Adaptative Up-Regulation of PRX2 and PRX5 Expression Characterizes Brain from a Mouse Model of Chorea-Acanthocytosis

The peroxiredoxins (PRXs) constitute a ubiquitous antioxidant. Growing evidence in neurodegenerative disorders such as Parkinson’s disease (PD) or Alzheimer’s disease (AD) has highlighted a crucial role for PRXs against neuro-oxidation. Chorea-acanthocytosis/Vps13A disease (ChAc) is a devastating, life-shortening disorder characterized by acanthocytosis, neurodegeneration and abnormal proteostasis. We recently developed a Vps13a−/− ChAc-mouse model, showing acanthocytosis, neurodegeneration and neuroinflammation which could be restored by LYN inactivation. Here, we show in our Vps13a−/− mice protein oxidation, NRF2 activation and upregulation of downstream cytoprotective systems NQO1, SRXN1 and TRXR in basal ganglia. This was associated with upregulation of PRX2/5 expression compared to wild-type mice. PRX2 expression was age-dependent in both mouse strains, whereas only Vps13a−/− PRX5 expression was increased independent of age. LYN deficiency or nilotinib-mediated LYN inhibition improved autophagy in Vps13a−/− mice. In Vps13a−/−; Lyn−/− basal ganglia, absence of LYN resulted in reduced NRF2 activation and down-regulated expression of PRX2/5, SRXN1 and TRXR. Nilotinib treatment of Vps13a−/− mice reduced basal ganglia oxidation, and plasma PRX5 levels, suggesting plasma PRX5 as a possible ChAc biomarker. Our data support initiation of therapeutic Lyn inhibition as promptly as possible after ChAc diagnosis to minimize development of irreversible neuronal damage during otherwise inevitable ChAc progression.


Introduction
The peroxiredoxins (PRXs) constitute a ubiquitous antioxidant system encompassing typical two-cysteine (Cys) Prxs (1-4), a single atypical two-Cys Prx (PRX5) and a single one-Cys Prx (PRX6) [1][2][3][4][5][6][7]. PRX2 and PRX5 use two Cys residues to accomplish the task of detoxifying a vast range of organic peroxides, H 2 O 2 and peroxynitrite. Hyperoxidation whereas PRX5 expression in wild-type and Vps13a −/− mouse red cells was similar. Analyses of Vps13a −/− mouse basal ganglia revealed increased protein oxidation and activation of the redox-related transcription factor, NRF2, associated with up-regulation of PRX2 and PRX5 expression and their reducing systems: thioredoxin-reductase (TRXR) and sulfiredoxin-1 (SRXN1). Basal ganglia from Vps13a −/− mice exhibited increased PRXSO3 levels, suggesting a relation between increased local oxidative stress and impaired autophagy [36]. We previously documented improvement of autophagy in double-knockout mice lacking both VPS13a and LYN [36]. In basal ganglia from Vps13a −/− ; Lyn −/− mice, the absence of LYN results in reduction of NRF2 activation and down-regulated expression of PRX2/5, SRXN1 and TRXR as compared to Vps13a −/− mouse basal ganglia. We also observed reduction in PRXSO3, supporting the observation that improvement of autophagy is linked to reduction of cellular oxidative stress. As proof of concept, we administrated Nilotinb to Vps13a −/− mice. In basal ganglia from nilotinib-treated Vps13a −/− mice we found reduced NRF2 activation and down-regulated expression of PRX2/5, TRXR and SRXN1, paralleled by a decrease in PRXSO3.
Our data collectively show for the first time that in Vps13a −/− mouse basal ganglia, PRX2/5 might represent neuroprotective systems against oxidation and neuroinflammation linked to impaired autophagy. Plasma PRX5 more than plasma PRX2 might reflect neurodegeneration related to the absence of VPS13a expression. Nilotinib improved autophagy, indirectly attenuated oxidation and reduced PRX2/PRX5 expression, supporting re-purposing of Lyn kinase inhibitors as a possible treatment for patients with ChAc. In addition, plasma PRX5 might represent a disease marker useful in follow-up of patients since plasma PRX5 is increased in Vps13a −/− mice and reduced by nilotinib treatment.

