fig. 2; bottom left panel). As expected, PrP from hamsters inoculated with control  brain homogenate, either alone (PrPc) or in combination with RADA (PrPc+RADA), were proteinase-K (PK) sensitive ( Fig. 2, top right panel). Brain homogenates from hamsters inoculated with PrPsc alone (PrPsc) or combined with RADA (PrPsc+RADA) had detectable PK-resistant prion, with the greatest amount detected in brain homogenate from animals that received the combined inoculant (Fig. 2, top right panel). Glial fibrillary acidic protein (GFAP) was used to validate the fidelity of the PK reaction and is completely digested by PK in all groups (Fig. 2, bottom right panel). Interestingly, more detectable GFAP protein was present in brain homogenate from the animals that received PrPsc+RADA (Fig. 2, bottom right panel).

Delayed PrPsc accumulation with RADA-peptide
In contrast to animals that received inoculation of PrPsc alone, no PK-resistant prion was detected by Western blot in brain homogenates at 40 d from animals that received PrPsc+RADA ( Fig. 3A; top panels; left side). At 75 d the animals that received PrPsc alone had late stage clinical Scrapie symptoms whereas those that received PrPsc with RADA had no observable symptoms (data not shown). At the 75 d time point PK-resistant prion was detected in brain of animals that received PrPsc alone or combined with RADA ( Fig. 3A; top panel; right side). Detection of GFAP protein by Western blot increased from day 40 to 75 in animals of both treatment groups ( Fig. 3A; middle panels), but was greater in brains of animals that received PrPsc+RADA at both time points ( Fig. 3C; P,0.001). Protein load was normalized by BCA and confirmed by the equal and unchanged detection of the contactin-1 protein which also served to validate PK digestion ( Fig. 3A; bottom panels). Quantification of Western blots shows at 40 d there is significantly more PK-resistant prion protein in animals inoculated with PrPsc alone ( Fig. 3B; P,0.001), whereas at 75 d there is significantly more prion protein in brain homogenate from animals inoculated with PrPsc+RADA ( Fig. 3B; P,0.001).

RADA induced ventricular enlargement and delayed spongiform degeneration
At the time of sacrifice the brains of hamsters inoculated with PrPsc+RADA were filled with an increased amount of cerebral spinal fluid (CSF) compared to animals inoculated with PrPsc alone, however gross wet weight (,1 g) did not differ significantly between these treatment groups (data not shown). Coronal brain sections showed enlarged lateral ventricles from animals that received PrPsc+RADA at 75 d and 115 d relative to those inoculated with PrPsc alone or PrPc+RADA ( Fig. 4A; top panels). Typical spongiform degeneration was observed in cortical brain sections at 75 d from animals inoculated with Scrapie alone and to a lesser extent those inoculated with RADA ( Fig. 4B; bottom panels). However, at 115 d the brain of PrPsc+RADA inoculated hamsters had abundant spongiform degeneration equivalent to levels observed at 75 d in animals inoculated with Scrapie alone. Importantly, no abnormal ventricular enlargement or spongiform degeneration was observed in brain sections from animals inoculated with PrPc+RADA. Hence, ventricular enlargement is a consequence of the combined action of PrPsc+RADA and not RADA alone. Moreover, our histopathology indicates that RADA delayed the progression of cortical spongiform degeneration induced by PrPsc.
Increased reactive gliosis and PrP protein aggregates in brain following PrPsc inoculation with RADA-peptide Hamster brain sections from animals at the time of sacrifice showed increased immunoreactive GFAP-positive astrocytes following inoculation with PrPsc alone (day 75) or PrPsc+RADA (day 115) as compared to animals inoculated with PrPc+RADA ( Fig. 5A; top panels). Few GFAP immunoreactive astrocytes were detectable at 115 d in control brain (PrPc+RADA) and PrP was undetectable by immunofluorescence. Animals that received PrPsc alone showed increased GFAP positive astrocytes with some detectable PrP protein; whereas those that received PrPsc+RADA showed a massive increase in immunoreactive GFAP and large PrP deposits. These immunoreactive PrP-positive aggregates were PK resistant (data not shown). In addition, PrPsc increased detection of the microglia/macrophage-specific calcium-binding protein IBA1 (ionized calcium binding adaptor molecule 1). Abundant immunoreactive IBA1-positive microglia were detected in 75 d PrPsc and 115 d PrPsc+RADA brain sections as compared to minimal detection in sections from animals inoculated with PrPc+RADA ( Fig. 5B; bottom panels). The absence of IBA-1 and GFAP co-localization demonstrates that the reactive gliosis induced by PrPsc and PrPsc+RADA involves two distinct glial cell populations.

