Radotinib Decreases Prion Propagation and Prolongs Survival Times in Models of Prion Disease

The conversion of cellular prion protein (PrPC) into pathogenic prion isoforms (PrPSc) and the mutation of PRNP are definite causes of prion diseases. Unfortunately, without exception, prion diseases are untreatable and fatal neurodegenerative disorders; therefore, one area of research focuses on identifying medicines that can delay the progression of these diseases. According to the concept of drug repositioning, we investigated the efficacy of the c-Abl tyrosine kinase inhibitor radotinib, which is a drug that is approved for the treatment of chronic myeloid leukemia, in the treatment of disease progression in prion models, including prion-infected cell models, Tga20 and hamster cerebellar slice culture models, and 263K scrapie-infected hamster models. Radotinib inhibited PrPSc deposition in neuronal ZW13-2 cells that were infected with the 22L or 139A scrapie strains and in cerebellar slice cultures that were infected with the 22L or 263K scrapie strains. Interestingly, hamsters that were intraperitoneally injected with the 263K scrapie strain and intragastrically treated with radotinib (100 mg/kg) exhibited prolonged survival times (159 ± 28.6 days) compared to nontreated hamsters (135 ± 9.9 days) as well as reduced PrPSc deposition and ameliorated pathology. However, intraperitoneal injection of radotinib exerted a smaller effect on the survival rate of the hamsters. Additionally, we found that different concentrations of radotinib (60, 100, and 200 mg/kg) had similar effects on survival time, but this effect was not observed after treatment with a low dose (30 mg/kg) of radotinib. Interestingly, when radotinib was administered 4 or 8 weeks after prion inoculation, the treated hamsters survived longer than the vehicle-treated hamsters. Additionally, a pharmacokinetic assay revealed that radotinib effectively crossed the blood–brain barrier. Based on our findings, we suggest that radotinib is a new candidate anti-prion drug that could possibly be used to treat prion diseases and promote the remission of symptoms.


Introduction
Prion diseases are progressive, incurable, and fatal neurodegenerative disorders that are caused by the conversion of cellular prion protein (PrP C ) into pathogenic prion isoforms (PrP Sc ) and the mutation of PRNP. PrP C is soluble and sensitive to proteases, but misfolded PrP Sc is insoluble, forms aggregates, and is partially resistant to proteases, especially proteinase K (PK). Pathologically, prion diseases are characterized by spongiform changes, neuronal loss, glial cell activation, and misfolded PrP Sc aggregation in the central nervous system [1]. These diseases are currently known to be transmissible zoonotic diseases, such as variant Creutzfeldt-Jakob disease (vCJD) (transmitted from bovines to humans) and chronic wasting disease (transmitted between cervids) [2]. Different human prion

Radotinib Decreases the Levels of Pathogenic Prion Isoforms in Prion-Infected Cells
In our laboratory, we have generated a ZW13-2 neuronal cell line [18] that continuously produces PK-resistant prion proteins (PrP Sc ) due to 22L and 139A scrapie agent infection. The cell model enables a relatively rapid assessment of the drug's effect on PrP Sc levels.
To investigate the effect of radotinib on the uninfected and 22L or 139A scrapie-infected ZW13-2 neuronal cells (called ZW13-2-22L or ZW13-2-139A cells), the viability of these cells was monitored after exposure to radotinib (Supplementary Figure S1). No decrease in viability was observed after exposure to 40 µM radotinib; therefore, we used concentrations below 40 µM. We found that radotinib treatment led to a mild reduction in PrP Sc levels in the ZW13-2-22L and ZW13-2-139A cells ( Figure 1A,B, respectively), indicating that radotinib may inhibit PrP Sc propagation or eliminate PrP Sc in these cell models.
concentrations below 40 μM. We found that radotinib treatment led to a mild reduction in PrP Sc levels in the ZW13-2-22L and ZW13-2-139A cells ( Figure 1A,B, respectively), indicating that radotinib may inhibit PrP Sc propagation or eliminate PrP Sc in these cell models. Figure 1. Radotinib treatment led to a mild reduction in PrP Sc levels in neuronal cells with sustained prion production. ZW13-2-22L (A) and ZW13-2-139A (B) cells were incubated with radotinib (0-40 μM) for 24 h. Prion protein (PrP) levels after proteinase K treatment (PK, 5 μg/mL, 37 °C, 1 h incubation) or control were assessed by Western blotting with an anti-PrP antibody (3F10) [19]. Molecular masses in kDa are indicated on the left-hand side. β-actin was used as a loading control.