Mouse Strains and Design of the Study
12-and 18-month-old C57B6/2J wild-type (WT) mice, Vps13a −/− mice and Vps13a −/− Lyn −/− mice were studied [36]. Whenever indicated, WT and Vps13a −/− mice were treated daily with vehicle (tap water) or nilotinib (25 mg/Kg/day). Nilotinib was administered by once daily gavage to WT or Vps13a −/− mice of ages 11 or 17 months for periods of either 3 or 6 months [36]. Isoflurane-anesthetized mice were euthanized and randomly assigned to experimental groups. Brains were acutely dissected to isolate basal ganglia (BG, consisting of corpus striatum), which was rapidly frozen in liquid nitrogen. Red cells and plasma collected from both Vps13a −/− mice and from ChAc patients were analyzed for levels of PRX2 and PRX5 by Western-blot analysis [37,[39][40][41].

Patient Characteristics
Plasma samples from ChAc patients described in two previous studies were analyzed [35,36,38]. In brief, diagnosis was based on clinical manifestations, the absence of chorein in Western blot analysis, and genetic testing. All patients gave their informed consent for sample preparation and publication of the data. Patients and healthy control blood donors were enrolled in ongoing studies on the pathogenesis and natural history of neurodegenerative diseases approved by the institutional review board of the Technische Universität Dresden, Germany (EK 45022009, EK 78022015) and Comitato Etico Verona e Rovigo (FGFITA3).

Statistical Analysis
Data were analyzed using either t-test or one-way ANOVA for multiple comparisons. A difference with a p < 0.05 was considered significant.

Plasma PRX2 and PRX5 Are Increased in Both Vps13a −/− Mice and Patients with ChAc
We previously reported that 12-and 18-month-old Vps13a −/− mice recapitulate the clinical and biologic phenotypes of ChAc [36]. As shown in Figure 1a, plasma levels of PRX2 and PRX5 were significantly higher in Vps13a −/− mice aged 12 and 18 months than in corresponding wild-type animals. PRX5 increased with age in Vps13a −/− mice but not in wild-type mice (Figures 1a and S1a). Plasma PRX2 levels were significantly higher in 18-month-old wild-type mice than in younger healthy animals (Figures 1a and S1a). Levels of PRX2 and PRX5 in plasma from ChAc patients was significantly higher than in plasma of healthy subjects (Figures 1b and S1b). Since the normal range of circulating PRXs remains undefined, we have assumed that plasma PRX concentrations in wild-type animals and in human healthy controls are not pathologic [44]. antibodies reported in Supplementary Methods. Oxidized proteins were monitored using OxyBlot Protein Oxidation Detection Kit (EMD Millipore) as previously reported [42,43]. In brief, carbonylated proteins were detected by reacting with 2,4-dinitrophenylhydrazine (DNPH) and blotted with anti-dinitrophenyl antibody [42,43].

Statistical Analysis
Data were analyzed using either t-test or one-way ANOVA for multiple comparisons. A difference with a p < 0.05 was considered significant.