PrP binds RADA-peptide and binding is inhibited by Congo Red
Normal and Scrapie-infected brain homogenates were incubated with RADA bound to 96-well plates. PrP binding was dependent on the concentration of RADA with no significant difference observed between normal or Scrapie brain (Fig. 6A). Pre-incubation with Congo red resulted in a dose-dependent inhibition of binding of PrP to RADA with a significant inhibition of PrP binding observed at 1 mM Congo red ( Fig. 6B; Mann-Whitney; P,0.001). Importantly, the inhibition of PrP binding to RADA by Congo red was dependent on the order of reagent addition; inhibition required Congo red pre-incubation with PrP prior to RADA exposure (Fig. 6C). RADA pre-incubated with Congo red before or after plate binding was ineffective in inhibiting the binding of PrP, suggesting that the Congo red interacts with PrP, not RADA, to inhibit binding.

PrPsc infectivity is not modified by RADA-peptide
We used four independent preparations of RADA combined with PrPsc to show increased survival time in hamsters was not limited to a single sample. All treatment groups showed a significant increase in animal survival compared to those inoculated with equivalent dose of PrPsc alone (Table 2). Moreover, when brain homogenate was prepared from animals inoculated with PrPsc combined with RADA and re-inoculated back into a new group of hamsters (PrPsc/RADA; dose 10 22 ) no increase in survival was observed (Table 3). This demonstrates that the property of Scrapie infectivity was not modified by exposure to the RADA in subsequent passage. The same inoculant when combined with RADA again (PrPsc/RADA+RADA), resulted in an increased survival time.

