Different tau fibril types reduce prion level in chronically and de novo infected cells

Neurodegenerative diseases are often characterized by the codeposition of different amyloidogenic proteins, normally defining distinct proteinopathies. An example is represented by prion diseases, where the classical deposition of the aberrant conformational isoform of the prion protein (PrPSc) can be associated with tau insoluble species, which are usually involved in another class of diseases called tauopathies. How this copresence of amyloidogenic proteins can influence the progression of prion diseases is still a matter of debate. Recently, the cellular form of the prion protein, PrPC, has been investigated as a possible receptor of amyloidogenic proteins, since its binding activity with Aβ, tau, and α-synuclein has been reported, and it has been linked to several neurotoxic behaviors exerted by these proteins. We have previously shown that the treatment of chronically prion-infected cells with tau K18 fibrils reduced PrPSc levels. In this work, we further explored this mechanism by using another tau construct that includes the sequence that forms the core of Alzheimer’s disease tau filaments in vivo to obtain a distinct fibril type. Despite a difference of six amino acids, these two constructs form fibrils characterized by distinct biochemical and biological features. However, their effects on PrPSc reduction were comparable and probably based on the binding to PrPC at the plasma membrane, inhibiting the pathological conversion event. Our results suggest PrPC as receptor for different types of tau fibrils and point out a role of tau amyloid fibrils in preventing the pathological PrPC to PrPSc conformational change.

Neurodegenerative diseases are often characterized by the codeposition of different amyloidogenic proteins, normally defining distinct proteinopathies. An example is represented by prion diseases, where the classical deposition of the aberrant conformational isoform of the prion protein (PrP Sc ) can be associated with tau insoluble species, which are usually involved in another class of diseases called tauopathies. How this copresence of amyloidogenic proteins can influence the progression of prion diseases is still a matter of debate. Recently, the cellular form of the prion protein, PrP C , has been investigated as a possible receptor of amyloidogenic proteins, since its binding activity with Aβ, tau, and α-synuclein has been reported, and it has been linked to several neurotoxic behaviors exerted by these proteins. We have previously shown that the treatment of chronically prion-infected cells with tau K18 fibrils reduced PrP Sc levels. In this work, we further explored this mechanism by using another tau construct that includes the sequence that forms the core of Alzheimer's disease tau filaments in vivo to obtain a distinct fibril type. Despite a difference of six amino acids, these two constructs form fibrils characterized by distinct biochemical and biological features. However, their effects on PrP Sc reduction were comparable and probably based on the binding to PrP C at the plasma membrane, inhibiting the pathological conversion event. Our results suggest PrP C as receptor for different types of tau fibrils and point out a role of tau amyloid fibrils in preventing the pathological PrP C to PrP Sc conformational change.
Prion diseases are a heterogeneous group of fatal and infectious neurodegenerative pathologies caused by the structural conversion of the cellular prion protein (PrP C ) into the disease-causing scrapie isoform (PrP Sc ), which accumulates mainly extracellularly (1). Human prion diseases are clinically and pathologically diverse, with most of the cases classified as sporadic Creutzfeldt-Jakob disease (CJD), and the remaining ones as genetic, such as familial CJD, Gerstmann-Sträussler-Scheinker disease (GSS), and fatal familial insomnia, and acquired, representing less than 1% of all cases (2).
Several groups reported the presence of phosphorylated tau deposits in GSS patients, assuming neurofibrillary tangles morphology with paired helical filaments (PHFs) (3)(4)(5)(6). In 1997, Tranchant et al. reported the presence of neurofibrillary tangles in a GSS patient affected by the A117V mutation in the prion protein gene. Interestingly, tau deposition in this patient could be related to either the protracted duration of the disease or the patient's age, as the clinical course was the longest in his family (7). Other works highlighted the colocalization between prion plaques and phosphorylated tau deposits in sporadic CJD and variant CJD cases (8)(9)(10).
Several lines of evidence suggest PrP C as the mediator of the neurotoxic effects and the spreading of pathological aggregates that, besides PrP Sc in prion diseases (11), are linked to other neurodegenerative disorders (12). PrP C has been identified as one of the binding partners for synaptotoxic oligomers of amyloid β (Aβ) (13), mediating their cognitive detrimental effects (14). Anti-PrP antibodies preventing the interaction between PrP C and Aβ efficiently hampered the Aβ-mediated disruption of synaptic plasticity (15). Similarly, pathological α-synuclein spreads faster in PrP C -overexpressing mice (16), thus suggesting an involvement of the cellular prion protein in α-synucleinopathies (17). Moreover, our group recently reported a PrP-dependent uptake and toxicity of TAR DNAbinding protein-43 (TDP-43) fibrils (18).
We have previously shown a similar interplay between PrP C and tau amyloids (19). The N-terminal region of PrP C was shown to interact with recombinant tau K18 fibrils and their uptake resulted less efficient in Neuro 2a (N2a) cells knocked out for PrP C . Moreover, tau K18 fibrils promoted the increased localization of PrP C at the plasma membrane.
Interestingly, tau amyloid administration significantly reduced PrP Sc level in N2a cells chronically infected with the RML prion strain, probably through an inhibitory effect on prion conversion. However, we were unable to exclude whether, besides PrP C binding, tau fibrils could affect prion load by other mechanisms.
To further investigate the mechanism responsible for the tau-mediated prion reduction and to compare the effect of distinct conformers of tau fibrils on PrP Sc , we designed a new tau construct that, compared to K18, has a six amino acids longer sequence (amino acids 244-378).