Plasma PRX2 and PRX5 Are Increased in Both Vps13a −/− Mice and Patients with ChAc
We previously reported that 12-and 18-month-old Vps13a −/− mice recapitulate the clinical and biologic phenotypes of ChAc [36]. As shown in Figure 1a, plasma levels of PRX2 and PRX5 were significantly higher in Vps13a −/− mice aged 12 and 18 months than in corresponding wild-type animals. PRX5 increased with age in Vps13a −/− mice but not in wild-type mice (Figures 1a and S1a). Plasma PRX2 levels were significantly higher in 18month-old wild-type mice than in younger healthy animals (Figures 1a and S1a). Levels of PRX2 and PRX5 in plasma from ChAc patients was significantly higher than in plasma of healthy subjects (Figures 1b and S1b). Since the normal range of circulating PRXs remains undefined, we have assumed that plasma PRX concentrations in wild-type animals and in human healthy controls are not pathologic [44].  Figure S1a (b) Western-blot (Wb) analysis of PRX2 and PRX5 in plasma samples from healthy control (Ctrl), ChAc patient 1 (P1) and ChAc patient 2 (P2), corresponding to ChAc patients reported in De Franceschi L et al. [36]. Densitometric analysis is shown in lower panel. Data are presented as means ± SEM, n = 4 * p < 0.05 compared to healthy controls. Loading controls are shown in Figure S1b Figure S1a (b) Western-blot (Wb) analysis of PRX2 and PRX5 in plasma samples from healthy control (Ctrl), ChAc patient 1 (P1) and ChAc patient 2 (P2), corresponding to ChAc patients reported in De Franceschi L et al. [36]. Densitometric analysis is shown in lower panel. Data are presented as means ± SEM, n = 4 * p < 0.05 compared to healthy controls. Loading controls are shown in Figure S1b As PRX2 and PRX5 are expressed in red cells [8,10,45], we analyzed PRX2 and PRX5 in cytosol fractions from erythrocytes of wild-type and erythrocytes (including circulating acanthocytes, Figure 1c) of Vps13a −/− mice ( Figure 1d, upper panel). PRX2 expression in Vps13a −/− mouse erythrocytes was significantly lower than in wild-type red cells, whereas PRX5 expression was similar in both mouse strains (Figure 1d, upper panel; Figure S1c). In agreement with our previous report in human ChAc red cells [38], the amount of PRX2 translocated to the membrane (ghost fraction) in Vps13a −/− red cells was significantly lower than in red cell membrane fractions from wild-type animals ( Figure 1d, lower panel; Figure S1c). This difference is related to the perturbation of multiprotein complexes involving band 3, the docking site for PRX2 [8].
Since PRX2 and PRX5 might be released into plasma by additional cell types besides red cells, the increased plasma levels of PRX2 and PRX5 observed in both Vps13a −/− mice and human ChAc patients also reflect the neurodegeneration and neuroinflammation observed in ChAc.

Activation of NRF2 and Up-Regulation of PRX2/5 Characterizes Basal Ganglia from
is a key redox-related transcription factor, linked to PRXs expression and function [11]. Studies of neurodegenerative disorders such as PD or AD have shown oxidation to be an early neurotoxicity marker of abnormal proteostasis [46][47][48][49][50][51][52][53]. We observed agedependent increases in protein oxidation in basal ganglia isolated from Vps13a −/− and wild-type mice of ages 12 and 18 months, as determined by OxyBlot (Figures 2a and S2). This increase was greater in basal ganglia from Vps13 −/− mice than in wild-type basal ganglia. We also found age-dependent NRF2 activation in basal ganglia isolated from Vps13a −/− and wild-type mice of ages 12 and 18 months (Figure 2b), greater in the former than the latter (Figure 2b).
12 months of age. Catalase was used as protein loading control. The blot shown is repre four with similar results. Densitometric analyses are presented in Supplementary Figur As PRX2 and PRX5 are expressed in red cells [8,10,45], we analyzed PRX2 in cytosol fractions from erythrocytes of wild-type and erythrocytes (including acanthocytes, Figure 1c) of Vps13a −/− mice ( Figure 1d, upper panel). PRX2 exp Vps13a −/− mouse erythrocytes was significantly lower than in wild-type red cel PRX5 expression was similar in both mouse strains (Figure 1d, upper panel; F In agreement with our previous report in human ChAc red cells [38], the amou translocated to the membrane (ghost fraction) in Vps13a −/− red cells was significa than in red cell membrane fractions from wild-type animals ( Figure 1d, lower ure S1c). This difference is related to the perturbation of multiprotein complexe band 3, the docking site for PRX2 [8].
Since PRX2 and PRX5 might be released into plasma by additional cell typ red cells, the increased plasma levels of PRX2 and PRX5 observed in both Vps and human ChAc patients also reflect the neurodegeneration and neuroinflam served in ChAc.

