PrP aggregation can be seeded by pre-formed recombinant PrP amyloid fibrils without the replication of infectious prions

Mammalian prions are unusual infectious agents, as they are thought to consist solely of aggregates of misfolded prion protein (PrP). Generation of synthetic prions, composed of recombinant PrP (recPrP) refolded into fibrils, has been utilised to address whether PrP aggregates are, indeed, infectious prions. In several reports, neurological disease similar to transmissible spongiform encephalopathy (TSE) has been described following inoculation and passage of various forms of fibrils in transgenic mice and hamsters. However, in studies described here, we show that inoculation of recPrP fibrils does not cause TSE disease, but, instead, seeds the formation of PrP amyloid plaques in PrP-P101L knock-in transgenic mice (101LL). Importantly, both WT-recPrP fibrils and 101L-recPrP fibrils can seed plaque formation, indicating that the fibrillar conformation, and not the primary sequence of PrP in the inoculum, is important in initiating seeding. No replication of infectious prions or TSE disease was observed following both primary inoculation and subsequent subpassage. These data, therefore, argue against recPrP fibrils being infectious prions and, instead, indicate that these pre-formed seeds are acting to accelerate the formation of PrP amyloid plaques in 101LL Tg mice. In addition, these data reproduce a phenotype which was previously observed in 101LL mice following inoculation with brain extract containing in vivo-generated PrP amyloid fibrils, which has not been shown for other synthetic prion models. These data are reminiscent of the “prion-like” spread of aggregated forms of the beta-amyloid peptide (Aβ), α-synuclein and tau observed following inoculation of transgenic mice with pre-formed seeds of each misfolded protein. Hence, even when the protein is PrP, misfolding and aggregation do not reproduce the full clinicopathological phenotype of disease. The initiation and spread of protein aggregation in transgenic mouse lines following inoculation with pre-formed fibrils may, therefore, more closely resemble a seeded proteinopathy than an infectious TSE disease. Electronic supplementary material The online version of this article (doi:10.1007/s00401-016-1594-5) contains supplementary material, which is available to authorized users.

2 Animal and Plant Health Agency, Pentlands Science Park, Midlothian.  This work used 3 different refolding methods to produce recPrP in a physiological α-helical conformation, a β-oligomeric conformation or an amyloid fibril form. This, in turn, necessitated following 3 different purification protocols that were prevalent in the literature at the time when these preparations were made. These protocols are described, in detail, in the following section. Each protocol was applied to the purification of both wildtype (WT-recPrP) and 101L mouse PrP (101L-recPrP). The subsequent data relates to the actual purifications used to generate the protein preparations that were inoculated into mice. In each case, we present chromatographic traces of steps carried out on AKTA FPLC and HPLC instruments, where protein elution was monitored by UV absorbance at 280 nm. Where appropriate, we also monitored conductivity. The percentage of the "B" buffer is given to indicate when elution starts. In almost all cases, SDS-PAGE gels are presented tracking the elution of the protein, apart from 1 purification were gels are not available. On the gels, samples highlighted with an asterisk were those taken forward to the next stage of purification or for refolding. In all cases, MoPrP migrates to ~25 kDa, consistent with its calculated molecular weight of 23 kDa.

PrP expression & lysis
For all preparations Escherichia coli expression bacteria (Rosetta strain) were grown (37 °C, shaking at 225 rpm) in 400 ml volumes of Terrific Broth (Sigma) containing ampicillin, to an OD600 of between 0.6-1.0. PrP expression was induced by addition of isopropyl thiogalactoside (IPTG) to a final concentration of 1 mM. Cells were left for at least 4 hours and up to overnight (~16 hours) and were harvested by centrifugation (15 min, 12k rpm, SLA1500 rotor). Cells were resuspended in 10 ml of lysis buffer (50 mM Tris-HCl pH 8, 100 mM sodium chloride, 1 mM ethylenediaminetetraacetic acid (EDTA)) per gram of cell pellet. Lysozyme was added to a final concentration of 20 μg/ml final and the suspension was incubated at 4 °C for 30 min. Sodium deoxycholate (1 mg/ml final) and DNase (5 μg/ml final) were added and 1 M MgSO4 was added to a final concentration of 1 mM. After 30 minutes incubation at room temperature, the inclusion bodies were isolated by centrifugation (15 min, 12k rpm, SLA1500 rotor).