Tau amyloid fibrils reduce prions burden in a time-dependent manner
To better understand the reduction in prion load observed in our previous study (19), we employed, together with tau K18, a second tau construct, herein called tau 244-378. This fragment encompasses the sequence shown to be important for the packing interface of the core of Alzheimer's disease (AD) tau PHFs and straight filaments (SFs) (20). The addition of the six C-terminal amino acids to tau K18 (amino acids 244-372) was performed using the restriction-free cloning technique, and the protein was expressed and purified as previously described for tau K18 (19) (Fig. S1A). To produce in vitro amyloid fibrils, both tau proteins were subjected to the same fibrillization protocol (19). As shown in Fig. S1B, both tau K18 and 244-378 showed comparable aggregation kinetics, even if K18 reached a higher thioflavin T (ThT) fluorescence signal compared to 244-378. The end-stage products of the fibrillization process were characterized by transmission electron microscopy (TEM) after sonication to break long fibrils into smaller species. TEM analysis showed that both constructs form fibrillar structures in vitro (Fig. S1C). To further characterize tau 244-378 fibrils, both fibrillization products were subjected to digestion with increasing concentrations of proteinase K (PK), followed by Western blot (WB) analysis, as it was already shown that the degree of resistance to proteases and the digestion pattern correlate with the structural organization of both in vitro and brain-isolated tau amyloids (21)(22)(23). As shown in Fig. S2A, both constructs form fibrils characterized by the resistance to PK digestion, highlighted by the presence of species with high molecular weight. In particular, tau K18 fibrils showed a higher resistance to PK digestion than tau 244-378. These latter were completely digested at concentrations higher than 1 μg/ml after 1 h of incubation at 37 C (Fig. S2A). In contrast, tau K18 fibrils were not completely digested neither at the highest concentration of PK used (100 μg/ml) nor after increasing the time of the treatment up to 24 h (Fig. S2, B and C). These results suggest that the two tau constructs acquire distinct conformations when subjected to in vitro fibrillization. 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay in RML prion-infected N2a (ScN2a RML) cells showed that tau 244-378 fibrils caused a significant alteration in their metabolic activity, even at the lowest concentration (Fig. S3A), while tau K18 did not (19). However, as we did not observe major effects on cell proliferation and viability at 2 μM, also 244-378 fibrils were used in cell culture experiments (Fig. S3B).
As previously shown, tau K18 fibril treatment caused a reduction in PrP Sc levels after 72 h of continuous incubation (19). To evaluate whether PrP Sc reduction needs 72 h of continuous incubation with tau fibrils or whether tau fibrils act earlier, we tested different tau K18 incubation times (72 h, 48 h, 24 h, 8 h, 4 h, 2h, and 1 h) before treatment removal and culturing up to 72 h. The PK-resistant PrP Sc content was evaluated by WB. As shown in Figure 1, A-D, shorter incubation times with tau K18 fibrils caused a reduction in PrP Sc levels at 72 h, suggesting that tau effect starts before the clearance of PrP Sc appears.
To investigate whether the treatment with tau 244-378 fibrils resulted in the same prion reduction, a comparison with K18 was also performed. For this experiment, we chose 4 h of incubation, since it was the lowest K18 incubation time showing around 50% of PrP Sc reduction. As shown in Figure 2, A (upper panel) and B, both tau species were able to reduce PrP Sc level. We also checked whether tau amyloids were actively internalized by cells, as they could act on intracellular degradation mechanisms to promote PrP Sc clearance. Therefore, we evaluated in the same samples the presence of intracellular tau amyloids at 72 h. Extracellular tau fibrils were removed by trypsinization to evaluate the signal coming exclusively from intracellular fibrils (24)(25)(26)(27). Interestingly, following 4 h of incubation, cells treated with K18 fibrils showed a strong tau signal corresponding to the intracellular aggregates which are still present after 72 h of culturing, while a faint tau signal was detected in cells treated with 244-378 fibrils ( Fig. 2A lower panel). However, the cells treated with tau 244-378 fibrils, checked right after 1, 2, 4, 8, and 24 h of incubation, showed higher amounts of internalized aggregates (Fig. 2C), demonstrating that both fibril types are actively internalized by cells.
Tau fibrils are internalized and localize in the autophagolysosomal pathway To further characterize the PrP Sc clearance mediated by tau fibrils, we assessed their localization following the uptake in ScN2a cells, as it may underlie specific pathways involved in the degradation of both tau fibrils and prions. Tau K18 fibrils were conjugated to Alexa-488 dye to produce fluorescent fibrils that were administered to cells for 24 h to obtain a significant internalization. Cells were then stained with antibodies targeting several cell compartments. Trypan blue was used before fixation to quench the signal of fibrils that were not internalized, as this dye cannot enter living cells (28). Images of a single cell were captured as multiple sections along the z-axis to take into account the whole cell volume and shown in twodimensional with orthogonal planes, to assess the colocalization both in the XZ and YZ axes. As shown in Fig. S4, tau K18 fibrils are efficiently internalized after 24 h of incubation, localizing in subcellular compartments such as Golgi apparatus (M6PR), autophagosomes (LC3B), and lysosomes (LAMP2). No colocalization was observed with the early endosome marker EEA1 and the endoplasmic reticulum one (Calnexin).
To investigate the potential role of the lysosomal pathway in mediating prion clearance induced by tau K18 fibrils, we took advantage of different inhibitors known to block the autophagolysosomal pathway at different steps. Despite the efficient inhibition of the autophagolysosomal pathway, as shown by the levels of LC3B proteins, neither 3-methyladenine, bafilomycin A1, nor chloroquine were able to impair the clearance of PrP Sc mediated by tau K18 fibrils (Fig. S5, A-C). At the same time, the treatment with MG132, known to inhibit proteasomal degradation, had no effect (Fig. S5D), suggesting that the most common degradation pathways, normally implicated in the removal of misfolded proteins, are not involved in the tau amyloid fibrils-mediated PrP Sc reduction.