Activation of NRF2 and Up-Regulation of PRX2/5 Characterizes Basal
is a key redox-related transcription factor, linked to PRXs expression tion [11]. Studies of neurodegenerative disorders such as PD or AD have shown to be an early neurotoxicity marker of abnormal proteostasis [46][47][48][49][50][51][52][53]. We obs dependent increases in protein oxidation in basal ganglia isolated from Vps13a − type mice of ages 12 and 18 months, as determined by OxyBlot (Figures 2a an increase was greater in basal ganglia from Vps13 −/− mice than in wild-type bas We also found age-dependent NRF2 activation in basal ganglia isolated from Vp wild-type mice of ages 12 and 18 months (Figure 2b), greater in the former tha ( Figure 2b).  Expression of PRX2 and PRX5 was significantly higher in basal ganglia from Vps13a −/− mice aged 12 and 18 months than in basal ganglia from corresponding wild-type animals ( Figure 3a). This observation was associated with PRX-SO3 accumulation in basal ganglia from 12-month-old Vps13a −/− mice as compared to age-matched wild-type animals (Figure 3a). No major difference in PRX-SO3 was evident between basal ganglia from Vps13a −/− and wild type mice aged 18 months (Figure 3a). Similarly, we found up-regulation of SRXN1 expression in basal ganglia from Vps13a −/− mice aged 12 and 18 months as compared to corresponding wild-type animals (Figure 3b). TRXR expression was significantly higher in basal ganglia from 12-month-old Vps13a −/− mice than in wildtype animals of the same age. TRXR expression in basal ganglia did not significantly differ between 18-month-old mice of the two strains ( Figure 3b). ganglia of wild-type (WT) and Vps13a −/− mice aged 12 and 18 months. GADPH was control. Representative of 4 blots with similar results. Densitometric analysis is panel. Data are presented as means±SEM, n = 4 * p < 0.05 compared to WT animal pared to 12-month-old WT mice.
Expression of PRX2 and PRX5 was significantly higher in basal ganglia mice aged 12 and 18 months than in basal ganglia from corresponding wil (Figure 3a). This observation was associated with PRX-SO3 accumulation i from 12-month-old Vps13a −/− mice as compared to age-matched wild-type a 3a). No major difference in PRX-SO3 was evident between basal ganglia and wild type mice aged 18 months (Figure 3a). Similarly, we found up SRXN1 expression in basal ganglia from Vps13a −/− mice aged 12 and 18 m pared to corresponding wild-type animals (Figure 3b). TRXR expression w higher in basal ganglia from 12-month-old Vps13a −/− mice than in wild-type same age. TRXR expression in basal ganglia did not significantly differ betw old mice of the two strains (Figure 3b). Taken together our data indicate that oxidative stress is higher in basa Vps13a −/− mice than in wild-type animals [35,36]. Chronic oxidative stress i from Vps13a −/− mice is associated with activation of NRF2 and up-regulatio and PRX5 cytoprotective systems. Although PRX2 expression is modulat both mouse strains, the increased age-independent expression of PRX5 Vps13a −/− mice appears related to the neurologic phenotype. Taken together our data indicate that oxidative stress is higher in basal ganglia from Vps13a −/− mice than in wild-type animals [35,36]. Chronic oxidative stress in basal ganglia from Vps13a −/− mice is associated with activation of NRF2 and up-regulation of both PRX2 and PRX5 cytoprotective systems. Although PRX2 expression is modulated by aging in both mouse strains, the increased age-independent expression of PRX5 and SRXN1 in Vps13a −/− mice appears related to the neurologic phenotype.

