https://orcid.org/0000-0001-5831-3307
Abstract
Machado–Joseph disease or spinocerebellar ataxia type 3 is an inherited neurodegenerative disease associated with an abnormal glutamine over-repetition within the ataxin-3 protein. This mutant ataxin-3 protein affects several cellular pathways, leading to neuroinflammation and neuronal death in specific brain regions resulting in severe clinical manifestations. Presently, there is no therapy able to modify the disease progression. Nevertheless, anti-inflammatory pharmacological intervention has been associated with positive outcomes in other neurodegenerative diseases. Thus, the present work aimed at investigating whether ibuprofen treatment would alleviate Machado–Joseph disease.
We found that ibuprofen-treated mouse models presented a significant reduction in the neuroinflammation markers, namely Il1b and TNFa mRNA and IKB-α protein phosphorylation levels. Moreover, these mice exhibited neuronal preservation, cerebellar atrophy reduction, smaller mutant ataxin-3 inclusions and motor performance improvement. Additionally, neural cultures of Machado–Joseph disease patients’ induced pluripotent stem cells-derived neural stem cells incubated with ibuprofen showed increased levels of neural progenitors proliferation and synaptic markers such as MSI1, NOTCH1 and SYP. These findings were further confirmed in ibuprofen-treated mice that display increased neural progenitor numbers (Ki67 positive) in the subventricular zone. Furthermore, interestingly, ibuprofen treatment enhanced neurite total length and synaptic function of human neurons. Therefore, our results indicate that ibuprofen reduces neuroinflammation and induces neuroprotection, alleviating Machado–Joseph disease-associated neuropathology and motor impairments. Thus, our findings demonstrate that ibuprofen treatment has the potential to be used as a neuroprotective therapeutic approach in Machado–Joseph disease.
Introduction
Machado–Joseph disease (MJD) is a progressive neurodegenerative disease caused by the over-repetition of the CAG triplet (that encodes for the glutamine amino acid) in the ATXN3 gene (1). This codon over-repetition is translated into an expanded glutamine tract within ataxin-3 protein, which when over 55 glutamine repetitions becomes toxic. The mutated protein accumulates in neuronal intranuclear inclusions and impairs several cellular functions, such as axonal transport, calcium homeostasis, mitochondrial function and the ubiquitin-proteasome pathway. Post-mortem pathoanatomical studies revealed widespread neuronal death in specific brain regions, including basal ganglia, thalamus, midbrain, pons and cerebellum (2,3). MJD patients display severe motor (ataxia, scanning speech and dysphagia) and non-motor (minor cognitive impairments and memory deficits) symptoms (4,5). Presently, there is no therapy able to modify the disease progression.
Neuroinflammation is a physiological response of the central nervous system (CNS) to a wide spectrum of homeostasis-disrupting conditions such as neurodegenerative diseases, as Alzheimer’s, Huntington’s or MJD (6). However, sustained neuroinflammation is harmful, leading to neuronal death and synaptic impairments, negatively interfering over neurodegenerative diseases progression (7,8). In MJD, an increased expression of inflammatory markers in patients, mouse and cellular models has been reported (9–11) and some cytokines variants, namely IL6*C allele, were associated to an earlier age onset (12).
Ibuprofen reduces neuroinflammation in the lentiviral-induced mouse model. C57/BL6 male mice with 8 weeks of age were injected with lentivirus encoding human mutant ataxin-3 in the striatum and treated (Ib) or not (Cnt) with ibuprofen during 4 weeks. Wild-type control (not injected) mice were also evaluated. (A) The relative Il6, Il1b and Tnfa mRNA levels in the striatum of mice, assessed with quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) and normalized for the Cnt, revealed a significant decrease in the proinflammatory interleukins levels. (B) Representative images (control: n = 5 mice and ibuprofen: n = 5 mice) and (C) quantification of western blot analysis of total IκBα (IκBα-T) and phosphorylated IκBα (IκBα-P) protein levels, normalized for the β tubulin I and for the Cnt mice, indicated a significant reduction of the IκBα-P in Ib-treated mice. (A and C): Cnt: n = 11 mice, Ib: n = 11 mice and Wt = 6 mice. Data are expressed as mean ± SEM, *P < 0.05, **P < 0.01 and ***P < 0.001; unpaired t-test.
In our laboratory, we have demonstrated that some beneficial therapeutic strategies act through reducing neuroinflammation. Namely, transplantation of neural stem cells (NSC) in truncated mutant ataxin-3 transgenic mice significantly improved motor impairments and neuropathology by reducing neuroinflammation (9) and the adenosine A2A receptor blockage reduces astrogliosis and microgliosis improving MJD-associated neuropathology (13). These evidences laid the groundwork for evaluation of anti-inflammatory approaches for therapy of MJD. Ibuprofen is a non-steroid anti-inflammatory drug with an extensive clinical use that inhibits cyclooxygenase-1 and 2 reducing proinflammatory prostaglandins production (14). This drug has led to promising results in Alzheimer’s (15,16) and Parkinson’s (17–20) diseases improving synaptic and cognitive functions.
Therefore, in this work we investigated in MJD patients’ iPSC-derived neural cultures and in two mouse models, a lentiviral-induced mutant ataxin-3 model and a truncated human ataxin-3 with an extended glutamine tract transgenic model, whether ibuprofen treatment would reduce neuroinflammation, improve MJD-associated neuropathology and motor impairments. Moreover, we also evaluated whether ibuprofen positively impacted neuronal function markers. We found that ibuprofen reduced neuroinflammation, alleviated neuronal death and mutant ataxin-3 inclusions size, while improving motor impairments. Additionally, ibuprofen treatment improved proliferation markers of neural progenitors in mice, as well as enhanced synapses function markers and the number of neurons with higher total neurite length in MJD patients’ iPSC-derived neural cultures. Thus, this work demonstrates that ibuprofen has potential to be tested as a therapeutic approach for clinical application in MJD treatment.
Results
Ibuprofen administration reduces neuroinflammation
Others and we previously showed that neuroinflammation is present and might contribute to MJD progression (9,11–13), suggesting that anti-inflammatory therapy might alleviate the disorder. Thus, in the present work, we evaluated the therapeutic potential of the anti-inflammatory drug ibuprofen in a lentiviral-induced mutant ataxin-3 mouse model (lentiviral-induced mouse model) (21), a truncated human ataxin-3 with 69 glutamines transgenic mouse model (transgenic mouse model) (22) and in patients’ iPSC-derived neural cultures.