Tau fibril uptake is not required for PrP Sc reduction We have shown that tau fibrils efficiently reduce prion level in ScN2a RML cells, and they are internalized by cells, localizing in different subcellular compartments. However, if internalization is involved, there should be other mechanisms by which tau fibrils promote PrP Sc reduction, since we have already ruled out the contribution of cellular degradation pathways. To assess whether tau fibril internalization is somehow related to PrP Sc reduction, we incubated ScN2a RML cells at 4 C. At this restrictive temperature, all the energy-dependent endocytic processes are inhibited, but tau amyloid fibrils are still able to bind to the plasma membrane (24). As already shown for the uninfected N2a cells (19), also for ScN2a RML, the incubation at 4 C almost completely inhibited the uptake of Alexa488-labeled tau K18 fibrils (Fig. S6A). We then evaluated the PrP Sc reduction at 72 h, after 4 h of incubation with tau K18 fibrils both at 37 C and 4 C. As shown in Fig. S6B, WB analysis of intracellular tau K18 fibrils at 4 C revealed an almost complete absence of signal when compared to 37 C, confirming the blockage of tau fibril endocytosis and the validity of our experimental design. Values are shown as percentage of PrP Sc relative to β-actin. Data are reported as mean ± SD and were analyzed with Friedman test with Dunn's multiple comparisons test: *p ≤ 0.05. N2a, Neuro 2a; PK, proteinase K; PrP Sc , isoform of the prion protein; ScN2a RML, RML prion-infected N2a; WB, Western blot.
However, no major differences were observed in PrP Sc clearance, with a similar decrease in K18-treated cells either at 37 C or 4 C (Fig. 3, A and B). This result suggests that despite their internalization, tau K18 fibrils do not mediate PrP Sc reduction by entering the cells.
We then assessed whether tau 244-378 fibrils, characterized by different biochemical and biological behaviors compared to K18, may act on PrP Sc reduction with similar mechanisms. As previously shown, tau 244-378 fibrils are less resistant to protease degradation compared to K18 fibrils, and despite being quickly internalized, they are efficiently removed after 3 days of incubation. To overcome this problem, the blocking of tau 244-378 fibril internalization at 4 C was performed by incubating cells for 4 h followed by cell lysis to prevent cells from eliminating fibrils. Fig. S6C (right lanes) shows how, as for tau K18 fibrils, 244-378 fibril internalization is efficiently reduced by incubating cells at 4 C. Moreover, the complete absence of tau signal when, after the treatment, cells are left in incubation up to 72 h shows that after the initial uptake, 244-378 fibrils are efficiently removed by cells, probably degraded (Fig. S6C, left lanes). However, after 3 days, the 244-378-mediated reduction in PrP Sc levels between cells incubated at 37 C and 4 C was comparable (Fig. 3, C and D).
These results show that despite being internalized and exhibiting different biochemical and biological properties, both K18 and 244-378 fibrils reduce prion level independently of their uptake.
Tau amyloid fibrils bind to PrP C and hinder PrP Sc de novo infection Since the effect of tau fibrils on prion clearance is independent from their internalization, their action could be exerted in the extracellular environment. Being mainly located on the outer leaflet of the plasma membrane, PrP C may represent the target of tau amyloid fibrils used in this study. As previously shown, the administration of tau K18 fibrils to N2a cells caused an increased localization of PrP C in the plasma membrane, probably due to its impaired recycling and its stabilization in this compartment (19). To confirm the binding observed between PrP C and tau amyloid fibrils (19,29), we performed an enzyme-linked immunosorbent assay (ELISA) experiment using both tau fibril types. As shown in Fig. S7, both tau K18 and 244-378 fibrils bind to the recombinant mouse PrP 23-231 with an average EC 50 of 102.8 and 174.47 nM respectively, whereas no appreciable affinity was detected with the bovine serum albumin used as negative control. This result confirms the ability of PrP C to bind not only amyloid aggregates of different proteins (29) but also distinct fibril types of tau, suggesting that the reduction in PrP Sc level, observed so far in our cellular model, could derive from a direct interaction between tau amyloids and PrP C , making the latter unavailable for the conversion into PrP Sc .
To reproduce the effect of tau fibrils on PrP Sc -mediated PrP C conversion, we took advantage of an in vitro real-time quaking-induced conversion (RT-QuIC) assay. Recombinant mouse PrP 23-231 (rPrP) was preincubated with tau K18 or 244-378 fibrils in 5:1 and 1:1 M ratios, before seeding with 1 ng of phosphotungstate precipitated PrP Sc . ThT fluorescent dye was added to monitor the real-time aggregation process. As shown in Fig. S8, both tau K18 (upper panel) and 244-378 (lower Tau fibrils slow down prion replication in cells panel) fibrils hinder the PrP Sc -mediated rPrP conversion in a molar concentration-dependent manner.
However, besides the binding to PrP C , a direct interaction between tau fibrils and PrP Sc cannot be ruled out.
Regardless of the binding to either PrP C or PrP Sc , tau fibril administration should be able to hinder the conversion of PrP C also in uninfected cells.
To test this hypothesis, we took advantage of a de novo prion infection protocol, which allows PrP C to PrP Sc conversion in uninfected cells using an inoculum obtained from prion-infected cells (30). To assess the contribution of tau fibrils in preventing the infection, N2a cells were pretreated for 4 h with 2 or 4 μM of tau K18 fibrils, washed with PBS to remove the treatment, and then exposed for 72 h to ScN2a RML cell lysate. The efficient de novo cell infection is shown by the increase of PrP Sc content from passage 2 to passage 3, suggesting that cells are actively replicating prions (Fig. 4, A and C).
Tau K18 fibril pretreatment, either at 2 μM and 4 μM, was efficiently able to reduce prion accumulation in both passages, confirming the impairment of the de novo infection process (Fig. 4, A and B).
As expected, a similar result was obtained performing the de novo infection experiment with tau 244-378 fibrils (Fig. 4, C and D).