Down-Regulates PRX2/5 Expression in Vps13a −/− Basal Ganglia
We previously reported that impaired autophagy in basal ganglia of Vps13a −/− mice results in accumulation of neurotoxic proteins and of active Lyn [35,36]. The genetic absence of Lyn (in Vps13a −/− ; Lyn −/− mice) or therapeutic inhibition of active Lyn by nilotinib improves autophagy and beneficially impacts neuroinflammation in Vps13a −/− mice. We therefore evaluated NRF2 activity and PRX2/5 expression in Vps13a −/− ; Lyn −/− mice. As shown in Figure 4a, we found that basal ganglia NRF2 activation was lower in Vps13a −/− ; Lyn −/− mice than in Vps13a −/− animals. In agreement, we observed down-regulation of NQO1, a NRF2-regulated antioxidant system [11]. Expression of PRX2 and PRX5 in basal ganglia of Vps13a −/− ; Lyn −/− mice was significantly lower than in Vps13a −/− animals ( Figures  4b and S3). Reduced accumulation of overoxidized PRXs (PRX-SO3) was also detected in basal ganglia of Vps13a −/− ; Lyn −/− mice (Figures 4b and S3). This was associated with lower basal ganglia expression of both SRXN1 and TRXR in Vps13a −/− ; Lyn −/− than in Vps13a −/− animals (Figure 4c). These data indicate that improvement of autophagy by genetic inactivation of LYN prevents activation of the NRF2 pathway, resulting in down-regulation of antioxidant systems such as NQO1, PRX2 and PRX5. These data indicate that improvement of autophagy by genetic inactivation of LYN prevents activation of the NRF2 pathway, resulting in down-regulation of antioxidant systems such as NQO1, PRX2 and PRX5.
As proof of concept, we treated Vps13a −/− mice with nilotinib, a specific LYN inhibitor that can permeate the BBB [36]. As shown in Figure 5a and Supplementary Figure S5a), nilotinib significantly reduced NRF2 activation in Vps13a −/− mice aged 12 and 18 months as compared to vehicle-treated animals. This result is consistent with down-regulation of the NRF2 dependent NQO1 antioxidant system and the reduction in protein oxidation observed in basal ganglia of nilotinib-treated Vps13a −/− mice aged 12 and 18 months as compared to vehicle-treated animals (Figure 5a, Supplementary Figure S4). compared to vehicle-treated animals (Figure 5a, Supplementary Figure S4).
Basal ganglia expression of PRX2 and PRX5 was significantly lower in nilotinibtreated than in vehicle-treated Vps13a −/− mice aged 12 and 18 months and was associated with reduced accumulation of overoxidized PRXs (PRX-SO3) (Figure 5b and Supplementary Figure S5b). Most notably, plasma PRX5 levels in mice treated with nilotinib were lower than in vehicle-treated animals (Figures 5c and S6).

Discussion
Our data indicate that oxidation is an early event in the pathogenesis of ChAc in Vps13a −/− mice, resembling other neurodegenerative disorders such as PD or AD [18,51,52,54]. Indeed, protein oxidation was detectable in basal ganglia from 12-month- Basal ganglia expression of PRX2 and PRX5 was significantly lower in nilotinib-treated than in vehicle-treated Vps13a −/− mice aged 12 and 18 months and was associated with reduced accumulation of overoxidized PRXs (PRX-SO3) (Figure 5b and Supplementary Figure S5b). Most notably, plasma PRX5 levels in mice treated with nilotinib were lower than in vehicle-treated animals (Figures 5c and S6).