Discussion
The complex molecular events that underlie most neurodegenerative diseases are poorly understood. In prion diseases the mechanism by which PrPc is converted to PrPsc is central to disease transmission and pathophysiology [1,13]. Indeed, deposition of PrPsc in the absence of endogenous PrPc is insufficient to cause disease. Consequently, the disruption of PrPc to PrPsc conversion and concomitant cellular events provides a useful basis for therapeutic intervention.
The structural change from primarily alpha-helical PrPc to beta-sheet PrPsc is thought to underpin the aggregation status, protease resistance and replication capacity of PrPsc [14,15]. The PrPsc template is necessary to promote conversion of PrPc in transmitted disease, but mutations in the PrP gene can impart intrinsic changes in PrPc primary structure that fosters spontaneous conformational conversion that leads to abnormal protein aggregation and prion diseases [1,16]. The existence of multiple prion strains with differing biochemical and pathophysiological signatures suggests that a range of stabilized structural isoforms of PrPsc can be transmitted to PrPc [17,18]. This functional diversity likely reflects the unique properties of stabilized prion structures and their ability to complex with distinct compliments of endogenous factors. Moreover, the neuropathological variability observed among prion strains may result from a selective targeting of a strain-specific PrPsc complex to unique cellular or extracellular sites within the brain [17,19,20]. Hence, disruption of the prion structure or its interaction with these endogenous factors would impede PrPc conversion, cellular distribution and neuropathology.
The disparity observed in the bioactivity of anti-prion agents such as polyene antibiotics and tricyclic compounds among prion strains may reflect their differing ability to disrupt PrPsc interaction with strain specific proteins [21]. These agents likely interrupt innate PrPsc+ protein interaction thereby modifying PrPsc distribution and associated cellular conversion events. These changes might reduce the rate or site of PrPsc accumulation thereby decreasing efficacy of prion transmission [20,22]. However, impermanent changes to PrPsc structure would inexorably lead to its replication from PrPc and eventual disease. This may explain the ability of anti-prion agents to reduce the rate of PrPsc accumulation and extend animal survival, but not cure disease. Synthetic peptides derived from the primary sequence of PrPc have been shown to have anti-prion activity [23]. Peptides generated against PrPc residues 106-126 or 109-141 prevented PrPsc conversion in cell-free systems [24,25]. Additionally, the use of PrP119-136 peptide was effective in decreasing PrPsc in chronically infected cells [26]. Soto et al generated a peptide corresponding to the amino acid residues of PrPc with the propensity to form beta-sheet structure and inserted incremental proline residues to disrupt the conformational requirements of an ordered beta-sheet. Pre-incubation of this peptide (iPrP13) with mouse-adapted Scrapie prior to intracerebral inoculation increased survival time in mice. Moreover, iPrP13 was able to reduce PrPsc in cultured cells following exposure to Scrapie infected brain homogenate [23]. This data demonstrates that a competitive peptide mimetic can impede PrPc to PrPsc conversion, result in reduced PrPsc accumulation and delay disease progression. Alternatively, transgenic mice expressing mutant PrP(P101L) at low level persist as clinically normal but can be induced to develop neurodegeneration and amyloid deposits by inoculation of a synthetic beta-sheet peptide (PrP89-143) carrying  Fig. 3A shows a Western blot from hamster brain homogenates inoculated with 1% Scrapie alone (Sc) or combined with 0.9% RADA (Sc+RADA) at 40 d and 75 d treated with (+) or without (2) proteinase-K (PK). The top panels show PrPc and PrPsc detection at each time point (20 mg). Middle panels show detection of GFAP protein and bottom panel contactin-1. Fig. 3B shows quantification of total PrP (2PK) and prion (+PK) from Western blots at each time-point. Fig. 3C show quantification of ,55 kDa GFAP band (2PK) from Western blots from each time-point. doi:10.1371/journal.pone.0004440.g003 the P101L substitution [27]. These results suggest that a peptide mimetic can also induce PrPc conversion to PrPsc and promote disease.
Here we report the anti-prion activity of a synthetic RADApeptide repeat that is unrelated to PrPc primary structure and selfassembles into a hydrated beta-sheet nanofiber matrix [11]. The RADA hydrogel provides a suitable substrate for neuron attachment, outgrowth, and regeneration [8,28]. RADA likely mimics the adhesive proteins present in the extracellular matrix (ECM) that contain the RGD tri-peptide motif and modulate cell recognition, adhesion and migration [29]. Indeed, the core RAD motif in the RADA-peptide is similar to the ubiquitous integrin receptor binding site RGD [30,31]. The PrP protein has been shown to interact with the ECM protein laminin, which contains a RGD domain and has self-assembling properties [32,33]. Additionally, PrP interacts with the 37 kDa/67 kDa laminin receptor which plays a role in the propagation of Scrapie in vitro [34]. The increase in survival of rodents treated with sulfated glycans after Scrapie infection is likely mediated by disruption of prion/laminin receptor interaction [35]. Interestingly, an inverted RGD domain is found in the rodent PrP protein at the C-terminus (DGR aa 229-227) that is absent in other species.