Discussion
Although Aβ, tau, α-synuclein, prion, and TDP-43 aggregates are usually involved in different pathologies, their accumulation often occurs in nondiseased individuals, and they can codeposit in the brain of diseased patients (31). The same also applies to prion diseases, with described cases of codeposition of PrP Sc with Aβ (10), α-synuclein (32) and phosphorylated tau (3)(4)(5)(6)(7)(8)(9)(10). On this regard, it is now widely recognized that the cellular form of the prion protein, PrP C , binds different βsheet-enriched amyloid proteins, suggesting its affinity to a common feature of these amyloids. Recently, Corbett et al.
showed that PrP C binds to soluble aggregates of tau, α-synuclein, and Aβ and that its genetic ablation reduced the toxic fibrils and then infected with ScN2a RML homogenate. Cells were passaged for three times to allow prion replication. β-actin of the same samples not treated with PK was used as loading control. D, graph shows the quantification of three independent experiments. Values are shown as percentage of PrP Sc relative to β-actin and data reported as mean ± SD. Data were analyzed with two-way ANOVA with Sidak's multiple comparisons test: ****p ≤ 0.0001. N2a, Neuro 2a; PK, proteinase K; PrP Sc , isoform of the prion protein; ScN2a RML, RML prion-infected N2a; WB, Western blot. effects that these aggregates normally exert (29). Similarly, our lab has shown an interplay between PrP C and α-synuclein amyloid fibrils, with an increased uptake and spreading of αsynuclein aggregates in the presence of PrP C , both in vitro and in vivo (17). Moreover, the N terminus of PrP C was found to interact with tau K18 amyloid fibrils, facilitating their uptake in N2a cell line (19). Interestingly, α-synuclein, tau, and TDP-43 fibrils were able to reduce the amount of PrP Sc when administered to prion-infected N2a cells, possibly by hindering the conversion event of PrP C . To further explore this effect, we took advantage of the previously used tau K18 construct, made of the amino acids 244-372, and of a new tau construct, which extends the tau K18 sequence of six amino acids (244-378). This fragment encompasses the sequence shown to be important for the packing interface of the core of AD tau PHFs and SFs, and it could result in the formation of filaments with a different structure compared to those of K18. TEM analysis of the end-stage products revealed that as for tau K18, also 244-378 forms fibrils in vitro. We further characterized 244-378 fibrils by subjecting them to protease degradation, as the resistance to proteolysis represents one of the features of in vitro and brain-extracted amyloid proteins (21)(22)(23)33), and it correlates with the structural organization and the amount of β-sheets (34). Surprisingly, we found that 244-378 fibrils, unlike the K18 ones, were very sensitive to PK digestion, suggesting different conformational structures and different β-sheet content, that may explain also the lower ThT fluorescence emission when fibrillized in vitro. This result may also suggest that 244-378, comprising the in vivo PHFs and SFs AD sequence, may reflect a more physiological conformational structure, as it was already shown that in vivo filaments are less resistant to protease degradation when compared to the in vitro produced ones (23,35). First, we tested the metabolic effect of 244-378 fibrils in our cellular model. In contrast to tau K18 fibrils, tau 244-378 fibrils exhibited a marked effect in the reduction of the metabolic activity in ScN2a RML cells, further confirming not only the biochemical but also the different biological nature of these two amyloid preparations. However, at 2 μM concentration, no effects on cell proliferation and viability were observed.
As previously shown, tau K18 fibrils reduce PrP Sc level after 72 h of incubation (19). Here, we have shown that both tau K18 and 244-378 fibrils were spontaneously internalized by these cells and caused a comparable reduction effect on prion load, despite the 244-378 signal was almost absent after 72 h, probably because of their degradation. Their active digestion by cells would be in agreement with the lower resistance of 244-378 fibrils to protease degradation, and it may also explain the reduced metabolic activity observed in the MTT assay. We also observed that, following the internalization, tau K18 fibrils localized in the Golgi apparatus and the autophagolysosomal pathway, suggesting a possible stimulation of the cellular degradation pathways and a consequent decrease in PrP Sc levels. However, the treatment with inhibitors, aimed at blocking the autophagic and proteasomal degradation, had no effects, thus ruling out a role of these pathways in the tau fibrils-mediated PrP Sc clearance. Unexpectedly, we observed that tau fibril internalization did not affect their ability to reduce PrP Sc , as prion reduction was fully comparable between cells incubated at 37 C or 4 C. These findings strongly suggest that tau fibrils-dependent PrP Sc clearance is mediated by a direct binding between PrP C /PrP Sc and tau fibrils at the plasma membrane, hindering one of the two partners necessary for prion conversion. We already observed the increased PrP C clustering at the plasma membrane following tau fibril exposure (19). By exploiting their action on the plasma membrane, the binding of tau fibrils with PrP Sc seems unlikely, since the latter is mostly present at the intracellular level (36,37) with little localization at the plasma membrane (38). We demonstrated here that both tau K18 and 244-378 fibrils efficiently bind to the recombinant full-length PrP, with EC 50 values slightly higher than the ones found for other amyloid species (29), confirming the ability of the cellular prion protein to act as a receptor not only for different amyloidogenic proteins but also for distinct types of tau fibrils. This result may suggest that the interaction between tau fibrils and PrP C reduces PrP Sc levels by sequestering one of the two partners needed for the infectious cycle. Indeed, both tau amyloids efficiently hindered the PrP Sc -mediated rPrP conversion in an in vitro RT-QuIC assay, thus confirming their interference in the prion infection process. After the observation of tau fibrilsmediated reduction of PrP Sc level in chronically prion-infected cells, we wondered whether this effect is limited to an established prion infection or whether tau amyloids can somehow affect the early stages of prion infection. To test this hypothesis, we took advantage of a de novo prion infection protocol, by exposing uninfected N2a cells to a homogenate of ScN2a cells. In this way, the PrP Sc present in the inoculum converts the PrP C of the uninfected N2a cells. After the initial infection step, cells were passaged three times to allow the removal of the initial inoculum and the onset of the de novo prions. The first passage was not analyzed to avoid the interference of the initial inoculum. Our results indicated that both tau K18 and 244-378 fibrils, administered before the exposure of cells to prion homogenate, reduced significantly the accumulation of PrP Sc following the de novo infection.
In addition, our de novo experiment suggests that, since the treatment is removed before the addition of PrP Sc inoculum, tau fibril binding to PrP C represents the event through which they exert their effect on PrP Sc . In the early phases of the infection process, tau binding to PrP C could prevent its physical interaction with PrP Sc , inhibiting the conversion, since the addition of tau fibrils is sufficient to hinder this process in vitro. Furthermore, tau fibril binding to PrP C could costrain the protein at the plasma membrane, as demonstrated by its clustering in this compartment following tau fibril exposure (19). This could result in the hampering of PrP C recycling through the endocytic compartments preventing its interaction with PrP Sc and the progress of the infectious cycle. During the established infection, tau fibril binding to PrP C may stabilize the protein at the plasma membrane, preventing its interaction with PrP Sc in the endocytic compartments, where the conversion most likely occurs (36,39,40).
In conclusion, our findings confirmed the role of tau fibrils in mediating the reduction of prions in ScN2a RML cells. Moreover, this effect seems to reside in the ability of tau fibrils to bind to the PrP C present on the plasma membrane, thus stabilizing the protein and hindering its conversion to PrP Sc . Here we show that PrP C may act as a receptor for different tau fibril types, that, despite showing different conformational properties, are able to comparably reduce prions in our chronically prion-infected model. In addition, tau amyloid fibrils impair the initial phases of prion conversion, hindering PrP C /PrP Sc interaction and/or stabilizing PrP C at the plasma membrane. These results may suggest a dual interplay between prion protein and other amyloidogenic proteins. While, as previously shown for tau, α-synuclein, and TDP-43, the presence of PrP C at the plasma membrane facilitates the uptake of pathological aggregates possibly accelerating the course of the disease, here we show that tau circulating species may protect PrP C from PrP Sc interaction, thus slowing prion conversion and prion disease progression. This may also explain why in some prion-affected patients, the presence of tau aggregates may correlate with a longer disease incubation. In this regard, understanding the reciprocal interactions between different amyloidogenic proteins and how these relationships may result in distinct disease phenotypes could represent a key step in the development of targeted therapies.