Ibuprofen treatment reduces DARPP-32 depletion and decreases Purkinje cells loss and cerebellar atrophy in mouse models. (A–E) C57/BL6 male mice with 8 weeks of age were injected in the striatum with lentiviral vectors encoding for human mutant ataxin-3 and treated (Ib) or not (Cnt) with ibuprofen during 4 weeks. (F–K) Transgenic female and male mice with 5–6 weeks old were treated (Ib) or not (Cnt) with ibuprofen for 8 weeks. (A and B) Immunohistochemical images of DARPP-32 staining in striatum and (C) DARPP-32-depleted volume quantification, normalized for the Cnt, demonstrated that ibuprofen reduces DARPP-32 neuronal marker loss; Ib: n = 11 mice and Cnt: n = 7 mice. (D) Representative images (control: n = 5 mice and ibuprofen: n = 5 mice) and (E) quantification of western blot analysis, normalized for β-tubulin I and the Cnt, demonstrated that Ib-treated mice have higher levels of DARPP-32 protein in the striatum; Ib: n = 11 mice and Cnt: n = 11 mice. (F and G) Immunohistochemical fluorescence images of Purkinje cells shown by calbindin immunoreactivity (red) revealed that ibuprofen-treated mice have a higher number of Purkinje cells; DAPI (blue). (I and J) Cresyl violet staining revealed that the molecular (ML) and granular (GL) layers are thicker in the mice treated with ibuprofen. (H) Quantification of Purkinje cell numbers, normalized to Cnt; Ib: n = 6 mice and Cnt: n = 6 mice. (K) Cellular layers thickness quantification normalized to Cnt; L5: lobule 5, L9: lobule 9; Ib: n = 7 mice and Cnt: n = 4 mice. Scale bars: A, B, F, G = 50 μm; I and J = 100 μm. Data are expressed as mean ± SEM, *P < 0.05, **P < 0.01 and ***P < 0.001; unpaired t-test.
Ibuprofen treatment increases number and decreases size of mutant ataxin-3 inclusions. (A–E) C57/BL6 male mice with 8 weeks of age were injected in the striatum with lentivirus encoding for human mutant ataxin-3 and treated (Ib) or not (Cnt) with ibuprofen during 4 weeks. (F–H) Transgenic female and male mice with 5–6 weeks old were treated (Ib) or not (Cnt) with ibuprofen for 8 weeks. (A and B) Immunohistochemical images of mutant ataxin-3 inclusions shown by ubiquitin staining in the striatum of mice demonstrated that ibuprofen increases the number of inclusions while reducing its size. (C) Quantification of inclusions number normalized for Cnt; Ib: n = 11 mice and Cnt: n = 9 mice. (D) Size distribution of inclusions shown in (A) and (B); Ib: n = 8 mice and Cnt: n = 8 mice. (E) Quantification of inclusions area shown in (A) and (B); Ib: n = 8 mice and Cnt: n = 8 mice. (F and G) Immunohistochemical fluorescence images of mutant ataxin-3 inclusions stained for hemagglutinin (HA, green, white arrows) of transgenic mice cerebella also shown that ibuprofen treatment increases the number of mutant ataxin-3 inclusions; the upper small inserts are an image detail showing inclusions. (H) Quantification of mutant ataxin-3 inclusions numbers normalized for Cnt; Ib: n = 6 mice and Cnt: n = 4 mice. Scale bars: A and B = 100 μm, F and G = 50 μm. Data are expressed as mean ± SEM, *P < 0.05; #P = 0.064, unpaired t-test.
The mouse models, expressing inflammatory markers as previously described (9,13), were treated with 375 mg of ibuprofen/kg of chow (23,24). This drug dose corresponds to 45–60 mg/kg intake of ibuprofen per day, which is considered enough to inhibit cyclooxygenases, the main pharmacological target of ibuprofen and, to reduce neuroinflammation (23,25) and is equivalent to 270 mg of ibuprofen/day in humans (26). Therefore, we first investigated whether ibuprofen would reduce the abnormally increased proinflammatory interleukins Il6, Il1b and Tnfa messenger RNA (mRNA) levels in the lentiviral-induced mouse model (Fig. 1A). Upon drug treatment, a significant and robust reduction in the Il1b and Tnfa levels (from 1 to 0.378 ± 0.11 and 0.281 ± 0.10, respectively) was observed. Nevertheless, the ibuprofen treatment did not reduce the proinflammatory interleukins to basal levels, given that the wild-type control (not injected) exhibited lower levels of Il1b and Tnfa mRNA as compared to ibuprofen-treated mice. NF-κB is an important inflammatory mediator that is regulated by the IκB-α protein. Reduction of IκB-α phosphorylation results in NF-κB inflammatory activity decrease. Thus, the neuroinflammation of ibuprofen-treated mice was also evaluated through the quantification of IκB-α protein phosphorylation levels. We observed that the phosphorylated IκB-α protein levels were reduced from 1 to 0.651 ± 0.080 in the treated mice (Fig. 1B and C). Therefore, our results indicate that ibuprofen administration reduces neuroinflammation.
Ibuprofen improves MJD-associated neuropathology and motor impairments
Subsequently, we investigated whether ibuprofen administration would alleviate MJD-associated neuropathology in two mouse models. In the lentiviral-induced mouse model, in which mutant ataxin-3 promotes depletion of the neuronal marker DARPP-32, we observed that the ibuprofen-treated mice exhibited a 28% reduction in the DARPP-32 depleted area (from 1 to 0.72 ± 0.053) (Fig. 2A–C). This result was confirmed by western blot, with ibuprofen-treated mice presenting 1.46 ± 0.13 fold higher DARPP-32 protein levels, as compared to non-treated mice (Fig. 2D and E).
This neuronal preservation triggered by ibuprofen treatment was also investigated in a severe transgenic mouse model, which presents extensive neuronal depletion, namely Purkinje cells loss (22). Ibuprofen administration reduced Purkinje cells (calbindin-positive) depletion resulting in a 1.163 ± 0.064 times higher number of cells (Fig. 2F–H), as compared to control mice. Additionally, the cellular layers thickness of two cerebellar lobules, 5 and 9, was evaluated. These two particular lobules were tested because as they are relatively far from each other, their evaluation allows a global vision of the cerebellar atrophy changes promoted by ibuprofen treatment. Lobules 5 and 9 evaluation is particularly important in MJD context given its association to sensorimotor functions (27–29). The thickness of cerebellar granular layers in lobules 5 and 9 were 1.21 ± 0.05 and 1.66 ± 0.04 times larger in ibuprofen-administered mice, respectively (Fig. 2I–K). Additionally, the atrophy of the molecular layer in lobule 5 was also decreased, as its thickness was 1.32 ± 0.06 times larger (Fig. 2I–K) in treated mice. Thus, ibuprofen administration resulted in both neuronal preservation and cerebellar atrophy reduction of a severe mouse model.
We next investigated whether ibuprofen administration would modify the number and size of mutant ataxin-3 inclusions. In the lentiviral-induced mouse model, ibuprofen-treated mice displayed a 1.34 ± 0.09 times increase in the number of mutant ataxin-3 inclusions (Fig. 3A–C). Additionally, a significant change in the size distribution and mean area of inclusions was detected (Fig. 3D and E). Ibuprofen-treated mice exhibited a significant higher number of inclusions smaller than 5 μm in diameter, while the number of inclusions over 45 μm in diameter was reduced (Fig. 3D). These changes were also accompanied by a significant reduction of 22.6% (from 1 to 0.774 ± 0.07) in the inclusions’ mean area of the drug-treated mice (Fig. 3E). The increase in the number of mutant ataxin-3 inclusions induced by ibuprofen was further investigated in the transgenic mouse model. As shown in Figure 3F–H, the number of inclusions was increased 1.473 ± 0.147 times in the drug-treated mice. Overall, our results demonstrate that ibuprofen changes the number and size of the mutant ataxin-3 inclusions. Interestingly, no significant changes in the levels of high molecular weight mutant ataxin-3 (above 200 kDa), soluble mutant ataxin-3 (67 kDa) and the respective toxic fragment of 34 kDa (30) were detected in the lentiviral-induced mouse model upon ibuprofen treatment (Supplementary Material, Fig. S1), indicating that ibuprofen does not change neither the levels of mutant ataxin-3 nor the balance between the different protein species.