Radotinib Inhibits Prion Propagation in the Ex Vivo Cerebellar Tissue Slice Culture Model
Above, we showed the effect of radotinib in sustained PrP Sc -produced cell models. The limitation of the cell model seen above is its inability to fully represent prion diseases. Particularly, to observe the effects of the drug, it is necessary to monitor for a longer period of time. Therefore, to evaluate the effect of radotinib on PrP Sc propagation, ex vivo Tga20 mouse (transgenic mice characterized by the 5-10-fold overexpression of mouse PrP) [20] and hamster cerebellar tissue culture models were established [17]. To determine whether cerebellar tissue culture with scrapie infection was successfully established, we confirmed that cerebellar cell death, accompanied by neuronal cell death and PrP Sc accumulation, was increased in 22L scrapie-infected Tga20 cerebellar tissues compared to controls (Supplementary Figure S2). In these ex vivo models, we found that treatment with 25 μM radotinib reduced PrP Sc propagation in 22L scrapie-infected Tga20 cerebellar tissue slices that were cultured for 3 or 5 weeks (Figure 2A,B). In addition, radotinib reduced PrP Sc propagation in 263K scrapie-infected hamster cerebellar tissue slices in a dose-dependent manner ( Figure 2C). Radotinib treatment led to a mild reduction in PrP Sc levels in neuronal cells with sustained prion production. ZW13-2-22L (A) and ZW13-2-139A (B) cells were incubated with radotinib (0-40 µM) for 24 h. Prion protein (PrP) levels after proteinase K treatment (PK, 5 µg/mL, 37 • C, 1 h incubation) or control were assessed by Western blotting with an anti-PrP antibody (3F10) [19]. Molecular masses in kDa are indicated on the left-hand side. β-actin was used as a loading control.

Radotinib Inhibits Prion Propagation in the Ex Vivo Cerebellar Tissue Slice Culture Model
Above, we showed the effect of radotinib in sustained PrP Sc -produced cell models. The limitation of the cell model seen above is its inability to fully represent prion diseases. Particularly, to observe the effects of the drug, it is necessary to monitor for a longer period of time. Therefore, to evaluate the effect of radotinib on PrP Sc propagation, ex vivo Tga20 mouse (transgenic mice characterized by the 5-10-fold overexpression of mouse PrP) [20] and hamster cerebellar tissue culture models were established [17]. To determine whether cerebellar tissue culture with scrapie infection was successfully established, we confirmed that cerebellar cell death, accompanied by neuronal cell death and PrP Sc accumulation, was increased in 22L scrapie-infected Tga20 cerebellar tissues compared to controls (Supplementary Figure S2). In these ex vivo models, we found that treatment with 25 µM radotinib reduced PrP Sc propagation in 22L scrapie-infected Tga20 cerebellar tissue slices that were cultured for 3 or 5 weeks (Figure 2A,B). In addition, radotinib reduced PrP Sc propagation in 263K scrapie-infected hamster cerebellar tissue slices in a dose-dependent manner ( Figure 2C).

