Fluvoxamine maleate alleviates amyloid-beta load and neuroinflammation in 5XFAD mice to ameliorate Alzheimer disease pathology

Chemicals- RPMI-1640 (Sigma-Aldrich-R6504), Dulbecco’s modified eagle medium (Sigma-Aldrich-D1152), Penicillin (Sigma-Aldrich-P3032), Streptomycin (Sigma-Aldrich-S6501), Sodium bicarbonate (Sigma-Aldrich-S5761), Fetal Bovine Serum (FBS) (GIBCO-10270106), Phosphate buffer saline (PBS) (Sigma-Aldrich-D5652), EDTA (Invitrogen-15575038), Lipopolysaccharide (LPS) (Sigma-Aldrich-L3129), Adenosine 5’-triphosphate (ATP) (Sigma-Aldrich-A6419), Acrylamide (MP Biomedical-193982), Glycine (MP Biomedical-194825),Triton X-100 (Sigma-Aldrich-T8787), Albumin Bovine Fraction V (MP Biomedical-160069), Phenylmethylsulfonyl fluoride (PMSF) (MP Biomedical-195381), Skimmed milk (Himedia-GRM1254), Strataclean resin (Agilent-400714-61), Bafilomycin A1 (Sigma-Aldrich-B1793), Rapamycin (Sigma-Aldrich-R8781),DAPI (Sigma-Aldrich-D9542), glycerol (Sigma-Aldrich-G5516), Tween 20 (Sigma-Aldrich-P7949), HEPES (Sigma-Aldrich-H3375), Paraformaldehyde (Sigma-Aldrich-P6148), MTT (Sigma-Aldrich-M5655), Trizma (Sigma-Aldrich-T6066), SDS (Sigma-Aldrich-L3771), Uric acid sodium salt (Sigma-Aldrich-U2875), Suberic acid (Sigma-Aldrich-S1885), OPTI-MEM media (Gibco-11058-021), β-Amyloid (Anaspec-AS-60479).

Antibodies- NLRP3 (CST-15101S), ASC (Santa Cruz Biotechnology-SC-22514), CASP1 (Santa Cruz Biotechnology- SC-56036), anti-mIL-1β (R & D biotechnology-AF-401-NA), pCAMKK2 (CST-12818S), PRKAA2 (CST-2532S), pPRKAA2 (CST-2535S), pULK1 (CST-14202S), ULK1 (Santa Cruz Biotechnology-SC-33182), FIP200 (CST-12436S), BECN1 (Santa Cruz Biotechnology-SC-48341), ATG5 (Santa Cruz Biotechnology–SC-133158), ATG7 (Santa Cruz Biotechnology-SC-33211), ATG13 (CST- D4P1K), NF-κB (Santa Cruz Biotechnology-SC-372), pNF-κB (Santa Cruz Biotechnology-SC-136548), MTOR (CST-2972S), pMTOR (CST-5536S), LAMP1 (Santa Cruz Biotechnology-SC-20011), HRP-linked anti-goat IgG (Santa Cruz Biotechnology-SC-2354), siRNA PRKAA2 (CST-6620S), GFAP (CST-80788S), IBA1 (CST-17198S), HRP-linked anti-rabbit IgG (CST-7074S), HRP-linked anti-mouse IgG (CST-7076S), anti-mouse IgG Alexa flour 488 (CST-4408S), anti-mouse IgG Alexa flour 555 (CST-4409S), anti-rabbit IgG Alexa flour 488 (CST-4412S), anti-rabbit IgG Alexa flour 555 (CST-4413S), anti-ACTB (Sigma-Aldrich-A3854), ANTI-LC3B-II (Sigma-Aldrich-L7543), Anti-p62/SQSTM1 (Sigma-Aldrich-P0067).