Our data shows that pre-incubation of RADA with PrPsc brain homogenate prior to inoculation is necessary for its anti-prion activity. Moreover, we show that both PrPc and PrPsc bind RADA and binding can be competitively inhibited with Congo red. The planar dye Congo red has anti-prion activity and is routinely used as an amyloid stain [36]. We also show that Congo red binding competes with RADA for PrP binding and not by binding RADA directly. This suggests that Congo red and RADA share a common binding motif in PrPsc and implies a similar mechanism by which they exert their anti-prion activity. Although the mechanism by which Congo red binds amyloid and PrPsc has not been elucidated, its anti-prion activity may be mediated through stabilization of the PrPsc conformation that then impedes the process required for PrPc conversion [37,38]. Interestingly, Congo red has a strong self-assembling activity, a property that other symmetric or planar dyes may share, in that they form ribbon-like micellar species that may provide an adhesive structure for binding beta-sheet peptide chains [39]. The symmetrical self-assembled beta-sheet rich RADA-scaffold may provide a complimentary surface for PrPsc binding and consequently uncoupling the process necessary for PrPc to PrPsc conversion.
Alternatively, the interaction of PrPsc with RADA may competitively disrupt PrPsc binding to endogenous ECM proteins. This could result in altered clearance of the PrPsc+RADA complex or a distinct extracellular distribution. A change in clearance of PrPsc mediated by RADA binding could result in a reduced PrPsc titer retained in the brain that would account for the delay in prion accumulation at 40 d and the increased survival. Yet, this does not explain the progressive increase in reactive gliosis from the time of inoculation, the increased rate of PrPsc accumulation from 40-to 75 d with a disconnect between the severity of clinical disease and total PrPsc. However, this could be explained by an altered extracellular Figure 5. Scrapie induced reactive gliosis and PrP aggregation is increased by RADA. Fig. 5A shows dual labeled immunofluorescent GFAP (red) and PrP (green) hamster brain micrographs. Limited staining of GFAP and no detectable PrP was observed in PrPc+RADA treated brain after 115 d (top right panel). PrPsc treatment alone results in an increased detection of GFAP-positive astrocytes (red) and weak PrP protein (green, white arrows) after 75 d (middle panel). Combined PrPsc+RADA treatment (right panel) results in abundant and strong detection of both GFAP-positive astrocytes and large PrP deposits (green, arrows). V = ventricle. Bv = blood vessel. DAPI nuclei (blue). Bars = 50 mm. Fig. 5B shows micrographs of hamster brain sections labeled with GFAP (red) and IBA1 (green). Increased IBA1-positive immunoreactive microglia are detected in brain following inoculation of PrPsc alone after 75 d (left panel; white arrows) and PrPsc+RADA after 115 d (middle panel; white arrows) as compared to PrPc+RADA after 115 d (right panel; white arrows). IBA1 and GFAP immunoreactivity do not co-localize in dual labeled sections from PrPsc and PrPsc+RADA brain. V = ventricle. DAPI nuclei (blue). Bars = 50 mm. doi:10.1371/journal.pone.0004440.g005 distribution of the PrPsc-RADA complex. Localization to secondary sites not normally involved in PrPc to PrPsc conversion could act to sequester the PrPsc from it preferred neuronal site.
Here the PrPsc+RADA complex could be more susceptible to protease degradation within the extracellular matrix or via intracellular pathways. Indeed, the conversion of PrPc to PrPsc can occur at multiple cellular sites [40,41]. If a predominant, neuronal site is required for the neuropathological impact of prion disease then accumulation at secondary sites, such as glia, could alter the neuropathology. The progressive increase in reactive gliosis observed in animals inoculated with PrPsc+RADA suggests an active disease process but may reflect a more robust neuro-protective response.
Future experiments will determine binding of prion from different species and strains to RADA and the capability of RADA to disrupt in vitro PrPc to PrPsc conversion. The participation of an intrinsic rodent RGD-like motif in PrPc+ protein interaction might account for discrepancies in the efficacy of prion conversion between species, strains, and anti-prion compounds observed in assays based on the rodent PrP protein.
A further evaluation of PrP interaction with synthetic peptides corresponding to RGD motifs may be useful in elucidating the molecular mechanism by which certain compounds exert their anti-prion activity and illuminate distinct biochemical properties inherent in the structural variability of prions. We conclude that RADA impedes the propagation of PrPsc from endogenous PrPc Figure 6. PrP binds RADA-peptide and binding can be inhibited with Congo red. Fig. 6A shows chemiluminescent detection of PrP binding to RADA dilutions from Scrapie (PrPsc) and normal (PrPc) brain homogenates. Fig. 6B shows quantification of PrP binding to RADA (Control RADA/PrP) following pre-incubation with dilutions of Congo Red. Fig. 6C compares the binding of PrP to RADA based on the order of reagent addition. Quantification of PrP binding to RADA: RADA followed by PrP (RADA+PrP), RADA followed by 1 mM Congo red (CR) then PrP (RADA+CR+PrP), RADA followed by preincubated PrP with 1 mM CR (RADA+CR/PrP), RADA pre-incubated with 1 mM CR followed by PrP (RADA/CR+PrP). RLU = relative light units. doi:10.1371/journal.pone.0004440.g006  resulting in an altered rate of PrPsc accumulation that delays disease onset and extends survival.