Figure 2.
The effect of radotinib treatment on PrP Sc deposition in 22L and 263K scrapie-infected cerebellar slice culture models. (A,B) Sliced Tga20 cerebellar tissues were exposed to 1% 22L scrapie agent for 1 h. Culture medium including radotinib (25 μM) was changed three times a week, and the cultures were maintained for 3 (A) or 5 (B) weeks. The cultured tissues were lysed using RIPA buffer, and then, PrP levels after proteinase K treatment (PK, 3 μg/mL, 37 °C, 1 h incubation) or control were assessed by Western blotting with an anti-PrP antibody (3F10). (C) Hamster cerebellar tissue slices were exposed to 1% uninfected or 263K scrapie-infected hamster brain homogenates for 1 h and then treated with radotinib (0-40 μM) for 5 weeks. PrP levels after PK treatment (1 μg/60 μg of total proteins, 37 °C, 1 h incubation) were assessed by Western blotting with an anti-PrP antibody (3F10). Molecular masses in kDa are indicated on the left-hand side. β-actin was used as a loading control.
To determine whether radotinib exerts a practical effect during disease progression, we designed two experiments ( Figure 3A). Hamster cerebellar tissue slices that were infected with the 263K scrapie strain (1) were cultured for 5 weeks in the presence of radotinib ( Figure 3B, lanes 10-12) or (2) were cultured for 2 weeks in the absence of radotinib treatment followed by 3 weeks in culture medium containing radotinib, which was changed three times a week ( Figure 3B, lanes 3, 4, 7-9). Then, the cerebellar tissue slices were lysed and subjected to immunoblotting, and we found that the levels of PrP Sc were decreased after 5 weeks of treatment with radotinib in a dose-dependent manner ( Figure  3B). Interestingly, later treatment with 10-40 μM radotinib also dramatically reduced the The effect of radotinib treatment on PrP Sc deposition in 22L and 263K scrapie-infected cerebellar slice culture models. (A,B) Sliced Tga20 cerebellar tissues were exposed to 1% 22L scrapie agent for 1 h. Culture medium including radotinib (25 µM) was changed three times a week, and the cultures were maintained for 3 (A) or 5 (B) weeks. The cultured tissues were lysed using RIPA buffer, and then, PrP levels after proteinase K treatment (PK, 3 µg/mL, 37 • C, 1 h incubation) or control were assessed by Western blotting with an anti-PrP antibody (3F10). (C) Hamster cerebellar tissue slices were exposed to 1% uninfected or 263K scrapie-infected hamster brain homogenates for 1 h and then treated with radotinib (0-40 µM) for 5 weeks. PrP levels after PK treatment (1 µg/60 µg of total proteins, 37 • C, 1 h incubation) were assessed by Western blotting with an anti-PrP antibody (3F10). Molecular masses in kDa are indicated on the left-hand side. β-actin was used as a loading control.
To determine whether radotinib exerts a practical effect during disease progression, we designed two experiments ( Figure 3A). Hamster cerebellar tissue slices that were infected with the 263K scrapie strain (1) were cultured for 5 weeks in the presence of radotinib ( Figure 3B, lanes 10-12) or (2) were cultured for 2 weeks in the absence of radotinib treatment followed by 3 weeks in culture medium containing radotinib, which was changed three times a week ( Figure 3B, lanes 3, 4, 7-9). Then, the cerebellar tissue slices were lysed and subjected to immunoblotting, and we found that the levels of PrP Sc were decreased after 5 weeks of treatment with radotinib in a dose-dependent manner ( Figure 3B). Interestingly, later treatment with 10-40 µM radotinib also dramatically reduced the levels of PrP Sc . These observations indicate that radotinib can markedly inhibit, reduce, or delay the deposition of PrP Sc .  To expand on the study about whether treatment with radotinib 1 or 2 weeks after infection inhibits PrP Sc propagation, 263K scrapie-infected hamster cerebellar tissue slices were cultured for 3 or 4 weeks, and then, radotinib was administered for the final weeks leading up to the 5-week endpoint. As shown in Figure 3C, treatment with radotinib in the final 2 weeks also effectively inhibited PrP Sc propagation, and the effects were similar to 5 weeks of treatment; however, one week of treatment with radotinib did not reduce PrP Sc propagation, similar to vehicle treatment of 263K-infected slices. Taken together, these results suggest that radotinib may exert anti-prion effects in vitro and ex vivo. To expand on the study about whether treatment with radotinib 1 or 2 weeks after infection inhibits PrP Sc propagation, 263K scrapie-infected hamster cerebellar tissue slices were cultured for 3 or 4 weeks, and then, radotinib was administered for the final weeks leading up to the 5-week endpoint. As shown in Figure 3C, treatment with radotinib in the final 2 weeks also effectively inhibited PrP Sc propagation, and the effects were similar to 5 weeks of treatment; however, one week of treatment with radotinib did not reduce PrP Sc propagation, similar to vehicle treatment of 263K-infected slices. Taken together, these results suggest that radotinib may exert anti-prion effects in vitro and ex vivo.

