Abstract
Huntington's disease (HD) is an autosomal-dominant neurodegenerative disease characterized by progressive motor and cognitive impairments, with no disease-modifying therapies yet available. HD pathophysiology involves evident impairment in glutamatergic neurotransmission leading to severe striatal neurodegeneration. The vesicular glutamate transporter-3 (VGLUT3) regulates the striatal network that is centrally affected by HD. Nevertheless, current evidence on the role of VGLUT3 in HD pathophysiology is lacking. Here, we crossed mice lacking Slc17a8 gene (VGLUT3–/–) with heterozygous zQ175 knock-in mouse model of HD (zQ175:VGLUT3–/–). Longitudinal assessment of motor and cognitive functions from 6 to 15 months of age reveals that VGLUT3 deletion rescues motor coordination and short-term memory deficits in both male and female zQ175 mice. VGLUT3 deletion also rescues neuronal loss likely via the activation of Akt and ERK1/2 in the striatum of zQ175 mice of both sexes. Interestingly, the rescue in neuronal survival in zQ175:VGLUT3–/– mice is accompanied by a reduction in the number of nuclear mutant huntingtin (mHTT) aggregates with no change in the total aggregate levels or microgliosis. Collectively, these findings provide novel evidence that VGLUT3, despite its limited expression, can be a vital contributor to HD pathophysiology and a viable target for HD therapeutics.
SIGNIFICANCE STATEMENT Dysregulation of the striatal network centrally contributes to the pathophysiology of Huntington's disease (HD). The atypical vesicular glutamate transporter-3 (VGLUT3) has been shown to regulate several major striatal pathologies, such as addiction, eating disorders, or L-DOPA-induced dyskinesia. Yet, our understanding of VGLUT3's role in HD remains unclear. We report here that deletion of the Slc17a8 (Vglut3) gene rescues the deficits in both motor and cognitive functions in HD mice of both sexes. We also find that VGLUT3 deletion activates neuronal survival signaling and reduces nuclear aggregation of abnormal huntingtin proteins and striatal neuron loss in HD mice. Our novel findings highlight the vital contribution of VGLUT3 in HD pathophysiology that can be exploited for HD therapeutic management.
Introduction
Huntington's disease (HD) is an inherited, autosomal-dominant neurodegenerative disease caused by an expansion in the polyglutamine repeat region at the N-terminal domain of huntingtin protein (HTT) that leads to its aggregation in neurons (MacDonald et al., 1993). HD phenotype includes involuntary body movements, such as chorea and dystonia, loss of cognitive abilities, psychiatric disturbances, and inevitable death within 15-20 years of disease onset (Li and Li, 2004). HD pathology is characterized by progressive degeneration of some neocortical regions and importantly medium spiny neurons (MSNs) in the striatum (MacDonald et al., 1993). To date, no disease-modifying therapies are available for HD, and therapeutic options are limited to palliative management (Tabrizi et al., 2022).
Impairment in glutamate neurotransmission is considered a hallmark of HD, primarily contributing to disease phenotype (Eidelberg et al., 2011; Lewerenz and Maher, 2015; Ribeiro et al., 2017). Expression of mutant HTT (mHTT) proteins is associated with selective degeneration in the MSNs via sensitization of glutamate receptor-mediated intracellular Ca2+ release leading to neuronal excitotoxicity (Sun et al., 2001; Fan and Raymond, 2007). Specifically, metabotropic glutamate receptors 1/5 (mGluR1/5) together with NMDARs augment membrane depolarization and intracellular Ca2+ flux in MSNs while sparing other striatal interneurons in HD animal models (Calabresi et al., 1999; Tang et al., 2003; Ribeiro et al., 2017). Additionally, inhibiting glutamate release via the chronic activation of mGluR2/3 improves motor deficits and reduces mHTT pathology in zQ175 HD mice (Li et al., 2021).
Vesicular glutamate transporters subtype 3 (VGLUT3) is a member of the SLC17a family that transports glutamate into synaptic vesicles within select neuronal populations in the brain (Ueda, 1986; Bellocchio et al., 2000; Ibrahim et al., 2020). Specifically, VGLUT3 is expressed in neuronal populations coreleasing glutamate with other neurotransmitters, such as ACh, GABA, and serotonin (Fremeau et al., 2002; Gras et al., 2002; Schäfer et al., 2002). VGLUT3 is expressed in the striatum, hippocampus, raphe nuclei, cerebral cortex, and transiently in the cerebellum (Gras et al., 2002, 2005; Amilhon et al., 2010). In particular, VGLUT3 is expressed by subpopulations of interneurons scattered within the cerebral cortex (Schäfer et al., 2002; Herzog et al., 2004); as well as by subsets of GABAergic interneurons and serotoninergic fibers projecting to the hippocampus (Somogyi et al., 2004; Amilhon et al., 2010). These VGLUT3 projections can modulate cognitive and learning behaviors in animals (Fazekas et al., 2019; de Almeida et al., 2023).