Tau production and fibrillization
Human tau 244-378 construct was obtained using the restriction-free cloning technique. Forward (GGGAGGCGG-CAACAAAAAAATTGAGACCCACAAGCTGACCTTCT) and reverse (GCTTTGTTAGCAGCCGGATCCTTATCAGAA GGTCAGCTTGTGGGT) primers were designed to add six amino acids to the C terminus of tau K18 (amino acids 244-372 of the human tau sequence) in a pET-11a plasmid.
Expression and purification of human tau proteins were performed as previously described (41). Briefly, pET-11a plasmids, containing the sequences encoding for the human tau constructs, were transformed in Escherichia coli BL21 DE3. Cells were grown in 2 L of Luria-Bertani medium supplemented with 100 μg/ml, and the expression of the protein was obtained with 0.8 mM isopropyl β-D-1-thiogalactopyranoside (PanReac AppliChem). Proteins were purified by a first step of cation-exchange chromatography (HiTrap SP-FF, Cytiva) and a second one of size-exclusion chromatography (HiLoad 26-600, Cytiva). Purified proteins were analyzed in sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), lyophilized, and stored at −80 C.
Fibrillization reactions were performed in a 96-well black plate with transparent bottom (BD Falcon) with a 3 mm glass bead (Sigma-Aldrich). The 200 μl final volume mixtures were composed of tau K18/244-378 0.5 mg/ml, 0.1 mM DTT, and 50 μg/ml heparin in PBS 1X pH 7.4. Due to ThT cytotoxic effect, 10 μM of the dye was added only in three wells to monitor the real-time aggregation. Plates were covered with a sealing tape (Fisher Scientific) and incubated at 37 C in orbital shaking (50 s of 400 rpm shaking and 10 s of rest) in FLUOstar Omega Microplate Reader (BMG LABTECH). Fluorescence was monitored every 30 min by bottom reading at 444 nm of excitation and 485 nm of emission. The reaction was stopped after 40 h, and fibrils were pelleted by centrifugation at 186.000g for 1 h at 4 C and resuspended in 1 ml of 1× PBS pH 7.4 before being stored at −80 C. Before use, aliquots were sonicated for 5 min in a sonicator Misonix s3000 at 250 W.