Radotinib Prolongs the Survival Times and Delays the Deposition of PrP Sc in 263K Scrapie-Infected Hamsters
Based on the in vitro and ex vivo results described above, we examined whether radotinib can prolong the survival time of hamsters that were intraperitoneally infected with the 263K scrapie strain. We administered radotinib (100 mg/kg in 0.5% carboxymethyl cellulose (CMC) sodium salt solution as a vehicle) once a day via the intragastric (IG) route for 6 days a week. As shown in Figure 4A, the survival time of the 263K scrapieinfected group was 135.1 ± 9.9 days postinoculation (dpi; range 118-152 dpi); however, the radotinib-treated group had a significantly prolonged survival time of 159.3 ± 28.6 dpi (range 124-223 dpi, p = 0.00217). Additionally, we injected radotinib via the intraperitoneal (IP) route 2 weeks after scrapie infection, which may exclude any possibility of interaction between the scrapie agent and drug, but there was no significant difference because only one hamster among six hamsters survived for 256 days (155 ± 50 dpi). This result suggests that the radotinib IG administration route is more advantageous for the treatment of prion diseases.
To determine whether IG administration of radotinib inhibits the propagation of PrP Sc in the hamster model, we evaluated the levels of PrP Sc in the brains of 263K scrapieinfected hamsters that were treated with radotinib and sacrificed at 146 dpi as a control ( Figure 4, lanes 1, 2), 263K scrapie-infected hamsters that were treated with vehicle and died naturally at 146 dpi ( Figure 4, lanes 3, 4), 263K scrapie-infected hamsters that were treated with radotinib and died naturally at 154 dpi ( Figure 4, lanes 5, 6) and healthy 263K scrapie-infected hamsters that were treated with radotinib and sacrificed at 146 dpi ( Figure 4, lanes 7, 8). PK-resistant PrP Sc was deposited in high levels in the brains of 263K scrapie-infected hamsters that were treated with the vehicle (0.5% CMC) ( Figure 4B, lane 4). Similarly, the PrP Sc levels in the brains of the 263K scrapie-infected hamsters that were treated with radotinib and died at 156 dpi were also high ( Figure 4B, lane 6). However, the PrP Sc levels in the brains of the 263K scrapie-infected hamsters that were treated with radotinib and sacrificed at 146 dpi were markedly decreased ( Figure 4B

Radotinib Prolongs the Survival Times and Delays the Deposition of PrP Sc in 263K Scrapie-Infected Hamsters
Based on the in vitro and ex vivo results described above, we examined whether radotinib can prolong the survival time of hamsters that were intraperitoneally infected with the 263K scrapie strain. We administered radotinib (100 mg/kg in 0.5% carboxymethyl cellulose (CMC) sodium salt solution as a vehicle) once a day via the intragastric (IG) route for 6 days a week. As shown in Figure 4A, the survival time of the 263K scrapieinfected group was 135.1 ± 9.9 days postinoculation (dpi; range 118-152 dpi); however, the radotinib-treated group had a significantly prolonged survival time of 159.3 ± 28.6 dpi (range 124-223 dpi, p = 0.00217). Additionally, we injected radotinib via the intraperitoneal (IP) route 2 weeks after scrapie infection, which may exclude any possibility of interaction between the scrapie agent and drug, but there was no significant difference because only one hamster among six hamsters survived for 256 days (155 ± 50 dpi). This result suggests that the radotinib IG administration route is more advantageous for the treatment of prion diseases.  Hamsters infected with the 263K scrapie strain received vehicle (n = 15) or radotinib (100 mg/kg/day for 6 days a week after 263K IP injection, n = 12) or intraperitoneally (10 mg/kg/day for 6 days a week 2 days postinoculation, n = 6). (A) The survival curve is presented as a Kaplan-Meier plot, and the log rank test was used for statistical analysis. ** p = 0.00217. NS, not significant. Vehicle, 0.5% CMC. (B) Brain homogenates from control hamsters that were sacrificed at 146 dpi (lanes 1 and 2), 263K scrapie-infected hamsters that were treated with vehicle and died naturally at 146 dpi (lanes 3 and 4), 263K scrapie-infected hamsters that were treated with radotinib and died naturally at 154 dpi (lanes 5 and 6), and 263K scrapie-infected hamsters that were treated with radotinib and sacrificed at 146 dpi (lanes 6 and 8) were incubated with or without PK (5 µg/mL) and then subjected to Western blotting. PrP levels were measured by Western blotting using an anti-PrP antibody (3F4). β-actin was used as a loading control. Molecular masses in kDa are indicated on the left-hand side. (C) Hamsters that were infected with the 263K scrapie strain, treated with vehicle, died naturally at 146 dpi (upper panels) as well as clinically healthy 263K scrapie-infected hamsters that were treated with radotinib and sacrificed at 146 dpi (100 mg/kg; bottom panels) were used for immunohistochemical staining. The anti-PrP antibody (3F4) was used. Scale bar = 200 µm.
Although the PrP Sc levels were decreased, the total expression of PrP in the 263K scrapie-infected hamsters that were treated with radotinib ( Figure 4B, lane 7) was similar compared to the control. Radotinib may inhibit or delay the deposition of PrP Sc even though there were differences among 263K scrapie-infected hamsters that were treated with radotinib. Next, we performed immunohistochemical staining for PrP Sc using 263K scrapieinfected hamsters that were treated with radotinib and sacrificed on 146 dpi. Compared to the brains of control-treated hamsters that were infected with the 263K scrapie strain, the total PrP and PK-resistant PrP Sc levels were definitely lower in the brains of infected hamsters that were treated with radotinib ( Figure 4C).