The possible positive modulation of ibuprofen in the MJD-associated motor impairments was evaluated in adult, severely impaired transgenic mice. Using the rotarod test at a constant speed, motor performance was assessed over 8 weeks (Fig. 4). Mice treated with ibuprofen displayed a robust motor improvement as compared to control mice. These differences were significant from 6 weeks after starting the treatment (Ib: 69.42 ± 11.33 s; Cnt: 19.00 ± 3.95 s) and were maintained until the end of the experiment.
Improvement of motor coordination impairments in transgenic mice treated with ibuprofen. Transgenic female and male mice with 5–6 weeks old treated (Ib) or not (Cnt) with ibuprofen and wild-type mice were assessed for motor coordination by rotarod performance (at constant velocity, 5 rpm) over 8 weeks. A significant improvement of the motor phenotype in the ibuprofen-treated mice as compared to Cnt was observed. Ib: n = 8 mice, Cnt: n = 6 mice and wild-type: n = 4 mice. Data are presented as mean ± SEM. Two-way ANOVA analysis with Bonferroni post-test; **P < 0.01.
Ibuprofen increases neural progenitors proliferation and synaptic markers. MJD patients’ iPSC-derived neural cultures were treated (Ib) or not (Cnt) with 500 μM of ibuprofen. (A) The relative NOTCH1, MSI1, MKI67, PCNA, DLG4 and SYP mRNA levels were assessed with qRT-PCR and normalized for the Cnt. The mRNA levels of the neural progenitors proliferation markers NOTCH1 and MSI1 and, the synaptic marker SYP were increased in Ib-treated cultures; n = 5 independent experiments. Immunocytochemistry fluorescence images of (B and C) Ki67 (green, white arrows) and (D and E) Msi1 (green) staining of patients’ iPSC-derived neural cultures (C and E) treated (ibuprofen) or (B and D) not (control) with ibuprofen during 7 days revealed that ibuprofen treatment increases neural progenitors proliferation marker Msi1. (F) Quantification of Ki67 and Msi1 number of particles normalized for the number of cells (DAPI positive cells, blue) and for the Cnt cultures; n = 6 independent experiments, scale bars = 50 μm. Data are expressed as mean ± SEM, *P < 0.05; unpaired t-test.
Ibuprofen increases neural progenitors proliferation, neurite length and synapse numbers of human neurons
Neuroinflammation has been associated with disturbed synaptic function, impaired neurite outgrowth and decreased neural progenitors proliferation and differentiation (31–34). To investigate whether ibuprofen treatment would alleviate these impairments in MJD, several synaptic and neural progenitors proliferation markers were investigated in cell and mouse models.
Thus, to investigate the effect of ibuprofen treatment in MJD patients’ neural progenitors proliferation, neuronal neurite length and functional synapse numbers, we took advantage of MJD patients’ iPSC-derived NSC that upon in vitro differentiation originate heterogeneous neural cultures composed by neurons and glia (35), which were treated with 500 μM ibuprofen. The ibuprofen concentration was selected in accordance with a previous work (36) and also because the oral administration of 400 mg of ibuprofen in humans result in a plasmatic concentration of 32 μg/ml (155 μM) (37) and ibuprofen can reach concentrations 3-fold higher (465 μM) in the cerebrospinal fluid as compared to the unbound drug plasmatic concentrations (38). Therefore, the concentration used in vitro in this work is in accordance with the physiological ones obtained in humans. The mRNA levels of synaptic (DLG4 and SYP), neural progenitors proliferation (MSI1 and NOTCH1) and cellular proliferation (MKI67, PCNA) markers were assessed in ibuprofen-treated patients’ iPSC-derived neural cultures. As can be observed in Figure 5A, cultures treated with 500 μM ibuprofen exhibit a significant increase in the mRNA levels of SYP and NOTCH1, and a strong tendency to increase for MSI1 by 1.50 ± 0.14, 1.28 ± 0.10 and 1.53 ± 0.20 times, respectively. To further assess the effect of ibuprofen in the proliferation of neural progenitors, we evaluated Msi1 and Ki67 particles levels in MJD patients’ iPSC-derived neural cultures (Fig. 5B–F). We found that the number of Msi particles increase, 1.614 ± 0.206 times, upon ibuprofen treatment (Fig. 5F), indicating that this drug increases neural progenitors proliferation markers in human neural cultures, without causing cytotoxicity (data not shown).
The possible positive modulation in the synaptic function was further analyzed both in MJD patients’ iPSC-derived neural cultures and in Neuro 2a cells. Treatment of Neuro 2a cells with 500 μM of ibuprofen clearly enhanced neurites length (Fig. 6A and B). In MJD patients’ iPSC-derived neural cultures, the percentage of neurons with total neurite length over 500 μmwas increased, as compared to controls, suggesting that ibuprofen promotes an enhancement of the neurite length (Fig. 6C–E). Moreover, in cultures treated with the drug, a 2.08 ± 0.36 times higher number of excitatory synapses, defined as instances of co-localization of PSD-95 and VGLUT1, was observed (Fig. 6F–H), indicating that ibuprofen enhances the number of functional synapses.
Ibuprofen increases neurite length and synaptic function of MJD patients’ iPSC-derived neurons. (A and B) Neuro 2a cells and (C–H) MJD patients’ iPSC-derived neural cultures treated (Ib) or not (Cnt) with 500 μM ibuprofen. (A and B) Immunocytochemistry fluorescence pictures of β3tubulin labeling (red) in Neuro 2a cells clearly demonstrated the neurite length increase resulting from ibuprofen treatment. (C and D) Immunocytochemistry fluorescence images of the neuronal marker MAP2 staining (red) in patients’ iPSC-derived neural cultures treated (ibuprofen) or not (control) with ibuprofen; scale bars = 100 μm. (E) Quantification of total MAP2-positive neurite length distribution indicates that Ib-treated cultures exhibit an increased percentage of neurons with total neurite length above 500 μm. Cnt: n = 89 neurons, Ib: n = 72 neurons from four independent experiments. (F and G) Immunocytochemistry fluorescence images staining for excitatory synapses, defined as instances of (F.1 and G.1) PSD-95 (green) and (F.2 and G.2) VGLUT1 (red) (F.3 and G.3) colocalized puncta (merge) (white arrows) in MAP2-positive neurite. (H) Quantification of the number of excitatory synapses per neurite length, normalized for the Cnt cultures, revealed that ibuprofen treatment increases the number of excitatory synapses in MAP2-positive neurons. Cnt: n = 41 neurons, Ib: n = 43 neurons from three independent experiments; scale bars = 5 μm. Data are expressed as mean ± SEM, *P < 0.05 and **P < 0.01; unpaired t-test.