Recombinant PrP production
Mouse PrP 23-231 construct was transformed in E. coli BL21 (DE3) cells (Stratagene), and its expression was induced with 1 mM isopropyl β-D-1-thiogalactopyranoside (PanReac AppliChem). Cells were grown at 30 C for 12 h and then lysed using PandaPLUS 2000. Inclusion bodies containing the recombinant protein were washed several times in bi-distilled water and then dissolved in 8 M guanidine hydrochloride. Protein was purified by size-exclusion chromatography into a pre-equilibrated HiLoad 26/60 Superdex 200 pg column (Cytiva). Protein refolding was performed by dialysis against 20 mM sodium acetate, pH 5.5. The purified protein was analyzed by SDS-PAGE under reducing conditions, dialyzed against PBS 1X, pH 5.8, and stored at −80 C. All salts used were from Sigma-Aldrich.

TEM analysis
Ten microliter of tau K18 or 244-378 fibril solution was dropped onto 200-mesh Formvar carbon-coated nickel grids (Electron Microscopy Sciences) for 20 min after which samples were stained with 25% uranyl acetate replacement (Electron Microscopy Sciences) for 10 min. Before the analysis, the staining solution was removed using Whatman filter paper, and the grids were air-dried for 5 min. Samples were visualized using a FEI Tecnai Spirit Transmission Electron Microscope operating at 120 kV and equipped with an Olympus MegaView G2 camera.