Animals
Female Syrian Golden hamsters (LVG) at 4 weeks of age were housed in pairs on a 12 h light-dark cycle. Animals were provided continual access to food and water. All protocols were approved by the USDA animal care and use committee and experimental procedures conducted in certified BL2 laboratory. PrPsc-infected hamsters were sacrificed when clinical Scrapie symptoms included; increased startle response, ataxia, and .5 s righting reflex.

Reagents
Infectious 263K Scrapie was propagated by serial passage in hamster brains for 72 days following intracerebral inoculation. Infected or normal brains were frozen on dry ice and stored at 280 C until use. A 16-mer RADA peptide hydrogel was purchased from BD Bioscience as Puramatrix (Product #354250; Lot#425714) or synthesized to 95% purity (AnaSpec, CA) for comparison. Congo red (Sigma-Aldrich, MO) was solubalized in water at 1 mM. Ultra pure low melting temperature agarose (AquaPor LM, National Diagnostics, GA) was dissolved in water at 65 C and combined with brain homogenate to form agarose plug controls.

Inoculation
Hamster brain homogenates (10%) were prepared by homogenization in 320 mM sucrose and pre-cleared by centrifugation at 3000 RPM for 10 min at 4 C. Resulting supernatant was processed for inoculation by 1:10 dilution in 320 mM sucrose or PuraMatrix (1% w/v synthetic 16-mer RADA peptide hydrogel) to yield a 1% brain homogenate (dose equivalent 10 22 ) in 0.9% peptide hydrogel. Additionally, brain homogenate or PuraMatrix was serially diluted in 320 mM sucrose and final concentration of brain homogenate was achieved by 1:10 dilution in PuraMatrix. Samples were vortexed and incubated at room temperature for 20 min prior to unilateral intracerebral inoculation (40 ml) using a 27-guage needle attached to a 1 ml syringe.

ELISA assay
RADA-peptide hydrogel was diluted in 0.1 M sodium bicarbonate buffer (pH 8.6) and adsorbed to black 96-well Maxisorb plates (Nunc, NY) at 4 C overnight. Plates were washed in TBS with 0.1% Tween-20 (TBST), blocked with 10% non-fat dry milk followed by incubation with dilutions of hamster brain homogenate at 4 C for 18 h. Plates were washed, blocked a second time and incubated with m-a-PrP (mAb IPC1; 10K) diluted in TBST with 0.1% IgG-free BSA (Jackson) followed by gt-a-m-HRP (1:10K) for 1 h. Plates were resolved by ECL (Supersignal Femto, Pierce) and chemiluminescence detected using a Wallac Victor II (PerkinElmer, MA).

Histochemistry
Hamster brains were fixed and 1 mm coronal sections cut using a steel hamster brain matrix (EM sciences) and gross morphology imaged using a Leica EZ4D stereo microscope. Thin cyrosections were collected stained with CAT Hematoxylin (Biocare Medical), dehydrated in graded-ETOH, cleared in Xylene, and coverslipped using DPX (EM sciences). Sections were imaged using a Leica DMI400B with an attached DFC320 R2 digital camera.

Statistics
Sigma Stat software was used for statistical analysis (Systat, CA). Kruskal-Wallis one way analysis of variance on ranks was used followed by pairwise comparison using Mann-Whitney U Rank Sum Test.