To further evaluate the effect mediated by ibuprofen over synaptic function markers, we evaluated mRNA levels of Syp, the presynaptic Slc17a7 (VGLUT1 gene; present in excitatory synapses), the postsynaptic Dlg4, Slc32a1 (VGAT gene; present in inhibitory synapses) and Gphn [gephyrin; involved in inhibitory synapses formation (39)] in the lentiviral-induced mouse model. As can be observed in Figure 7A, in ibuprofen-treated mice, the Dlg4 and Slc32a1 mRNA levels were 1.36 ± 0.07 and 1.35 ± 0.11 times increased, respectively, indicating an increase in synaptic markers levels upon ibuprofen treatment.
Ibuprofen increases neural progenitors proliferation and synaptic markers in the lentiviral-induced mouse model. C57/BL6 male mice with 8 weeks of age were injected in the striatum with lentivirus encoding for human mutant ataxin-3 and treated (Ib) or not (Cnt) with ibuprofen during 4 weeks. (A and B) The relative Mki67, Pcna, Msi1, Notch1, Agrn, Cttn, Neurod1, Syp, Slc17a7, Slc32a1, Dlg4 and Gphn mRNA levels in the striatum of Ib and Cnt mice were assessed and normalized for the Cnt mice. The neural progenitors proliferation marker Msi1, the postsynaptic Dlg4 and the inhibitory synaptic marker Slc32a1 mRNA levels were significantly increased in the ibuprofen-treated mice; Cnt: n = 11 mice and Ib: n = 11 mice. (C and D) Immunohistochemical fluorescence labeling of Ki67 positive particles (white) present in the subventricular zone (SVZ) demonstrated that Ib-treated mice display higher number of proliferating neural progenitors; scale bars = 500 μm. (E) Quantification of Ki67 positive particles in the SVZ normalized for the Cnt mice; Cnt: n = 10 mice and Ib: n = 9 mice. Data are expressed as mean ± SEM, *P < 0.05, **P < 0.01 and ***P < 0.001; unpaired t-test.
To evaluate the possible in vivo modulation of ibuprofen in the neural progenitors proliferation and differentiation, mRNA levels of Mki67, Pcna, Msi1, Notch1, Neurod1 [involved in the terminal differentiation of neuronal cells (40)], Agrn [agrin; implicated in synaptic development, brain plasticity and signaling (41)] and Cttn [cortactin; involved in neuronal growth cone morphology and spreading (42)] were also evaluated. As can be observed in Figure 7B, ibuprofen-treated mice showed a significant 1.76 ± 0.19 fold increase of Msi mRNA levels as compared to Cnt mice. To evaluate whether the positive modulation mediated by ibuprofen in mRNA levels of neural progenitors proliferation and synaptic markers is also possible in non-inflammatory conditions, the Msi1, Neurod1, Dlg4, Slc32a1 and Syp mRNA levels of wild-type control mice (not inject) treated or not with ibuprofen were evaluated (Supplementary Material, Fig. S2). We found that both Neurod1 and Slc32a1 mRNA levels were significantly increased upon ibuprofen treatment.
The in vivo increased levels of neural progenitors proliferation markers were further investigated through the evaluation of the number of Ki67 positive cells in the neurogenic subventricular zone (SVZ). Ibuprofen-treated mice exhibited a close to significance 1.16 ± 0.06 times higher number of proliferating neural progenitors cells in the SVZ, as compared to non-treated mice (Fig. 7C–E).
Discussion
Neuroinflammation contribution to progression of neurodegenerative disorders motivated the test of anti-inflammatory drugs, including ibuprofen, resulting in promising effects in models, and patients of diseases such as Alzheimer’s (15,16) and Parkinson’s (17,18) diseases. In the particular case of MJD, increased levels of inflammatory markers in patients, mouse and cellular models have been reported (9–11). Nevertheless, the potential therapeutic effect of anti-inflammatory therapies in this disease was not investigated so far. Here we found that oral administration of ibuprofen reduced neuroinflammation, neuropathology, motor coordination impairments and increased neurite length and markers of synaptic function and neural progenitors proliferation.
Upon oral administration, ibuprofen is rapidly absorbed, displaying an almost complete bioavailability (43), without major changes in pharmacokinetics parameters with repeated administrations (14). Oral administration besides being an efficient administration route for ibuprofen is also not expensive, easy to implement in clinical practice and does not require specialized professionals for its application, which translates into higher compliance of patients to medication. Therefore, this administration route was adopted in the present work.
The ibuprofen dose tested in both mouse models was 375 ppm (375 mg of drug/kg of chow) that is equivalent to 270 mg of ibuprofen/day in humans (26), which has potential to be implemented in prolonged treatments, as can be inferred from clinical trials and epidemiological studies revealing the mild toxicological profile of ibuprofen administration for long periods of time (>6 months) (14,44).
Two mouse models (a lentiviral-induced mutant ataxin-3 mouse model and a transgenic mouse model carrying a truncated form of human ataxin-3 with 69 glutamines) and two cellular models (Neuro 2a and MJD patients’ iPSC-derived neural cultures) were used in this study. Models carrying the full-length mutant ataxin-3, the lentiviral-induced mouse model and the iPSC-derived neural cultures, were used for molecular and cellular evaluations (Supplementary Material, Fig. S3). The transgenic mouse model, which has a robust representation of the major MJD-associated motor and neuropathological hallmarks (22), namely motor coordination impairments and neuronal death, was used for the evaluation of these parameters. Nevertheless, we are aware of the limitations of this model; therefore, neuropathological hallmarks, as neuronal loss and mutant ataxin-3 aggregation, were also assessed in the mouse model carrying the full-length mutant ataxin-3 (Supplementary Material, Fig. S3).
We found that ibuprofen treatment significantly decreases proinflammatory interleukins Il1b and Tnfa mRNA levels, as well as the Iκ-Bα protein phosphorylation in the lentiviral-induced mouse model. Iκ-Bα is the inhibitor of the proinflammatory NF-κB; as the phosphorylation of Iκ-Bα protein leads to its elimination, consequently, a reduction in Iκ-Bα phosphorylation result in the enhancement of Iκ-Bα levels available to inhibit proinflammatory NF-κB activity (45). Thus, the decreased Iκ-Bα phosphorylation observed with ibuprofen treatment is an indication of neuroinflammation reduction. Nevertheless, the uncompleted reduction of the proinflammatory interleukines mRNA levels towards normalization, observed with the ibuprofen dose tested in this work, is an indication that a higher dose of ibuprofen should be tested, in order to evaluate whether a dosage increase might result in better outcomes.
The prolonged exposure to high levels of proinflammatory modulators leads to neuronal death (7). Therefore, we evaluated whether ibuprofen administration reduces neuronal loss. In the lentiviral-induced mouse model, the drug promoted the preservation of the DARPP-32 neuronal marker, significantly reducing the neuronal loss promoted by the mutant ataxin-3. Accordingly, in the transgenic mouse model, a significant decrease in the Purkinje cells loss, the most affected neurons in this model, and preservation from cerebellar atrophy (cellular layers thickness reduction) was observed. Thus, our results clearly demonstrate that ibuprofen improves the disease-associated neuropathology, by decreasing neuronal loss and preserving cerebellar neuronal architecture.