Kits and reagents- PVDF Membrane (Millipore-ISEQ00010), Precision plus protein markers (Bio-Rad-161-0375), ECL-kit (Millipore-WBKLS0500), Bradford reagent (Bio-Rad-500-0006), FuGENE HD (Promega-E2313), mouse TNF-α ELISA kit (Invitrogen-88-7324-88), mouse IL-6 ELISA kit (Invitrogen-88-7064-88), mouse IL- 1β ELISA kit (Invitrogen-88-7013-88), mouse IL- 18 ELISA kit (Invitrogen-88-50618-88), Human Aβ42 ELISA kit (Invitrogen-KHB3441).

The synthesis of fluvoxamine maleate was done by a well-established method (Figure 1A) (WO2014178064, US 6,433,225 B1). The scheme for the synthesis of fluvoxamine maleate (Scheme 1) and data related to its characterization are given in the Supplementary Material File.

In order to activate the NLRP3 inflammasome in primary astrocytes, N9 microglial cells were seeded in 24 well plates and allowed to grow for 24 h. The cells were then treated with LPS (1 μg/ml) and incubated for next 24 h. The LPS activated microglial conditioning media was collected and primary astrocytes were pretreated with the conditioned medium for 24 h. Next day, primary astrocytes were primed with LPS (1 μg/ml) for 4 h and treated with different concentrations of FXN and standard MCC 950 (100 nM) for 1 h in serum free media, followed by ATP (5 mM) stimulation for 30 min.

After activation of NLRP3 inflammasome and drug treatments (as mentioned in the previous section), the supernatants were collected and analyzed to assess the levels of IL-1β and IL-18 using ELISA following the manufacturer’s instructions (Invitrogen). For detection of LPS induced pro inflammatory cytokines, TNF-α and IL-6, primary astrocytes were seeded in 24 well plate and pretreated with N9 conditioning media. Then the astrocytes were incubated with LPS for 24 h and the levels of pro-inflammatory cytokines were examined in the supernatants. The levels of cytokines were normalized by dividing the values with total protein content present in the sample. For in vivo studies, brain cortex was homogenized in RIPA buffer and centrifuged at 12000 rpm for 20 min at 4°C. The supernatants were collected and analyzed for IL-1β, TNF-α and IL-6 cytokine levels. After completion of 2 months oral dosing, blood was collected from retro-orbital plexus and the Aβ₄₂ levels in plasma of 5X FAD mice were analyzed using ELISA following the manufacturer’s guidelines (Invitrogen).

After drug treatments under NLRP3 inflammasome activation condition, primary astrocytes were lysed in the cold buffer consisting of 1% cocktail, sodium orthovanadate (1 mM), KCL (150 mM), PMSF (0.1 mM), HEPES-KOH (20 mM) and 1% of NP-40. The lysed cells were centrifuged at 330g for 10 min at 4°C and the supernatant was collected to analyze the protein expression by western blotting. The cell pellets were washed with PBS and 500 μL of cold PBS was added to the cells. Suberic acid (2 mM) was added and incubated with the pellets for 30 min at 37°C for cross-linking of ASC proteins. Afterwards the pellets were centrifuged at 330g for 10 min at 4°C and 2X Laemmli buffer was added to the cell pellets. Cell lysates were heated at 95°C for 5 min and proteins were analyzed by western blotting.

For in vitro protein analysis by western blotting, primary astrocytes were seeded in 60 mm petri dishes and treated with FXN and standard MCC 950 under NLRP3 inflammasome activation condition. Cytokines secreted in the media were concentrated using Strataclean resin according to the instructions provided by manufacturer and cells were lysed with RIPA buffer [PMSF (2mM), Na3OV4 (0.5mM), NaF (50mM) and 1% cocktail] and centrifuged at 12,000 rpm for 20 min at 4°C. For in vivo studies, hippocampi from 5X FAD mice brain were homogenized and the lysed samples were centrifuged. Supernatants were collected and the protein content was estimated using Bradford’s reagent. For in vitro studies 60 μg protein and for in vivo studies 20 μg protein was loaded for separation using SDS PAGE and proteins were transferred to PVDF membrane for 2 h at 4°C. The blots were blocked using 5% BSA or 5% skimmed milk for 1 h at RT and incubated overnight with primary antibody at 4°C. Afterward blots were incubated with HRP-conjugated secondary antibody for 2 h at RT. Chemiluminescent HRP substrate (Millipore) was used for detection of protein bands and visualized by Chemidoc system (Syngene G:BOX chemi XT 4).