Cell treatments
ScN2a RML or N2a were plated in 6 cm or 10 cm Petri dishes according to the experimental setting. Recombinant preformed fibrils at specific concentrations were directly administered to the medium of cultured cells and incubated for a different amount of time. In some experiments, non-internalized fibrils were removed by trypsinization after washing cells with PBS 1X.
For immunofluorescence experiments, tau K18 fibrils were conjugated to Alexa-488 succinimidyl ester (Thermo Fisher) according to the manufacturer's instructions, and unbound dye was removed by subsequent dialysis (19).
To block tau fibril endocytosis, ScN2a RML cells were preincubated for 10 min at 37 C or 4 C, after which 2 μM of tau fibrils were directly added to the medium, and cells were kept in culture for 4 h at 37 C or 4 C. Then, cells were washed twice with 1X PBS to remove the treatment and kept in culture up to 72 h at 37 C.

Analysis of internalized tau amyloids
For the evaluation of Alexa488-tau K18 fibril internalization, after incubation, cells were washed twice with sterile PBS 1X and incubated for 5 min with a sterile 1:1 solution of trypan blue in PBS. Trypan blue quenches the fluorescence coming from extracellular tau fibrils (28). Cells were then rinsed three times in PBS 1X and fixed with 4% paraformaldehyde for 30 min. Permeabilization was performed for 5 min in 0.1% Triton X-100 in PBS and coverslips stained with HCS Cell-Mask Blue Stain (Thermo Fisher Scientific) 1:2500 labeling whole-cell cytoplasm. Coverslips were mounted with Fluoromount-G (Thermo Fischer Scientific) and stored at 4 C. Images were acquired using a Nikon confocal microscope (Nikon A1plus).

Metabolic cell activity and cell counts
The metabolic influence of tau 244-378 amyloid fibrils was evaluated with MTT assay. 10.000/well ScN2a cells were plated in a 96-well, and the day after, different concentrations of sonicated tau 244-378 fibrils were directly added to the medium, and cells were incubated for 72 h. Cells were then incubated for 3 h at 37 C with 20 μl of 5 mg/ml MTT (Sigma-Aldrich) solution in PBS 1× followed by the solubilization with 1:1 DMSO/2-Propanol solution. The absorbance was measured at 570 nm in Enspire multimode plate reader (Per-kinElmer) with a reference wavelength of 650 nm. Each condition was tested in six replicates and in three independent experiments.
Tau 244-378 fibril effects on proliferation and cell death were assessed by cell counting. 15.000/well ScN2a RML cells were plated in a 12-well, and the day after, they were treated with 2 μM tau 244-378 fibrils. Cells were left in incubation up to 72 h, after which they were detached with trypsin and counted using Scepter 2.0 Handheld Automated Cell Counter (Millipore) with 60 μM sensors. Three independent experiments were conducted, each one in three technical replicates.