The presence of mutant ataxin-3 inclusions is also a major MJD hallmark. Aggregated proteins in neurons and glia trigger neuroinflammation (46,47), which per se can interfere with protein conformation promoting protein unfolding that may contribute to the disease pathogenesis (48). On the other hand, ibuprofen has been identified as a protein folding stabilizer (49). Therefore, we evaluated whether ibuprofen administration would modify the number and size of mutant ataxin-3 inclusions. In both mouse models used a significant increase in the number of the mutant ataxin-3 inclusions was observed upon ibuprofen treatment. Interestingly, this was accompanied by a decrease in the inclusions’ mean size, potentially neuroprotective. In fact, we have previously demonstrated that larger inclusions are associated with increased caspase-3 activation and cell death (50). Thus, our results demonstrate that ibuprofen treatment interferes with mutant ataxin-3 aggregation. Either, by changing the protein milieu, such as the inflammatory modulators levels or by directly interfering with the protein folding state. Furthermore, the increase in the number of mutant ataxin-3 inclusions in parallel with an improved MJD-associated neuropathology observed in the present work is in accordance with our previous work with caffeine (13), which supports the hypothesis that the aggregation of mutant ataxin-3 may be a cellular defensive mechanism. These observed improvements, triggered by ibuprofen in the neuropathology, were accompanied by a significant reduction of the motor impairments.
Neuroinflammation is associated with impairments in the proliferation of neuronal progenitors (7,51) and in the synaptic function (7). Thus, we assessed the impact of ibuprofen treatment in the mRNA levels of markers of synaptic function, neural progenitors proliferation and differentiation, as well as synaptic development and brain plasticity in the lentiviral-induced mouse model and in MJD patients’ iPSC-derived neural cultures. In both models, a clear increase of the mRNA levels of neural progenitors proliferation markers, such as Msi1 and Notch1 was observed. This result was further confirmed by the observed enhancement of proliferating cells (Ki67 positive) in the neurogenic SVZ of mice treated with ibuprofen and by the Msi1 enhanced protein levels in the neural cultures treated with the drug. Therefore, our results demonstrate that ibuprofen enhances the levels of markers of neural progenitors proliferation both in a mouse model and in a human in vitro model. The observed changes in the Msi1 levels called our attention because Msi family members, Msi1 in particular, are key regulators in the proliferation and differentiation of neural progenitors (52). Additionally, it has been reported that Msi1 subcellular localization changes during neuronal differentiation (53) and the abrogation of this protein result in neuronal maturation decrease (54). Therefore, the positive modulation played by ibuprofen in the neural progenitors proliferation, illustrated by Msi1 levels change, may contribute to an improvement of the brain neuronal network, by supplying new neurons to replace the affected neurons, in accordance with other neurogenic strategies (55–57).
MJD is also associated with disruption of dendritic development (58) and synapse loss (59). Thus, as ibuprofen has been reported to improve synaptic function (16), we evaluated whether this drug could improve synaptic markers in the lentiviral-induced mouse model and in MJD patients’ iPSC-derived neural cultures. Ibuprofen treatment promoted an enhancement of the mRNA levels of synaptophysin and PSD-95 in these models, together with an increase of excitatory synapse numbers and in the number of neurons with larger total neurite length in the neurons obtained from MJD patients’ iPSC-derived NSC. Additionally, a neurite length increase was also clearly demonstrated in Neuro 2a cells. Thus, our data indicates that ibuprofen significantly improves the synaptic function markers in MJD experimental models, in accordance to the previously described positive synaptic modulation triggered by ibuprofen in Alzheimer’s disease (16,60) and with the described increased neurite length of human NTERA-2 cells-derived neurons treated with 500 μM of ibuprofen (36). Nevertheless, some questions require further investigation, namely whether the beneficial effect of ibuprofen treatment in MJD is maintained upon continued administration and whether secondary effects might occur with continued administration of this drug.
This study demonstrated that ibuprofen administration improves MJD by decreasing neuroinflammation, increasing neurite length and enhancing levels of synaptic function and neural progenitors proliferation markers. Moreover, ibuprofen treatment reduces neuronal loss, while preserving cerebellar architecture and improving motor coordination. Therefore, we demonstrated that ibuprofen produces a significant positive modulation in some of the more important clinical hallmarks of MJD and, consequently, has potential to be used as a therapeutic approach for MJD treatment.
Materials and Methods
Viral vectors production
Viral vectors encoding full-length human mutant ataxin-3 with 72 glutamines (61) and the virus encoding for the multicistronic all-in-one SIN vector expressing the four reprogramming factors Oct4, Sox2, c-Myc and Klf4 (62) were produced in 293 T cells using a four-plasmid system (63). The lentiviral particles were stored at −80°C and its concentration was determined by measuring HIV-1 p24 antigen (RETROtek, Gentaur).
MJD patients’ iPSC-derived neural cultures
The human fibroblasts were obtained under informed consent and following the protocols and guidelines instructed by the Ethics Committee of the Medical Faculty of the University of Coimbra.
Human iPSC were obtained by reprogramming fibroblasts of MJD patients through the introducing of four reprogramming factors, Oct4, Sox2, c-Myc and Klf4 (64) using lentivirus. Then, iPSC were induced to NSC, as previously described (65). NSC were maintained in DMEM/F12 culture medium supplemented with 2.4 g/L of bicarbonate sodium, 1% penicillin-streptomycin (Gibco), 2% B27 (Gibco), 20 ng/ml of bFGF and epidermal growth factor (EGF) (Peprotech) and 5 μg/ml heparin (Sigma) (35). To differentiate NSC into neural cultures, 600 000 cells were plated in MW6 and 300 000 cells were plated in MW12 with coverslips, plates were covered with poly-L-lysine and laminin (Sigma). bFGF, EGF and heparin were removed from the culture medium and 2 μM of all-trans retinoic acid (Sigma) and 5 μM of forskolin (Sigma) were added. Cells were allowed to differentiate and culture medium was replaced every 3 days. For neurite length and synapse numbers assessment cells were differentiated for 4 days, afterwards culture medium was replaced and cells were incubated with 500 μM ibuprofen (Sigma) for 3 days. For qRT-PCR, cells were differentiated for 1 day and then incubated with ibuprofen for 3 days. For Ki67 and Msi1 immunocytochemical staining, cells were immediately incubated with ibuprofen after being plated. To Cnt cells, the same volume of the ibuprofen’s solvent (ethanol) was added.
Neuroblastoma cell culture (Neuro-2a)
Mouse neuroblastoma cells (Neuro-2a cells) obtained from the American Type Culture Collection bank (CCL-131) were maintained in DMEM medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (Gibco) (complete medium). The 100 000 cells plated in MW12 with coverslips covered with Poly-L-lysine were incubated in complete medium supplemented with 20 μM of all-trans retinoic acid, used to induce neurites (66). Ibuprofen (500 μM) was immediately added after cells plating and maintained for 5 days; to Cnt cells the same volume of the ibuprofen’s solvent (ethanol) was added.