Autophagy flux was measured by treating primary astrocytes with different concentrations of FXN in the presence and absence of bafilomycin A1 for 24 h. The protein expression of LC3B-II and SQSTM1 were analyzed by western blotting.

For confocal microscopic examination of the proteins, primary astrocytes were seeded in 6-well plate over coverslip. After different treatments, cells were washed thrice with PBS and fixed with 4% paraformaldehyde for 15 min. The cells were permeabilized with 0.2% Triton X-100 for 7 min and blocked using blocking buffer (2% BSA and 0.2% Triton X-100) for 1 h. The cells were incubated overnight with primary antibody at 4°C. Next day, the cells were incubated with secondary antibody (Alexa flour 555 or Alexa flour 488) for 1 h at RT and washed thrice with PBS. After counterstaining the nuclei with DAPI, the coverslips were mounted using the mounting media and the images were acquired using Yokogawa CQ1 Benchtop High-Content Analysis System at 40x or 60x.

Aβ₄₂-HiLyte flour488 peptide was prepared following the manufacturer’s protocol (Anaspec Inc.). Primary astrocytes seeded on coverslips were treated with FXN or rapamycin. After 12 h, 2 μg/ml fluorochrome tagged (Hilyte flour488) Aβ₄₂ protein was added to cells for another 12 h. Bafilomycin A1 was given 3 h prior to experiment termination. Cells were washed with PBS and fixed with 4% PFA, afterwards cells were permeabilized with Triton-X 100, and the nuclei were counterstained with DAPI. Slides were prepared and images were taken and analyzed in the Yokogawa CQ1 Benchtop High-Content Analysis System at 40x or 60x.

Primary astrocytes were grown in 6-well plate for confocal microscopic analysis and 60-mm dishes for western blotting. For transfection of PRKAA2, cells were incubated with OPTI-MEM media and PRKAA2 siRNA in FuGENE HD was added for 24 hours. Transfected primary astrocytes were incubated with microglial conditioning media for 24 h, followed by FXN (78 nM) treatment under NLRP3 inflammasome activation condition and the samples were analyzed by western blotting and confocal microscopy at 60x (Yokogawa CQ1 Benchtop High-Content Analysis System).

For in vivo studies, FXN was formulated in 5% of DMSO, 30% PEG400, 20% PEG200 and 45% distilled water. LPS and ATP were dissolved in PBS.

Six months old 5XFAD transgenic mice were used in the study and randomly divided into three groups (n = 5). C57BL6/J mice were used as Control wild type (WT). They were housed under a 12-hour light/dark cycle in a temperature (65–75°F; ˜18-23°C) and humidity-controlled (40–60%) environment, supplied with free access to the food and water (ad libitum). Prior to initiation of study all the animals were acclimatized for one week under standard laboratory conditions, animals were drug naive with no prior procedures performed. All testing were performed from 1 to 4 p.m. Mice were randomized in groups based on their body weights for all behavioral assays and testing were performed by an experimenter blinded to the treatment groups. Total study duration was 2 months. All experiment protocols were approved by the Institutional Animal Ethics Committee (IAEC) (IAEC approval no.- 321/82/2/2023), and followed the Committee for Control and Supervision of Experiments on Animals (CCSEA; Ministry of Environment and Forest, Government of India) guidelines for animal care.