Wide Concentration Range of Radotinib Increases the Survival Time of 263K Scrapie-Infected Hamsters
Above, we showed the effect of radotinib when administered at different doses and for different time intervals in inhibiting PrP Sc propagation in in vitro and ex vivo models. Therefore, we examined how various concentrations and times of radotinib treatment affected survival time in vivo. One hour after the IP infection with the 263K scrapie strain, hamsters were treated IG with 30, 60, or 200 mg/kg radotinib, and then, survival curves were generated; the results of each group were compared with those of the 100 mg/kg group ( Figure 5A). The mean survival time of the 30 mg/kg radotinib group was 145.1 ± 5.3 dpi (range 140-154 dpi, p = 0.05848), that of the 60 mg/kg radotinib group was 161.0 ± 34.8 dpi (range 112-208 dpi, p = 0.03022), and that of the 200 mg/kg radotinib group was 157.0 ± 33.9 dpi (range 112-210 dpi, p = 0.01944). This result suggests that certain concentrations of radotinib are efficacious for prolonging the survival time of animals with prion diseases. 157.0 ± 33.9 dpi (range 112-210 dpi, p = 0.01944). This result suggests that certain concentrations of radotinib are efficacious for prolonging the survival time of animals with prion diseases. , or for the initial 10 weeks (radotinib 10W-initial, n = 10). The survival curve is presented as a Kaplan-Meier plot, and the log rank test was used for statistical analysis (bottom panel). * p < 0.05, ** p < 0.01, NS, not significant.
Additionally, we found that the initial 10 weeks of treatment with radotinib after 263K scrapie infection also significantly prolonged the survival time (149.6 ± 25.3 dpi, range 112-191 dpi, p = 0.0446). These results indicate that early treatment with radotinib is effective for the treatment of prion diseases.

Radotinib Crosses the Blood-Brain Barrier (BBB) of Hamsters
Previously, radotinib was shown to cross the BBB in a Parkinson's disease mouse model [13]. However, it is unknown whether radotinib penetrates the BBB in hamster models. Firstly, we the efficacy of two c-Abl tyrosine kinase inhibitors, imatinib and nilotinib hydrochloride monohydrate (hereafter referred to as nilotinib), in inhibiting (or retarding) the deposition levels of PrP Sc in hamster cerebellar tissue slice culture models. We confirmed that nilotinib reduced PrP Sc levels, but imatinib had no effect (Supplementary Figure S3). Next, we investigated whether radotinib crosses the BBB of hamsters and compared its ability to do so with nilotinib. Nilotinib is a second-generation Abl inhibitor that was approved for the treatment of chronic myelogenous leukemia by the U.S. Food and Drug Administration (FDA) in 2007 [21]. Because the structures of radotinib and nilotinib are similar [22], nilotinib is a suitable agent for comparison in evaluating the BBB permeability of radotinib. We confirmed that radotinib (AUC last ) crossed the BBB, and it was present in brain tissues at approximately 1.4 times higher levels than nilotinib after administration at 60 and 100 mg/kg ( Table 1). The time to peak drug concentration (T max ) of radotinib was slower than that of nilotinib, but the maximum concentration (C max ) was similar. Notably, B/P ratio (%), AUC last in brains treated with 60 mg/Kg and 100 mg/Kg radotinib was 13.7 ± 2.1 and 23.2 ± 15.5, respectively, which is a high absorption rate and is 2-3 times higher than nilotinib treatment. This result suggests that radotinib also effectively crosses the BBB in hamsters.   The values are shown as the mean ± SD from the plasma (P) and brain (B) of three golden hamsters. ‡ A single value was obtained from one hamster since the parameter was not assessed in two other hamsters. NC, not calculated. *** p < 0.001 compared to the B/P ratio (%) and AUC last of the groups treated with both concentrations of nilotinib. ** p < 0.01 compared to the B/P ratio (%) and C max of the groups treated with both concentrations of nilotinib. † † p < 0.01 compared to the B/P ratio (%) and C max of the 60 mg/kg nilotinib group. † † † p < 0.001 compared to the B/P ratio (%) and C max of the 100 mg/kg nilotinib group. By using the program R (version 4.0.2), both the B/P ratio (%) and AUC last, as well as the B/P ratio (%) and C max, were over 95% in the Shapiro test (normality test), but only the B/P ratio (%) and C max were over 95% in the Bartlett test (homogeneity of variance). Thus, after the Box-Cox transformation of the B/P ratio (%) and AUC last values, both the Shapiro test and the Bartlett test were performed again (p > 0.05 in both the normality and the homogeneity of variance tests). Using the B/P ratio (%), C max values, and the transformed B/P ratio (%) and AUC last values, statistical analysis was carried out by one-way ANOVA. The significance was determined using the post hoc test (Tukey HSD).