Immunocytochemistry
Cultures plated in coverslips were fixed with 4% paraformaldehyde (Sigma) and permeabilized with 1% TritonTM X-100 (Sigma) followed by 1 h blocking with 3% bovine serum albumin (BSA) (Sigma) and overnight incubation with the primary antibodies at 4°C. The following primary antibodies diluted in 3% BSA/PBS were used: mouse anti-MAP2 (1:250, M1406, Sigma), rabbit anti-GFAP (1:400, Z0334, Dako), mouse anti-β3tubulin (1:500, 480099, Invitrogen), rabbit anti-Musashi 1 (1:500, ab21628, Abcam) and rabbit anti-Ki67 antibody (1:500, ab16667, Abcam). Subsequently, coverslips were washed three times with PBS and the incubation with the secondary antibodies was performed for 2 h at room temperature (RT). Secondary antibodies were diluted in 1% BSA/PBS: goat anti-mouse Alexa Fluor-594 (1:200, A-11005, Invitrogen) and goat anti-rabbit Alexa Fluor-488 (1:200, A-11008, Invitrogen). Nuclei were stained with 4, 6-diamidino-2-phenylindoline, DAPI (Applichem). Fluorescence images were acquired with a Zeiss Axiovert 200 imaging microscope.
Total neurite length measurement and excitatory synapses quantification
MJD patients’ iPSC-derived neural cultures, cultivated as previously described, were fixed with 4% paraformaldehyde/4% sucrose in PBS for 15 min at RT, permeabilized with 0.25% Triton X-100 in PBS and blocked with 10% BSA. Primary antibodies diluted in 3% BSA/PBS, rabbit anti-PSD95 (1:1000, D27E11, Cell Signaling), guinea pig anti-VGLUT1 (1:5000, AB5905, Chemicon) and mouse anti-MAP2 (1:250, M1406, Sigma), were incubated overnight at 4°C. Coverslips were then washed three times in PBS and incubated with the secondary antibodies, Alexa Fluor 488 anti-rabbit IgG (1:200, A-11008, Thermo), Alexa Fluor 647 anti-guinea pig IgG (1:500, A-21450, Thermo) and Alexa Fluor 568 anti-mouse IgG (1:200, A-11004, Thermo), for 45 min at 37°C. Coverslips were mounted with Fluorescence Mounting Medium (Dako).
For neurite length assessment, images of whole MAP2-expressing neurons were acquired with an Axio Observer Z1 microscope (Carl Zeiss) using 10× or 20× objectives. Cell body and neurites were digitally traced using the Neurolucida software and measurements were made with the Neurolucida Explorer software (both from MBF Bioscience).
Observation and imaging of synaptic puncta was performed with an Axio Observer Z1 microscope with a 63× oil objective, after randomly selecting MAP2-positive cellular tracts. Image analysis was performed using Fiji software. Instances of co-localization of pre- and postsynaptic markers were quantified as functional synapses. Values were normalized per dendritic section length to the mean value of the control. Imaging and quantification was performed blindly to condition.
In vivo experiments
Animals
The experiments with mice were approved by the Responsible Organ for the Animals Welfare of the Faculty of Medicine and Center for Neuroscience and Cell Biology of the University of Coimbra and were carried out in accordance with the European Community directive (86/609/EEC) for the use of laboratory animals.
Two mouse models were used: a lentiviral-induced mouse model and a transgenic mouse model. For the lentiviral-induced model, C57/BL6 male mice with 8 weeks of age (Charles River) were unilaterally injected in the striatum with lentivirus (400 000 ng of p24), expressing human mutant ataxin-3 with 72 glutamines (21) at a rate of 0.25 μl/min with an automatic injector (Stoelting Co., Wood Dale, USA). The injection was done at the coordinates calculated from bregma: caudal-rostral +0,6 mm, medial–lateral ±1,8 mm and dorsal–ventral −3,3 mm and tooth bar: 0 and the treatment with ibuprofen started the day after the stereotaxic injection. The transgenic mouse model (C57BL/6 background) expresses the N-terminal-truncated human ataxin-3 lacking 286 amino acid residues and with 69 glutamine repeats and an N-terminal hemagglutinin (HA) epitope, under the control of the L7 promoter that is specific of Purkinje cells (22). Transgenic mice and their wild-type littermates with 5–6 weeks old were used in this study. Ibuprofen-treated mice were fed with chow containing 375 ppm (375 mg of drug/kg of chow), control mice were fed with chow without drug, as performed in previous studies (15,23,24). These mice strains have a daily food intake of 3–4 g of chow/day, (67); thus mice’s ibuprofen estimated daily intake was 1.125–1.5 mg, considering that mice average weight is 25 g this translates into the 45–60 mg ibuprofen/kg. Mice of the lentiviral-induced model were treated with ibuprofen during 4 weeks and transgenic mice were treated during 8 weeks. Food and water were provided ad libitum.
Motor phenotype evaluation (rotarod test)
Transgenic mice treated or not with ibuprofen were evaluated for motor coordination with a rotarod apparatus (Letica Scientific Instruments). For this purpose, mice were placed on the rotarod, at a constant speed (5 rpm), for a maximum of 5 min and the latency to fall was recorded. All tests were performed in the same dark room after 30 min of acclimatization. Mice performed four trials for each time point with a 15–20 min rest between trials. For analysis, the mean latency to fall off the rotarod of three trials was used. Mice were trained and performed this test every 1–2 weeks until 8 weeks after the beginning of the treatment. An experienced operator performed the tests in a blind-fashion manner.
Tissue preparation for immunohistochemistry
After deep anesthesia (ketamine/xylazine), mice were intracardiacally perfused with PBS followed by fixation with 4% cold paraformaldehyde. The brains were removed and post-fixed in 4% paraformaldehyde for 24 h at 4°C and cryoprotected by incubation in 25% sucrose/PBS for 48 h at 4°C. Then, brains were frozen at −80°C and 30 μm sagittal (for transgenic mouse model) or coronal (for lentiviral-induced mouse model) sections were cut using a cryostat (LEICA CM3050 S). Slices were collected and stored in 48 well plates as free-floating sections in PBS/0.05 mm sodium azide. The slices were stored at 4°C until processing.
Brightfield immunohistochemistry
Ten brain sections (of the same column) were quantified for each animal. Sections were washed with PBS and incubated with phenylhydrazine, (1:1000) prepared in PBS, for 30 min at 37°C. Then, were washed twice with PBS and incubated in 0.1% Triton X-100 and 10% normal goat serum (NGS)/PBS for 1 h at RT, followed by incubation with primary antibody overnight at 4°C under moderate agitation. The primary antibodies, rabbit anti-ubiquitin (1:500, Z0458, Dako Cytomation) and rabbit anti-DARPP-32 (1:1000, AB10518, Millipore), were prepared in 0.1% Triton X-100 and 10% NGS/PBS. Then, sections were washed three times with PBS in moderate agitation and incubated with secondary antibody, biotinylated goat-anti-rabbit IgG (H + L) (1:200, BA-1000, Vector, BA-1000), prepared in PBS with 0.1% Triton and 10% NGS, for 2 h at RT. Sections were washed three times with PBS and incubated for 30–40 min at RT with a solution of the avidin-biotin system for signal amplification, ABC Kit (Vectastain® Vector). Afterwards, three washes with PBS were performed and sections were revealed with 3,3′-diaminobenzidine until brown coloration had appeared, followed by three washes with PBS. Then, sections were mounted on gelatinized slides. After being dried, slides were dehydrated with water for 30 s, ethanol 75% for 3 min, ethanol 95% for 3 min, ethanol 100% for 3 min and xylene for 3 min. Finally, brain sections were mounted in Eukitt® mounting medium (Fluka). Images were acquired with a 5× objective on a Zeiss PALM Axiovert 200 imaging microscope and analyzed with ImageJ software (NIH, USA).