De novo prion infection
De novo prion infection of N2a cells was performed as previously described by Arshad et al. (30) with some modifications. Briefly, 150.000 N2a cells were plated in 12-wells, and the day after, pretreated with 2 or 4 μM tau K18/244-378 amyloids for 4 h and washed twice with PBS 1X to remove the treatment before the addition of the seed. PrP Sc seed was prepared scraping a 10 cm Petri dish of ScN2a RML and sonicating cells in ice 2 × 5 s at 70% amplitude (Sonics VCX 130 PB). Total protein content of the lysate was quantified using bicinchoninic acid assay (Sigma-Aldrich), and 100 μg of proteins were added to N2a in a final volume of 500 μl of completed minimal essential medium without penicillin-streptomycin. Cells were kept in incubation with the seed for 72 h after which cells were detached with trypsin and plated 1:2 in a 6 cm Petri dish. Cells were split two more times every 4 days at 1:5 dilution, and the rest was pelleted and resuspended in 50 μl of lysis buffer and analyzed in WB for PrP Sc presence after PK digestion. In all passages, penicillin-streptomycin solution was omitted. PrP Sc amount from passage one was not analyzed to avoid the signal deriving from the initial seed.

Proteinase K digestion
To detect PrP Sc content, 150 μg of total proteins were digested with 20 μg/ml of PK (Sigma-Aldrich) at 37 C for 1 h. The reaction was stopped by adding 2 mM of phenylmethylsulphonyl fluoride (Sigma-Aldrich), and samples were ultracentrifuged for 1 h at 186.000g at 4 C. The resulting pellet was resuspended in 1× loading buffer and boiled for 10 min.

Western blotting
N2a and ScN2a were rinsed in PBS 1X and resuspended in lysis buffer (10 mM Tris HCl, 150 mM NaCl, 0.5% NP-40, 0.5% sodium deoxycholate). Lysates were centrifuged for 5 min at 5900g at 4 C. Total protein content of N2a and ScN2a cells was quantified using bicinchoninic acid assay (Sigma-Aldrich). Proteins (20 μg) were diluted in loading buffer and samples boiled for 10 min at 100 C.

ELISA assay
The binding between tau fibrils and the recombinant prion protein was assessed by ELISA. 96-well plates were coated overnight at 4 C with 500 ng of K18 or 244-378 tau fibrils. Plates were washed three times with PBS + 0.05% Tween-20 (PBS-T) and blocked with 5% bovine serum albumin in PBS for 1 h at RT before the addition of different concentrations of recombinant mouse PrP  . After three subsequent washes with PBS-T, mouse anti-PrP W226 1 μg/ml was incubated for 1 h at RT, and then plates were incubated with goat-antimouse IgG horse-HRP-conjugated antibody (Dako) for 1 h. Plates were developed with 50 μl of 3,3 0 ,5,5 0 -tetramethylbenzidine (Sigma-Aldrich), stopped by the addition of 1 M H 2 SO 4 and read at 450 nm and 570 nm in Enspire multimode plate reader (PerkinElmer).

RT-QuIC
To test the effect of tau fibrils on PrP conversion, recombinant PrP 23-231 was preincubated for 30 min at RT with both tau K18 and 244-378 fibrils at 1:1 and 5:1 M concentration. The reaction mixture was then completed with 130 mM NaCl, 1 mM EDTA, 0.002% SDS, and 10 μM ThT in PBS 1X pH 7.4. All solutions were filtered with a 0.22 μM filter before use. In the seeded conditions, PrP Sc was obtained by PTA precipitation (45), after which, 1 ng of seed was added. A 3 mm glass bead was added in each well. After sealing, the plate was incubated in FLUOstar Omega Microplate Reader (BMG LABTECH) at 37 C, with cycles of 60 s shaking at 600 rpm and 60 s rest. ThT was used to monitor the real-time aggregation, and the fluorescence intensity was measured every 30 min by using 450 ± 10 nm (excitation) and 480 ± 10 nm (emission) wavelengths.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 8.0 software. Values were expressed as mean ± standard deviation (SD). Values are shown as percentage of PrP Sc /β-actin and compared to control group. Groups were analyzed with Friedman test (Dunn's multiple comparisons test) and twoway ANOVA (Sidak's multiple comparisons test). p-values ≤ 0.05 were considered statistically significant.

Data availability
All data generated or analyzed during this study are included in this article (and its supporting information file).
Supporting information-This article contains supporting information.