Purification of PrP for refolding to α-monomeric protein
This purification regime was standard in our laboratory to produce protein that was natively folded. Inclusion bodies were solubilised in 10 ml of IMAC buffer A (100 mM sodium phosphate, 10 mM Tris, 8 M urea, 10 mM β-mercaptoethanol, pH 8.0) per gram of pellet, at room temperature for 1-2 hours using constant stirring. Insoluble material was removed by centrifugation (15 min, 12k rpm, SLA1500 rotor) and the supernatant was applied to a home-poured Nickel NTA affinity column (Qiagen). Bound material was eluted by a step elution to IMAC buffer B (100 mM sodium phosphate, 10 mM Tris, 8 M urea, 10 mM 2-mercaptoethanol, pH 4.5). PrP-containing fractions, as assayed by SDS-PAGE, were pooled and diluted 1:1 with IE A buffer (8 M urea, 50 mM HEPES, pH 8.0). This solution was applied to a home-poured cation exchange column (SP-sepharose, Pharmacia). Bound proteins were eluted by a gradient elution to 50 % IE B buffer (8 M urea, 50 mM HEPES, pH 8.0 containing 1 M NaCl). PrP containing fractions, as assayed by SDS-PAGE, were pooled and the protein was diluted to a concentration of ~0.1 mg/ml by the addition of 8 M urea. 5-fold molar excess of copper ions (CuCl2) were added and the solution was stirred overnight to allow copper-catalysed formation of the disulphide bond. The oxidised protein was dialysed extensively against 50 mM sodium acetate, pH 5.5 (6 changes of buffer, the first two of which contained 1 mM EDTA to remove copper ions) and finally the protein was concentrated by use of an Amicon filtration cell. The final concentrations of WT-recPrP and 101L-recPrP were 0.55 and 1.35 mg/ml respectively. The proteins contained predominately αhelical structure as confirmed by circular dichroism spectropolarimetry. Prior to inoculation the proteins were filtered (0.22 µm) and the final concentrations of protein were 0.48 and 1.33 mg/ml for WT-recPrP and 101L-recPrP protein respectively. . Inclusion bodies were re-suspended in 10 ml/g IMAC buffer A (9 M urea, 0.1 M sodium phosphate, 0.01 M Tris-HCl (pH 8)). The suspension was repeatedly vortexed and stirred continuously to aid protein solubilisation. Solubilised inclusion bodies were centrifuged (15 min, 12k rpm, SLA1500 rotor), to remove the insoluble fraction. The supernatant was applied to a home-poured column containing Ni-NTA resin (Qiagen) polypropylene column (Qiagen) and bound protein species were eluted with IMAC buffer B (500 mM imidazole, 0.02 M MOPS (pH 7)). recPrP-containing fractions were buffer-exchanged using a PD-10 desalting column (GE Healthcare) equilibrated with 20 mM sodium citrate (pH 3.4). The eluate was collected in 500 µl fractions and fractions containing the most recPrP were pooled. Protein concentrations at this stage were 1.87 and 1.56 mg/ml respectively for WT-recPrP and 101L-recPrP, and the concentration of WT-recPrP was adjusted to that of the 101L-recPrP by dilution in sodium citrate. The protein samples were incubated at 45 °C and 10 µl samples were analysed by size exclusion chromatography by use of a 300 x 7.8 nm TSKG4000SWxl HPLC column equilibrated with 20 mM sodium citrate (pH 3.4) at a flow rate of 1 ml/min. The eluate was monitored by UV absorbance at 280 nm. Overnight incubation in sodium citrate buffer allowed formation of oligomers, as shown by a shift to smaller elution volumes. After filtration for sterilisation, final concentrations of oligomeric preparations were 0.44 and 0.33 mg/ml for WT-recPrP and P101L-recPrP respectively.

Purification of PrP for refolding to amyloid fibrils
This Inclusion body pellets were resuspended in IMAC buffer A (8 M urea, 100 mM sodium phosphate, 10 mM tris-HCl pH 8 containing 10 mM reduced glutathione). PrP was purified using IMAC chromatography (Ni-NTA superflow, Qiagen) eluting components bound to the column with a linear gradient from 0 to 100 % IMAC buffer B (8 M urea, 100 mM sodium phosphate, 10mM tris-HCl pH 4.5 containing 10mM reduced glutathione). Protein-containing fractions, as determined by Coomasie-stained SDS-PAGE gels, were pooled and desalted by use of a Superdex 75 column (GE Healthcare) eluting protein with a solution containing 6 M urea, 0.1 M Tris-HCl, pH 7.5. Oxidised glutathione (50 mM) was added to protein containing fractions to a final concentration of 0.2 mM and EGTA to a final concentration of 5mM and the solution was left overnight to oxidise the single disulphide bond. Oxidised protein was purified further by reversed phase HPLC (in batches to avoid overloading the column) by dilution of the protein two-fold into RP-HPLC buffer A (0.1 % (v/v) trifluoroacetic acid in water). The protein was loaded onto a reversed phase column (214TP101522, Vydac) by use of a superloop (GE Healthcare) and bound protein was eluted by an increasing gradient of RP-HPLC buffer B (0.1 % (v/v) trifluoroacetic acid in acetonitrile). Monomeric, full length recombinant PrP eluted at ~35 % RP-HPLC buffer B and the fraction from the peak of the elution profile was lyophilised so as to avoid including other chemically modified species, which elute prior to or subsequent to this peak. Elutions from different HPLC batches were pooled prior to lyophilisation. Lyophilised protein was stored at -20 °C prior to use.
For fibrillisation, lyophilised recPrP was dissolved in 6 M guanidine HCl pH 6.0 at a concentration of 3 mg/ml. A fibrillisation reaction consisting of 2 M guanidine HCl, 10 mM thiourea, 120 µg/ml recPrP, 50 mM 2-(N-morpholino)ethanesulfonic acid (MES) pH 6.0, 10 µM thioflavin T was prepared. Reaction mixture (160 µl) was placed into individual wells of a 96 well plate and 3 Teflon spheres were added to aid mixing. The plate was shaken at 37 °C on a plate shaker. Aliquots were removed and assayed for fibril formation by thioflavin T fluorescence (excitation at 444 nm and emission at 485 nm) on a plate reader. The presence of fibrils was confirmed by both electron microscopy (see figure 1 in the main paper) and also by the presence of a 16 kDa band following maturation of the fibrils at 80 °C and digestion with proteinase K, as published previously by Breydo et al. Fibrils were dialysed into 50 mM sodium acetate, pH 5.5 for use. Final concentration of fibrils were 0.4 and 0.07 mg/ml respectively for WT-recPrP and P101L-recPrP.