Primers used in qRT-PCR analysis
| Gene | Specie | Reference | Gene | Specie | Reference |
|---|---|---|---|---|---|
| Il6 | Ms | QT00098875; QPA | Dlg4 | Ms | M-Dlg4–1; KPP |
| Tnfa | Ms | QT00104006; QPA | Slc32a1 | Ms | M-Slc32a1–1; KPP |
| Il1b | Ms | QT01048355; QPA | Slc17a7 | Ms | QT00148841; QPA |
| Hprt | Ms | QT00166768; QPA | Gphn | Ms | M-Gphn-1; KPP |
| Mki67 | Ms | M-MKi67–1; KPP | Syp | Ms | M-Syp-1; KPP |
| Pcna | Ms | M-Pcna-1; KPP | DLG4 | Hs | H-DLG4–1; KPP |
| Msi1 | Ms | M-Msi1–1; KPP | SYP | Hs | H-SYP-1; KPP |
| Notch1 | Ms | M-Notch1–1; KPP | MSI1 | Hs | H-MSI1–1; KPP |
| Neurod1 | Ms | M-Neurod1–1; KPP | NOTCH1 | Hs | H-NOTCH1–1; KPP |
| Agrn | Ms | M-Agrn-1; KPP | MKI67 | Hs | H-MKI67–1; KPP |
| Cttn | Ms | M-Cttn-1; KPP | PCNA | Hs | H-PCNA-1; KPP |
| HPRT | Hs | QT00059066; QPA |
| Gene | Specie | Reference | Gene | Specie | Reference |
|---|---|---|---|---|---|
| Il6 | Ms | QT00098875; QPA | Dlg4 | Ms | M-Dlg4–1; KPP |
| Tnfa | Ms | QT00104006; QPA | Slc32a1 | Ms | M-Slc32a1–1; KPP |
| Il1b | Ms | QT01048355; QPA | Slc17a7 | Ms | QT00148841; QPA |
| Hprt | Ms | QT00166768; QPA | Gphn | Ms | M-Gphn-1; KPP |
| Mki67 | Ms | M-MKi67–1; KPP | Syp | Ms | M-Syp-1; KPP |
| Pcna | Ms | M-Pcna-1; KPP | DLG4 | Hs | H-DLG4–1; KPP |
| Msi1 | Ms | M-Msi1–1; KPP | SYP | Hs | H-SYP-1; KPP |
| Notch1 | Ms | M-Notch1–1; KPP | MSI1 | Hs | H-MSI1–1; KPP |
| Neurod1 | Ms | M-Neurod1–1; KPP | NOTCH1 | Hs | H-NOTCH1–1; KPP |
| Agrn | Ms | M-Agrn-1; KPP | MKI67 | Hs | H-MKI67–1; KPP |
| Cttn | Ms | M-Cttn-1; KPP | PCNA | Hs | H-PCNA-1; KPP |
| HPRT | Hs | QT00059066; QPA |
QPA = QuantiTect Primer Assays, Qiagen; KPP = KiCqStart Pre-designed Primers, Sigma-Aldrich. Ms = mouse, Hs = human.
Primers used in qRT-PCR analysis
| Gene | Specie | Reference | Gene | Specie | Reference |
|---|---|---|---|---|---|
| Il6 | Ms | QT00098875; QPA | Dlg4 | Ms | M-Dlg4–1; KPP |
| Tnfa | Ms | QT00104006; QPA | Slc32a1 | Ms | M-Slc32a1–1; KPP |
| Il1b | Ms | QT01048355; QPA | Slc17a7 | Ms | QT00148841; QPA |
| Hprt | Ms | QT00166768; QPA | Gphn | Ms | M-Gphn-1; KPP |
| Mki67 | Ms | M-MKi67–1; KPP | Syp | Ms | M-Syp-1; KPP |
| Pcna | Ms | M-Pcna-1; KPP | DLG4 | Hs | H-DLG4–1; KPP |
| Msi1 | Ms | M-Msi1–1; KPP | SYP | Hs | H-SYP-1; KPP |
| Notch1 | Ms | M-Notch1–1; KPP | MSI1 | Hs | H-MSI1–1; KPP |
| Neurod1 | Ms | M-Neurod1–1; KPP | NOTCH1 | Hs | H-NOTCH1–1; KPP |
| Agrn | Ms | M-Agrn-1; KPP | MKI67 | Hs | H-MKI67–1; KPP |
| Cttn | Ms | M-Cttn-1; KPP | PCNA | Hs | H-PCNA-1; KPP |
| HPRT | Hs | QT00059066; QPA |
| Gene | Specie | Reference | Gene | Specie | Reference |
|---|---|---|---|---|---|
| Il6 | Ms | QT00098875; QPA | Dlg4 | Ms | M-Dlg4–1; KPP |
| Tnfa | Ms | QT00104006; QPA | Slc32a1 | Ms | M-Slc32a1–1; KPP |
| Il1b | Ms | QT01048355; QPA | Slc17a7 | Ms | QT00148841; QPA |
| Hprt | Ms | QT00166768; QPA | Gphn | Ms | M-Gphn-1; KPP |
| Mki67 | Ms | M-MKi67–1; KPP | Syp | Ms | M-Syp-1; KPP |
| Pcna | Ms | M-Pcna-1; KPP | DLG4 | Hs | H-DLG4–1; KPP |
| Msi1 | Ms | M-Msi1–1; KPP | SYP | Hs | H-SYP-1; KPP |
| Notch1 | Ms | M-Notch1–1; KPP | MSI1 | Hs | H-MSI1–1; KPP |
| Neurod1 | Ms | M-Neurod1–1; KPP | NOTCH1 | Hs | H-NOTCH1–1; KPP |
| Agrn | Ms | M-Agrn-1; KPP | MKI67 | Hs | H-MKI67–1; KPP |
| Cttn | Ms | M-Cttn-1; KPP | PCNA | Hs | H-PCNA-1; KPP |
| HPRT | Hs | QT00059066; QPA |
QPA = QuantiTect Primer Assays, Qiagen; KPP = KiCqStart Pre-designed Primers, Sigma-Aldrich. Ms = mouse, Hs = human.
Fluorescence immunohistochemistry
For each animal were analyzed six (transgenic model) or four (lentiviral-induced model) brain sections (of the same column). Sections were washed three times with PBS, followed by 2 h incubation in 0.1% Triton X-100 and 10% NGS/PBS at RT. Sections were incubated with primary antibodies overnight at 4°C. Primary antibodies, rabbit anti-calbindin (1:1000, Millipore), mouse anti-HA (1:1000, InvivoGen) and rabbit anti-Ki67 antibody (1:500, ab16667, Abcam) were prepared in 2% NGS and 0.05% Triton X-100/PBS. Then, sections were washed three times with PBS and incubated in secondary antibody, goat anti-mouse Alexa Fluor-594 (1:200, A-11005, Invitrogen) and goat anti-rabbit Alexa Fluor-488 (1:200, A-11008, Invitrogen) prepared in 2% NGS without Triton, for 2 h at RT. Nuclei staining was performed with DAPI (10 min at RT). Then, sections were washed three times with PBS and mounted on gelatinized slides. After being dried, slides were mounted in Mowiol® reagent (Sigma). Widefield fluorescence images were acquired with AxioCam HR camera, with a 5–20× objective on a Zeiss PALM Axiovert 200 imaging microscope and exposure times conserved in the experiments. The images were analyzed with ImageJ software.