The open field arena box (60 cm x 45 cm x 25 cm) made up of white colored non-reactive plastic was used for all the assessments. The mice were acclimatized to the testing room 30 minutes prior to experiment. Individual mice were placed in the center of the open field arena and allowed to freely explore it for 5 minutes. The arena box was cleaned with a 70% ethanol after every trial. The locomotor activity and exploratory behavior of mice were recorded using a video camera connected to AnyMaze software. Mice were assessed in the open field and the following parameters were recorded automatically: Total distance travelled (cm), average speed (m/s), time spent in the center area (sec), and time spent in the corners (sec).

The eight-armed radial arm maze (UGO Basile) was used to assess spatial memory behavior. The mice were habituated to the maze for three days before the experiment. Each mouse was trained to reach to the baited arm having a butter cookie as food reward and flag as a visual cue. Mice were trained for five days to locate the baited arm. Memory retention trials include recording of several parameters for three consecutive days. The movement of mice was tracked with the help of automated AnyMaze software connected with a video tracking camera.

The neuromuscular coordination of mice was recorded by rotarod. For this assessment mice were trained for three consecutive days with three trials per day. The maximum time allowed on the rotarod during the experiment was 300 seconds. The rotarod was initially set at a speed of 4 rpm at the start of the test and accelerated to 40 rpm over 300 seconds. The latency to fall from the rotarod was recorded for each mouse, and the mean latency to fall for each group was calculated.

After the treatment completion, the hippocampi were isolated from the mice brains and the 10 μm thick tissue-sections were prepared. For IHC, the slides were deparaffinized and histochemical staining was carried out. The sections were incubated with primary antibody (overnight at 4°C) under humid conditions after blocking using 2% BSA solution in TBS. For analysis of total Aβ, sections were incubated with anti-Aβ antibody (1-42) D3E10 followed by secondary antibody (Alexa fluor 488) and for localizing reactive astrocytes, anti-GFAP was used followed by secondary antibody (Alexa fluor 555). The alteration in autophagy levels were assessed by examination of LC3B levels in the tissue sections of mice hippocampi. The sections were immuno-stained with LC3B followed by secondary antibody (Alexa fluor 488). The nuclei were counterstained using DAPI. The images were acquired and quantified using Confocal Quantitative Image Cytometer at 40X (Yokogawa CQ1).

To assess cytotoxicity, primary astrocytes were cultured in 96-well plates and exposed to varying concentrations of FXN for 24 and 48 hours. Four hours before termination, MTT dye was added to each well. Termination involved the addition of DMSO to dissolve the formazan crystals, followed by measuring the absorbance at 570 nm.

Statistical analyses were analyzed using GraphPad Prism 9 software. The data shown here are Mean±SD of three independent experiments. The response of independent mice from the group was noted for animal experiments and the mean was calculated. The statistical analysis of data was calculated using one-way ANOVA, followed by post-hoc Bonferroni test. The p-value<0.05 was considered to be statistically significant with values assigned as ****p < 0.0001, ***p < 0.001, **p< 0.01, *p< 0.05 and ns= not significant.