Discussion
Our data indicate that oxidation is an early event in the pathogenesis of ChAc in Vps13a −/− mice, resembling other neurodegenerative disorders such as PD or AD [18,51,52,54]. Indeed, protein oxidation was detectable in basal ganglia from 12-month-old Vps13a −/− mice, whereas neuronal loss appeared not earlier than in 18 months old Vps13a −/− mice. These changes were absent from younger Vps13a −/− mice and from age-matched wild-type animals [36].
In Vps13a −/− mice prolonged oxidation is associated with activation of NRF2 (pNRF2) and up-regulation of PRX2, PRX5 and SRXN1. pNRF2 levels increased with age in both WT and mutant mouse strains but were higher in Vps13a −/− mice. This was paralleled by up-regulation of PRX2 and TRXR. In contrast, expression of PRX5 and SRXN1 was increased in Vps13a −/− basal ganglia in an age-independent manner. PRX5 might thus be a more reliable marker of neuro-oxidative damage than PRX2 in Vps13a −/− mice. A possible anti-inflammatory activity of PRX5 might also be suggested in Vps13a −/− mice, based on previous reports in models of AD and PD [18]. The up-regulation of SRXN1 further emphasizes the severity of oxidation in basal ganglia from Vps13a −/− mice ( Figure 6). SRXN1 restores PRXSO3 to the TRXR cycle, preventing overoxidation and inactivation of PRXs [28,53,55]. Indeed, both Srxn1 and TrxR have ARE-elements under NRF2 regulation, indicating their importance in resistance against neuro-oxidation ( Figure 6) [28]. Moreover, a neuroprotective role for SRXN1 has been previously suggested in in vitro models of neuronal cells exposed to oxidation [56,57].
Antioxidants 2022, 11, x FOR PEER REVIEW 10 of 14 Figure 6. Schematic diagram. Vps13a −/− mice are characterized by impaired autophagy and neuroinflammation, with increased production of reactive oxygen species (ROS). Therapeutic activation of redox-related transcriptional factor NRF2 binds to antioxidant-responsive elements (ARE) of specific genes, promoting transcription of antioxidant proteins such as NQO1, SRXN1 and TRXR, in order to counteract producction of ROS and other reactive species. TRXR is required to reduce and reactivate PRX2 and PRX5. In highly oxidized environments, Prxs can be overoxidized (PRX-SO3). SRXN1 can restore overoxidized Prxs to their functional reduced state. These antioxidant proteins together orchestrate preservation of a beneficial equilibrium between antioxidants and oxidants. In double knockout Vps13a −/− ; Lyn −/− mice and in Vps13a −/− animals treated with nilotinib, amelioration of authophagy beneficially affects oxidation, resulting in reduced ROS production.
In conclusion, our study highlights the novel role of oxidation as a contributor to the pathogenesis of ChAc in mice genetically lacking the Vps13 gene product, chorein. In Vps13a −/− animals, activation of Nrf2 and related up-regulation of antioxidant systems is inadequate in the context of the prolonged oxidation linked to the impaired autophagy that characterizes ChAc. Since pathological protein oxidation is related to accumulation of neurotoxic proteins, therapeutic strategies to improve protein clearance represents an interesting therapeutic approach to ChAc. PRX5 might thus act both in cytoprotective and anti-inflammatory manners. Our observation that the increased basal ganglia expression and plasma levels of PRX5 in Vps13a −/− mice are reduced by nilotinib treatment lends support to the use of plasma PRX5 levels as disease biomarker possibly transferable to human studies. Collectively, our data further support the initiation of therapeutic Lyn inhibition as soon as possible after diagnosis of ChAc, to produce the highest likelihood of preventing irreversible cellular damage related to the natural history of the disease.  . Schematic diagram. Vps13a −/− mice are characterized by impaired autophagy and neuroinflammation, with increased production of reactive oxygen species (ROS). Therapeutic activation of redox-related transcriptional factor NRF2 binds to antioxidant-responsive elements (ARE) of specific genes, promoting transcription of antioxidant proteins such as NQO1, SRXN1 and TRXR, in order to counteract producction of ROS and other reactive species. TRXR is required to reduce and reactivate PRX2 and PRX5. In highly oxidized environments, Prxs can be overoxidized (PRX-SO3). SRXN1 can restore overoxidized Prxs to their functional reduced state. These antioxidant proteins together orchestrate preservation of a beneficial equilibrium between antioxidants and oxidants. In double knockout Vps13a −/− ; Lyn −/− mice and in Vps13a −/− animals treated with nilotinib, amelioration of authophagy beneficially affects oxidation, resulting in reduced ROS production.