Quantification of Purkinje cells and mutant ataxin-3 inclusions
Quantitative analysis of calbindin-positive cells (Purkinje cells) and hemagglutinin inclusions (inclusions with mutant ataxin-3) was performed by scanning the whole cerebellum, while the mutant ataxin-3 inclusions in the lentiviral-induced model was made by scanning the whole striatum of the labeled sections with a Zeiss PALM Axiovert 200M microscope. An experienced operator performed the quantification in a blind-fashion manner using ImageJ software.
Cresyl violet staining and quantification of cerebellar cellular layers thickness
Sections were pre-mounted on gelatinized slides, dried overnight at RT and dehydrated with ethanol 96%, ethanol 100% and xylene, ethanol 100%, ethanol 96%, ethanol 70% and water, all for 5 s. Then, sections were stained with cresyl violet for 5 min, differentiated in ethanol 70% and dehydrated by passing through ethanol 96%, ethanol 100% and xylene solutions, during 5 s and mounted with Eukitt mounting medium (Fluka). Quantification of molecular and granular cerebellar layers thickness was made over six sections per animal and for each section three regions in each lobule, in fixed regions, were measured. A Zeiss PALM Axiovert 200M microscope was used for acquisition. An experienced operator performed the quantification in a blind-fashion manner using ImageJ software.
Western blot
After deep anesthesia (ketamine/xylazine), the striatum of the mice was removed and the brain punches were frozen at −80°C. Then, brain punches were processed with lysis buffer composed by 150 mm NaCl, 50 mm Tris, 5 mm EDTA, 1% Triton, 0.5% sodium deoxycholate and 0.1% SDS fresh added with protease (Complete Mini) and phosphatase inhibitors (PhosStop Easy pack, Roche), 10 ug/ml DTT (Sigma-Aldrich) and 1 mm PMSF (Sigma-Aldrich). Lysed samples were subjected to strong vortex and one or two cycles of 10 s of sonication. Concentration of protein lysates was estimated through the BCA method (Pierce™ BCA Protein Assay Kit, Thermo Scientific). Samples were prepared with the same protein amount (50 μg), and sample buffer (0.5 M Tris–HCl pH 6,8; 10% SDS; 30% glycerol; 0.6 M DTT and bromofenol blue) was added. Then, vortex and denaturation at 95°C for 5 min was performed. Samples were loaded in 4.5% stacking and 10% resolving sodium dodecyl sulphate polyacrylamide (SDS-polyacrylamide) gels. Electrophoresis was carried out at 70 V for 15 min and 90 V for ~1 h. Proteins were transferred to polyvinylidene difluoride membranes (Merck, Millipore) during 2 h at 0.75A in CAPS/0.5% methanol buffer at 4°C. Followed by blocking with 5% milk in TBS-Tween (25 mm Tris, 150 mm NaCl and 0.1% Tween-20) or 5% BSA in TBS-Tween for the phosphorylated IκBα protein. Then, membranes were incubated overnight at 4°C with the primary antibodies, mouse anti-total-IκBα (1:1000, L35A5, Cell Signaling); rabbit anti-phospho-IκBα (1:1000, 4812S, Cell Signaling); rabbit anti-DARPP-32 (1:2000, AB10518, Millipore); mouse anti-spinocerebellar ataxin type 3, clone 1H9 (1:2000, MAB5360, Millipore); and mouse anti-beta-tubulin I (1:5000, SAP.4G5, Sigma), prepared in the blocking solution. Subsequently, membranes were washed three times with 0,1% TBS-Tween and incubated with an alkaline phosphatase-conjugated secondary antibody (rabbit, 1:10000; NIF 1317, GE Healthcare or mouse 1:10000; 31328, Pierce Antibody, Thermo Scientific) with gentle agitation for 2 h, at RT. Finally, membranes were washed three times in TBS-Tween and incubated with the enhanced chemifluorescence substrate (ECF, Amersham Biosciences). Chemifluorescence signal was visualized in a Bio-Rad VersaDoc 3000 Imaging System. The analysis of band intensity was performed using Image J software (NIH, USA).
Quantitative real-time polymerase chain reaction (PCR)
Four weeks after the treatment have started, mice were sacrificed by cervical dislocation and a punch of the striatum was immediately frozen at −80°C. MJD patients’ iPSC-derived neural cultures were washed with PBS and immediately frozen at −80°C until RNA extraction. Frozen punches/cells were slightly thawed at RT and RNA extraction was performed with TRI Reagent (Sigma) and NucleoSpin RNA (Macherey-Nagel) Kit. Briefly, the RNA at the aqueous phase obtained with the TRI Reagent was cleaned up with the NucleoSpin RNA columns using DNase digestion at the spin column, according to the manufacturer’s recommendations. The RNA concentration and purity were determined with NanoDrop 2000 (Thermo Scientific). cDNA synthesis was performed with iScript cDNA Synthesis Kit (Bio-Rad) from 1 μg of total RNA, according to the manufacturer’s instructions. Real-time quantitative PCR was performed with the Sso Advanced SYBR Green Supermix Kit (Bio-Rad), as previously described (9). Threshold cycle (Ct) values were generated automatically by the StepOne Software (Applied Biosystems). In each experiment, and for all genes, a standard curve was performed and quantitative PCR efficiency was determined by the software. Additionally, no template and no reverse transcriptase controls were performed. The mRNA relative quantification with respect to control samples was determined by the Pfaffl method, taking into consideration the different amplification efficiencies of all genes. The primers used are indicated in Table 1.
Statistical analysis
All data are presented as mean ± SEM. Graphs and statistical analysis were performed in Graph Pad Prism 6 software. Statistical significance was assessed by unpaired t-test and two-way analysis of variance (ANOVA).
Supplementary Material
Supplementary Material is available at HMG online.
Conflict of Interest statement. The authors declare no conflict of interests.
Funding
European Union through the Regional Operational Program CENTRO2020; Competitiveness Factors Operational Program (COMPETE 2020); and National Funds through FCT (Foundation for Science and Technology)—projects BrainHealth2020 [CENTRO-01-0145-FEDER-000008]; ViraVector [CENTRO-01-0145-FEDER-022095]; CortaCAGs [POCI-01-0145-FEDER-016719, POCI-01-0145-FEDER-030737 and POCI-01-0145-FEDER-029716]; the National Ataxia Foundation; the French Muscular Dystrophy Association (AFM-Téléthon), Trampoline Grant #20126; SynSpread; ESMI; and ModelPolyQ projects under the EU Joint Program—Neurodegenerative Disease Research; the last two co-funded by the European Union H2020 program, GA No. 643417; the American Portuguese Biomedical Research Fund; and the Richard Chin and Lily Lock Machado–Joseph disease Research Fund.