NLRP3 inflammasome is an important player in neuroinflammation, so we treated the LPS-primed primary astrocytes with FXN before induction with ATP to analyze the anti-NLRP3 inflammasome activity of FXN. The ELISA results showed that FXN could inhibit NLRP3 inflammasome even at very low concentrations of nine nanomolar, which was evidenced by the measurement of released IL-1β and IL-18 as inflammasome activation products. FXN inhibited the release of IL-1β and IL-18 with IC50 values of 16 nM & 25.96 nM, respectively (Figures 1B, C). We also analyzed the inhibitory effect of FXN on NLRP3 inflammasome by using different types of second stimulus in three different types of cells. In nigericin-induced microglial N9 cells, FXN inhibited the release of IL-1β with an IC50 value of 7.02 nM, in ATP induced macrophage J774A.1 cells, it showed IC50 value of 18.40 nM, and in hemin induced THP-1 cells, FXN displayed the IC50 value 17.24 nM (Supplementary Figures S1A–C). Based on the concentration-dependent inhibitory activity of FXN; we chose 78 nM concentration for further detailed studies in astrocytes. We also confirmed the inhibition of release of cleaved IL-1β and CASP1 through western blotting of the proteins collected from the supernatant of primary astrocytes treated with FXN (Figure 1D). The analysis of whole cell lysates revealed reduced levels of pro-IL-1β and NLRP3, while pro-CASP1 was unchanged (Figure 1D; Supplementary Figure S1D). Further, the oligomerization of ASC is an important step in the activation of NLRP3 inflammasome. Therefore, we analyzed the effect of FXN on ASC by using two methods, including confocal microscopy and western blotting. Both experiments showed similar results with the potent inhibitory effect of FXN on ASC oligomerization (Figures 1E, F; Supplementary Figures S1E, F). We also analyzed the co-localization of CASP1 and NLRP3 in the presence of FXN. Data revealed that FXN and MCC950 (used as a standard) had a similar inhibitory effect on the co-localization of CASP1 and NLRP3 (Figure 1G; Supplementary Figure S1G).

In the initial experiments, we treated the LPS-primed astrocytes with FXN to know whether it could inhibit the assembly of the NLRP3 inflammasome complex. We also wanted to know if it can inhibit the NLRP3 inflammasome at the transcriptional level by inhibiting the NF-κB. Therefore, we treated the cells with FXN before priming them with LPS. We found that FXN could significantly inhibit NLRP3 inflammasome at a concentration of 78 nM contrary to nine nanomolar, observed during FXN treatment post-priming. The ELISA measurement of IL-1β showed an IC50 value of 131.1 nM (Figure 2A). We confirmed the inhibitory effect of FXN on NF-κB by direct and indirect methods. The dwindling levels of NF-κB dependent proinflammatory cytokines TNF-α and IL-6 after treatment with FXN at 78 nM indicated the inhibition of NF-κB in astrocytes (Figures 2B, C). Direct evidence for inhibition of NF-κB pathway came from western blot analysis of NF-κB in the cytosolic and nuclear fraction of astrocytes treated with FXN. The data showed that the movement of NF-κB, p65 from the cytoplasm to the nucleus is hindered in the cells treated with FXN in the presence of LPS and ATP stimuli, wherein the levels of NF-κB in the cytoplasm were significantly higher than in the untreated control. On the contrary, its level in the nucleus was significantly reduced (Figures 2D–F).

After confirming the inhibitory effect of FXN on NLRP3 inflammasome, we wanted to know its inhibition mechanism. Therefore, we focused on autophagy as a major mechanism regulating inflammasome activity. We found that FXN could induce autophagy at a low concentration of 39 nM; however, its effect was enhanced with the increasing concentration (Figure 3A; Supplementary Figure S2A). Further, to confirm the completion of autophagy, we calculated the autophagy flux by using LC3B-II and SQSTM1 expression after treatment of astrocytes with FXN (78 nM) in the presence and absence of end-stage autophagy inhibitory bafilomycin A1. We found an autophagy flux at all the tested time points through 24 h of astrocyte treatment with FXN (Figure 3B; Supplementary Figure S2B). Induction of autophagy by FXN was also confirmed by confocal microscopy analysis of co-localization of LC3B-II and LAMP1 in primary astrocytes. The cells treated with FXN and rapamycin showed a significant amount of co-localization of these proteins, while in the presence of bafilomycin A1, FXN-treated cells did not show any observable co-localization of LC3B-II and LAMP1 (Figures 3C, D). After confirming the induction of autophagy by FXN, we checked if it can induce autophagy in primary astrocytes under inflammatory conditions where the cells have been challenged with LPS and ATP. We found that the treatment of astrocytes with FXN before ATP induced the expression of LC3B-II, while SQSTM1 was downregulated, thus confirming the induction of autophagy (Figure 3E; Supplementary Figure S2C). We further analyzed the expression of various proteins involved in the induction of autophagy. We found that FXN induced autophagy through CAMKK2 mediated induction of PRKAA2 pathway leading to activation of various proteins including pPRKAA2 (Thr 172), pULK (Ser 317), ATG13, FIP 200, BECN1, ATG5 and ATG7 and downregulation of pMTOR (Figure 3F; Supplementary Figures S2C). However, induction of autophagy can also play a detrimental role in cell survival. Therefore, to rule out any toxicity due to induction of autophagy by FXN, we treated the primary astrocytes with various concentrations of FXN for up to 48 h and checked their viability by MTT assay. However, we did not find any significant toxicity of FXN even at 320 times higher concentration (25 µM) of FXN (Supplementary Figure S2D).