To investigate the possibility that potentiation of cytoprotective systems can counteract oxidation in neurodegenerative disorders, different therapeutic strategies to restore redox balance have been tested in both cell and animal models of AD, PD and multiple sclerosis (MS) [51,52]. These strategies are based largely on pure antioxidants (such as N-acetylcysteine and ω3-fatty acid supplementation), or antioxidant/potentiators of NRF2 (such as quercitin and myricytin), or NRF2 antagonists (e.g., sulphoraphane and dimethylfumarate) [51,52,58]. Although beneficial antioxidant effects of these molecules have been reported in models for neurodegenerative disorders, the time of treatment initiation during the course of disease progression seems to be an important determinant of treatment impact.
We approached this dilemma by targeting autophagy to prevent accumulation of neurotoxic proteins. We took advantage of our previous studies showing amelioration of autophagy in double knockout Vps13a −/− ; Lyn −/− mice as well as in Vps13a −/− animals treated with nilotinib [35,36]. In basal ganglia from Vps13a −/− ; Lyn −/− mice, the absence of LYN activity prevented NRF2 activation and up-regulation of NRF2-related cytoprotective systems, including NQO1, SRX1 or TRXR. These effects were associated with down-regulation of PRX2 and PRX5 and decrease of PrxSO3, supporting the strategy of improving autophagy to limit oxidation. As proof of concept, we found reduced protein oxidation and NRF2 activation in basal ganglia from Vps13a −/− mice treated with nilotinb ( Figure 6), a drug previously tested in patients with PD, dementia with Lewy Bodies and AD in different clinical studies [59][60][61][62][63][64]. A major limitation of these human studies is late initiation of treatment long after onset and progression of PD. Thus, initiation of treatment(s) directly targeting oxidation (e.g., antioxidant) or reducing oxidation by improvement of autophagy (e.g., Lyn inhibitors) must begin as early as possible in the disease course to prevent irreversible cellular damage. Indeed, PRX5 levels are reduced in nilotinibtreated Vps13a −/− mice and appear to be more specific than PRX2 levels as indicators of neuro-oxidation and neuroinflammation, whereas plasma PRX5 levels might serve to guide clinical decision-making in human ChAc. Consistent with this observation, the modulation of plasma PRX5 levels in rodents and humans after brain damage [23,26,65,66] supports plasma PRX5 as both a marker of neurologic damage and a possible therapeutic target.
In conclusion, our study highlights the novel role of oxidation as a contributor to the pathogenesis of ChAc in mice genetically lacking the Vps13 gene product, chorein. In Vps13a −/− animals, activation of Nrf2 and related up-regulation of antioxidant systems is inadequate in the context of the prolonged oxidation linked to the impaired autophagy that characterizes ChAc. Since pathological protein oxidation is related to accumulation of neurotoxic proteins, therapeutic strategies to improve protein clearance represents an interesting therapeutic approach to ChAc. PRX5 might thus act both in cytoprotective and anti-inflammatory manners. Our observation that the increased basal ganglia expression and plasma levels of PRX5 in Vps13a −/− mice are reduced by nilotinib treatment lends support to the use of plasma PRX5 levels as disease biomarker possibly transferable to human studies. Collectively, our data further support the initiation of therapeutic Lyn inhibition as soon as possible after diagnosis of ChAc, to produce the highest likelihood of preventing irreversible cellular damage related to the natural history of the disease.