Discussion
Our study provides evidence of the potential anti-prion efficacy of radotinib in scrapieinfected neuronal cell models, brain tissue slice culture models, and, most interestingly, scrapie-infected hamster models. In these models, radotinib reduced PrP Sc propagation, which may have resulted in prolonged survival times of prion-affected hamster models. To date, it has been a challenge to find or develop effective anti-prion drugs as well as drugs for the treatment of other neurodegenerative diseases, such as Alzheimer's disease and Parkinson's disease [23][24][25][26]. In particular, the duration of illness is typically shortened by approximately 18 months after signs and symptoms appear. Among the types of prion diseases, familial CJD accounts for 10-15% of cases [3,4]. In addition to complete disease treatment, prolonged clinical remission and delayed symptom onset need to be addressed. On the other hand, many compounds that are effective in preclinical tests are unsuitable for use in clinical trials due to low efficacy or unexpected toxicity. In addition, the development of new drugs for targeted disease treatment requires considerable time and costs [27]. Therefore, drug repositioning or repurposing methods are a very helpful strategy for investigating effective anti-prion drugs. In this regard, radotinib could be a drug candidate that could be quickly used in the treatment of prion diseases.
One of the major challenges to drug effectiveness is the blood-brain barrier (BBB) since it does not allow many drugs (either chemical compounds or proteins) to cross into the brain from the peripheral circulation. Abnormalities in PrP in the central nervous system are definite causes of prion diseases because PrP-deficient animal models do not progress to prion diseases [28]. Notably, radotinib effectively enters the mouse brain [13]. Here, we also confirmed the effective BBB penetration of radotinib and the absorption rate into the brain is better than nilotinib in hamsters (Table 1); therefore, radotinib may be active in directly inhibiting the conversion of PrP C to PrP Sc or the clearance of PrP Sc in the brains of animal models (Figure 4) as well as in vitro and ex vivo models (Figures 1-3). However, reductions in PrP Sc levels are not always necessary for prolonged survival time. For example, the eIF2a-P inhibitors trazodone hydrochloride and dibenzoylmethane prevent neurodegeneration and significantly prolong survival without reducing the PrP Sc levels in RML-infected hemizygous Tg37 +/− mice [9].
We showed that radotinib is effective in delaying neurodegeneration and prolonging survival time, although its effects vary (Figures 4 and 5). Several radotinib-treated samples exhibited clinical health, the markedly low spread of pathological PrP Sc in brain tissues, and delayed endpoints of disease, even in the context of 263K scrapie infection. Most importantly, late treatment with radotinib was also effective in ameliorating disease ( Figure 5B). The survival times observed after 4 weeks of late treatment (158.0 ± 31.9 dpi after prion infection) were approximately 10 days longer than those observed after 8 weeks of late treatment with radotinib (146.9 ± 23.7 dpi). Additionally, an initial 10 weeks of treatment with radotinib resulted in prolonged survival times (149.6 ± 25.3 dpi), but the mean survival time was similar to that observed after 8 weeks of late treatment. Similarly, we also found that 2 weeks of late treatment with radotinib reduced (or inhibited) PrP Sc deposition, but similar results were not observed after treatment with radotinib during only the final week of the cerebellar tissue slice culture models ( Figure 3B). These results indicate that the early and continuous administration of radotinib is more effective in delaying disease progression. Possible or definite CJD patients are diagnosed at late time points after the onset of symptoms by biopsy, 14-3-3 protein and tau detection, and real-time quakinginduced conversion (RT-QUIC) [5]. In other words, in order to be a valuable therapeutic option, late drug treatment must have an effect on targeting the ongoing disease. Radotinib may meet this criterion.