For the confirmation of the involvement of autophagy in FXN-mediated inhibition of NLRP3 inflammasome, we inhibited the autophagy through genetic and pharmacological methods. We pretreated the primary astrocytes with siPRKAA2 and analyzed the expression of autophagic and NLRP3 inflammasome-related proteins. The inhibition of PRKAA2 led to the reversal of the autophagic effect of FXN, which was reflected in the form of a marked decrease in the expression of LC3B-II and BECN1, while SQSTM1 expression was significantly increased (Figure 4A; Supplementary Figure S3A). Similarly, the inhibition of NLRP3 inflammasome by FXN was also reversed in the cells pretreated with siPRKAA2. The cleaved IL-1β levels were significantly increased, whereas the reduction in expression of NLRP3 was also reversed in the cells treated with FXN in the presence of siPRKAA2 (Figure 4A; Supplementary Figure S3A). We also confirmed these findings by using confocal microscopy, where the co-localization of CASP1 and NLRP3 was restored to the level present in cells primed with LPS and ATP when they were treated with FXN along with knockdown of PRKAA2 by using siRNA (Figure 4B; Supplementary Figure S3B). The knockdown of PRKAA2 was confirmed by using western blotting, where the expression of PRKAA2 and pPRKAA2 was found to be significantly reduced (Figure 4C). We found that a similar kind of reversal of the inhibitory activity of FXN against NLRP3 inflammasome could be achieved if the autophagy is inhibited by using the end-stage pharmacological inhibitor bafilomycin A1. The cleavage of IL-1β was almost completely inhibited, while the expression of NLRP3, NF-κB (p65) along with autophagy markers LC3B-II, SQSTM1 was restored when the cells were treated with FXN in the presence of bafilomycin A1 (Figure 4D; Supplementary Figure S3C).

To confirm whether the autophagy induced by FXN can help in the clearance of Aβ_42, we treated the cells for a total of 24 h with FXN, including 12 h treatment with hilyte fluor 488 tagged Aβ₄₂ The confocal analysis revealed that the cells treated with FXN and rapamycin were almost completely devoid of green fluorescence indicating clearance of Aβ₄₂ in comparison to control cells. Further, 3 h treatment of cells with autophagy inhibitor bafilomycin A1 prior to termination of the experiment completely repressed the clearance of Aβ_42, as indicated by the same level of green fluorescence as that of control cells (Figures 5A, C). The involvement of autophagy in the clearance of Aβ₄₂ by FXN was further confirmed when we inhibited the autophagy by using siRNA against PRKAA2. The clearance of Aβ₄₂ by FXN in the presence of siPRKAA2 was similarly stopped as we observed with bafilomycin. A1 (Figures 5B, D).

After confirming the mechanism of inhibition of NLRP3 inflammasome by FXN, we wanted to know if its dual pharmacological activity can improve Alzheimer disease pathology. Therefore, we treated the 6-month-old 5XFAD mice for two months and analyzed multiple behavior parameters. In the radial arm maze test, the untreated 5XFAD mice displayed significant deterioration of neuronal functions in comparison to wild-type control (Figure 6A). However, the groups treated with FXN (5 and 10 mg/kg) took significantly less time and traveled a reduced distance in finding the baited arm and spent more time in the baited arm apart from fewer entries into non-baited arms in comparison to untreated 5XFAD mice, indicating improvement in the spatial memory (Figures 6B–E). In the open field test, untreated 5XFAD mice spent significantly more time in the center than in corners and showed little interest in exploring the surroundings (Figure 6F). However, the mice treated with FXN covered a significantly higher distance at greater speed and showed normal behavior by spending more time in the corners than in the center (Figures 6G–K). We also observed a significant improvement in the neuromuscular coordination in the mice treated with FXN, which showed delayed latency to fall than the untreated 5XFAD mice (Figure 6L). In all the behavior parameters studied, the mice treated with FXN behaved more like wild-type animals than the untreated 5XFAD control group.

Improvement in neuronal health of 5XFAD mice after treatment with FXN led us to investigate the underlying cause for such change. Therefore, through immunohistochemistry (IHC), we checked the levels of Aβ₄₂ in the hippocampus after two months of treatment with FXN. We found a significant decline in the levels of Aβ₄₂ in the hippocampus of both groups of mice treated at 5 and 10 mg/kg, respectively compared to the control group, as indicated by reduced green fluorescence (Figure 7A). The average number of Aβ₄₂ plaques, skeletal length of Aβ₄₂ plaques, branch count and circumference of Aβ₄₂ plaques were significantly reduced in FXN treated groups as compared to 5XFAD control group (Figures 7B–E). Aβ₄₂ is strongly related to activation of astrocytes, we therefore checked if reduced levels of Aβ₄₂ had any impact on the reactive phenotype of astrocytes. The analysis of GFAP in the vicinity of Aβ₄₂ revealed that as the levels of Aβ₄₂ were depleted after treatment with FXN, the expression of GFAP reduced proportionately, as indicated by reduced red color fluorescence (Figures 7A, F). Further, the reduction of Aβ₄₂ levels in the brain also led to reduced circulatory levels of Aβ₄₂ in the blood plasma of 5XFAD mice treated with FXN in comparison to untreated control mice, as observed through ELISA (Figure 7G).

Based on the effect of FXN on in vitro and in vivo clearance of Aβ₄₂, we hypothesized that autophagy induced by FXN is responsible for the reduced level of Aβ₄₂ in the hippocampi of 5XFAD mice brains. Therefore, to test the hypothesis, we analyzed the expression of LC3B-II in the hippocampi and found a highly significant increase in its levels in the mice treated with FXN (Figure 8A). Further, analysis of expression of various autophagic proteins including pCAMKK2 (Ser 511), pPRKAA2 (Thr 172), pMTOR (Ser 2448), LC3-IIB, pULK1 (Ser 317), BECN1, and ATG5 revealed a significant upregulation of these proteins in the hippocampi of mice treated with FXN in comparison to untreated mice, while the effect of FXN at 10 mg/kg was more pronounced as compared to the dose of 5 mg/kg (Figure 8B; Supplementary Figures S4A).

The analysis of pro-inflammatory proteins in the hippocampi also supported the anti-inflammatory effect of FXN. We found that treatment of mice with FXN significantly reduced the levels of both NF-κB, p50 and p65 subunits confirming its effect on the most important inflammatory pathway. Further analysis of NLRP3 inflammasome proteins showed significant down-regulation of NLRP3, which significantly reduced ASC oligomerization and cleaved CASP1 and IL-1β in the mice treated with FXN (Figures 8C, D; Supplementary Figures S4B). The reduction of inflammation in the hippocampi was indicated by reduced levels of microglial marker protein IBA1 (Figure 8C; Supplementary Figures S4B). We also observed a significant decline in the levels of pro-inflammatory cytokines TNF-α, IL-6 and IL-1β, which were measured in the cortex region of the brain by using ELISA (Figures 8E–G). These data emphasized a potent effect of FXN on the pathology of AD in 5XFAD mice.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.