Radotinib is a novel and second-generation c-Abl tyrosine kinase inhibitor whose chemical structure resembles that of imatinib mesylate and nilotinib, and it is approved for the chronic-phase treatment of chronic myeloid leukemia [12,22]. Interestingly, we found that radotinib and nilotinib reduced PrP Sc accumulation, but imatinib had no effect in hamster cerebellar tissue slice culture models (Supplementary Figure S3). Previous studies have suggested that imatinib showed clearance of PrP Sc in scrapie-infected Neuro2A cells and mice [7,14]. Since we only tested these compounds in hamster cerebellar tissue slice culture models, further studies are needed to investigate the anti-prion effects of various c-Abl tyrosine kinase inhibitors in different prion animal models.
Recently, c-Abl tyrosine kinase targeting has been considered a therapeutic approach for neurodegenerative diseases, including Alzheimer's disease and Parkinson's disease [29]. The synthetic neurotoxic prion fragment PrP106-126 activates c-Abl tyrosine kinase, and blocking or knocking down c-Abl protects neuronal cells from PrP106-126-induced mitochondrial dysfunction, reactive oxygen species production, and apoptosis [30]. On the other hand, reactive oxygen species (ROS) induce the localization of c-Abl tyrosine kinase to mitochondria, thereby mediating mitochondrial dysfunction, including the loss of mitochondrial transmembrane potential, depletion of ATP, and necrosis-like cell death [31]. Dynamin-related protein 1 (Drp1) is a critical factor in regulating mitochondrial dynamics. c-Abl tyrosine kinase phosphorylated Drp1, which enhances the GTPase activity of Drp1, and promotes Drp1-mediated mitochondrial fragmentation and cell death [32]. Recently, scrapie infection induced neuronal cell death accompanied by mitochondrial dysfunction, including the induction of mitochondrial ROS, the loss of mitochondrial membrane potential, and reduced ATP production [33]. In scrapie-infected mice, an imbalance in mitochondrial fusion and fission proteins such as Mfn1, Fis1, Mfn2, and Dlp1 was observed [34]. Additionally, the accumulation of endothelial nitric oxide synthase in mitochondria and downregulation of mitochondrial superoxide dismutase, cytochrome c, and ATP activity were observed in scrapie-infected mice [35]. Particularly, radotinib protects against α-synuclein preformed fibrils (PFF)-induced neuronal cell death and restores α-synuclein PFF-induced activation of Iba-1-positive microglia and GFAP-positive astrocytes in Parkinson's disease models. Therefore, radotinib has the potential to act as a neuroinflammation suppressor [13]. Another tyrosine kinase inhibitor, STI571 (Gleevec or imatinib mesylate), causes the cellular clearance of PrP Sc in prion-infected cells through STI571, which strongly activates the lysosomal degradation of PrP Sc [14]. STI571 also protects neuronal cells from neurotoxic PrP fragment-induced apoptosis [15]. Based on these previous findings, we believe that radotinib may prevent dysregulated mitochondrial homeostasis, consequently leading to a delay in neuronal cell death caused by PrP Sc toxicity while also ameliorating glial cell-mediated neuroinflammation. With a reduced accumulation of PrP Sc (although the exact mechanism remains unknown), radotinib may prolong the survival time of prion-affected animal models.
In this study, we demonstrated for the first time that radotinib possesses anti-prion activity in in vitro, ex vivo, and in vivo prion models. Interestingly, late treatment with radotinib did not reduce its ability to decrease PrP Sc deposition or prolong survival time. Therefore, we suggest that radotinib is a valuable drug candidate that could be rapidly used for the treatment of prion diseases. Moreover, our findings, along with the observations of other researchers, suggest strong associations between c-Abl tyrosine kinase (or c-Abl tyrosine kinase inhibitor) and pathological prion proteins. Further study of c-Abl tyrosine kinase inhibitors will advance the treatment of prion diseases.

Prion Organotypic Slice Culture Assay
Cerebellums were harvested from postnatal 12-day-old Tga20 mice [20] or postnatal 12-day-old golden hamsters, and a prion organotypic slice culture system was established as previously described [16,17]. The cerebellar tissues were sectioned to thicknesses of 300 µm using a Vibratome 3000 sectioning system (Technical Products International, St. Louis, MO, USA), and the sections were maintained in 6-well plates with membrane inserts (Millicell-CM inserts, 0.4 µm, 30 mm, Merck-Millipore, Billerica, MA, USA) in culture medium (100 mL 2 × minimal essential medium powder, 100 mL basal medium Eagle without glutamine, 100 mL horse serum, 4 mL GlutaMAX-I, 4 mL penicillin/streptomycin, 5.5 mL D-(+)-glucose solution and 86.5 mL ddH 2 O, pH 7.2-7.4) in a 5% CO 2 incubator for 3 or 5 weeks. The medium was replaced 3 times per week. For prion inoculation, the cerebellar tissue slices were incubated with 1% 22L or 1% 263K scrapie agent in 6-well plates at 4 • C for 1 h and then placed on membrane inserts.

Western Blotting
Cells, tissue slices, and brain tissues were lysed using RIPA buffer. The proteins were separated with SDS-PAGE gels and transferred to nitrocellulose membranes. For the detection of target proteins, mouse monoclonal anti-PrP (3F4), mouse monoclonal anti-PrP (3F10) [19], and mouse monoclonal anti-β-actin (Sigma-Aldrich, Burlington, MA, USA) antibodies were used.

Immunohistochemistry
Neutral buffered formalin-fixed, paraffin-embedded brain tissue slices (5 µm thick) were used for immunohistochemical staining. The tissue slides were blocked with horse serum, incubated with an anti-PrP antibody (3F4), incubated with a biotinylated horse anti-mouse IgG secondary antibody using an ultrasensitive ABC peroxidase mouse IgG staining kit, and visualized with a 3,3 -diaminobenzidine (DAB) substrate kit (Thermo Fisher Scientific). After the sections were counterstained with hematoxylin, they were observed under a light microscope (BX51; Olympus, Southend-on-Sea, UK).

Animal Experiments
The animal experiments were approved by the Institutional Animal Care and Use Committees of Hallym University Medical Center (HMC 2019-0-0305-5-24). Golden hamsters (6 weeks of age, Japan SLC, Inc., Shizuoka, Japan) were intraperitoneally injected with 100 µL 1% brain homogenate from 263K scrapie-infected hamsters. Radotinib was dissolved in 0.5% CMC sodium salt solution, and intragastrically administered to the hamsters at doses of 30 to 200 mg/kg once daily for one week at the indicated times. After the terminal stage of scrapie agent incubation, when clinical manifestations of the disease were evident, the hamsters were sacrificed with CO 2 gas, and brain tissues were harvested.

Pharmacokinetic Assay
Radotinib and nilotinib hydrochloride monohydrate were suspended in 0.5% CMC at a concentration of 60 mg/kg/10 mL or 100 mg/kg/10 mL, and these reagents were administered to 6-week-old male golden hamsters in a single dose by oral gavage. At 1, 2, 3, 4, 6, 8, and 24 h after administration, blood samples were collected by venipuncture of the abdominal vena cava under anesthesia with Zoletil 50 (Virbac)/Rompun (Bayer). All the blood samples were transferred to heparin tubes, and plasma was obtained by centrifugation at 3000 rpm for 10 min at 15 • C. After collecting the blood samples, the hamsters were transcardially perfused with PBS, and whole brains were isolated. The whole brains were weighed and homogenized in PBS (1:3 dilution). Then, 2-fold acetonitrile was added to the homogenized brain samples, and the mixtures were centrifuged at 14,000 rpm for 5 min at 4 • C. The supernatants were used to quantify the concentrations of radotinib and nilotinib. The concentrations of radotinib and nilotinib in the plasma and brain tissue samples were measured using LC-MS/MS. Pharmacokinetic parameters, including area under the curve from the time of dosing to the last measurable positive concentration (AUC last ), maximum concentration (C max ), time to reach C max (T max ), and terminal elimination half-life (T1/2), were calculated with the BA Calc 2007 analysis program (Ministry of Food and Drug Safety (MFDS), Republic of Korea).

Statistical Analysis
The probability of statistically significant differences between experimental groups is presented in each figure legend or table legend using OriginPro 2023 software (OriginLab Corporation, Northampton, MA, USA) or the program R (version 4.0.2).