Within the striatum, VGLUT3 mediates the release of glutamate primarily from tonically active cholinergic interneurons (TANs) that provides monosynaptic and disynaptic inputs onto various striatal neurons (Nelson et al., 2014; Rehani et al., 2019). Interestingly, genetic silencing of VGLUT3 signaling diminishes TAN postsynaptic responses on striatal MSNs and fast-spiking interneurons, an effect that is considered to be mediated via ionotropic glutamate receptors (Higley et al., 2011; Nelson et al., 2014). VGLUT3 ablation also alters the expression of mGluR5 and mGluR2/3 within the striatum supporting the role of VGLUT3 in finetuning general striatal glutamatergic signaling (Ibrahim et al., 2022). Furthermore, cotransmission via VGLUT3-mediated glutamate and ACh exerts a complex dual control on dopamine signaling which is a major regulator of striatal output in locomotor and reward-guided activity (Trudeau and El Mestikawy, 2018). For example, VGLUT3 ablation elicits strong protective effects against L-DOPA-induced dyskinesia in a mouse model of Parkinsonism (Gangarossa et al., 2016). Furthermore, VGLUT3 and TANs are pivotal regulators of habit formation and hence are involved in major psychiatric disorders, such as addiction or eating disorders (Sakae et al., 2015; Favier et al., 2020). Despite this intricate role in regulating striatal circuits in normal and pathologic conditions, the contribution of VGLUT3 to HD pathophysiology is still largely unexplored.
Here, we investigated the potential contribution of VGLUT3 in HD pathogenesis using VGLUT3-null (VGLUT3–/–) mice crossed with the heterozygous knock-in zQ175 mouse model of HD (zQ175:VGLUT3–/–). We report that VGLUT3 deletion rescues motor coordination and short-term memory deficits in both male and female zQ175 mice. The phenotypic rescue is accompanied by a reduction in neuronal loss in the striatum of zQ175:VGLUT3–/– mice of both sexes. Furthermore, VGLUT3 deletion lowers the number of intranuclear mHTT aggregates and further augments the activation of ERK1/2 and Akt cellular pro-survival pathways with no evident changes in microgliosis in the striatum of zQ175 mice. Together, these findings provide evidence that VGLUT3 can be a viable target for the treatment of HD symptoms.
Materials and Methods
Reagents
HRP-conjugated anti-rabbit (G-21234) and anti-mouse (G-21040) (H + L) cross-adsorbed IgG secondary antibodies were purchased from Fisher Scientific. Rabbit anti-ERK1/2 (94484) and - ionized calcium-binding adaptor molecule-1 (Iba1) (178847) antibodies were from Abcam. Rabbit anti-ERK1/2-pT202/Y204 (9101S), -Akt (9272S), ULK1-pS757 (14202S), and -Akt-pS473 (9271S) antibodies were from Cell Signaling Technology. Rabbit anti-NeuN (ABN78), and mouse anti-Huntingtin clone EM48 (MAB5374) antibodies were from Sigma-Aldrich. Western blot reagents were purchased from Bio-Rad Laboratories, and all other biochemical reagents were purchased from Sigma-Aldrich.
Animals
Experimental protocols were approved by the Institutional Animal Care Committee at the University of Ottawa (protocol #CMMe-3346) and were in compliance with the Canadian Council of Animal Care guidelines. Animals were group-housed under a constant 12 h light/dark cycle and given food and water ad libitum. Heterozygous VGLUT3 KO mice (Gras et al., 2008) (VGLUT3–/+) and heterozygous knock-in zQ175 mice (The Jackson Laboratory, stock #370476) were bred to establish littermate-control male and female mice of the following genotypes: WT (VGLUT3+/+), VGLUT3–/–, zQ175, zQ175:VGLUT3–/– mice. zQ175 mice carry humanized exon 1 of the htt gene with ∼186 ± 15 CAG repeat expansions on one of the alleles (Menalled et al., 2012). Experiments were designed to assess both motor and cognitive functions of the mice starting from 6 months of age, as HD phenotype typically manifests at 8-10 months of age (Menalled et al., 2012). Upon the conclusion of the behavioral assessment, 15-month-old mice were euthanized by live cervical dislocation followed by brain dissection, and randomization for immunostaining and immunoblotting experiments.
Behavioral analysis
Mice were habituated in the testing room for at least 30 min before the experiment, and all behavioral assessment procedures were performed during the animal's dark cycle. Behavioral studies were conducted with first the open field test, followed by the novel object recognition test, rotarod test, horizontal ladder test, and forelimb grip strength test.
Novel object recognition
Mice were placed in a white box (45 cm × 45 cm × 45 cm), and their activities were tracked using an overhead-mounted camera fed to a computer in a separate room and analyzed using the Noldus EthoVision 11 software (EthoVision XT). First, mice were allowed to explore an empty box for 5 min; and 5 min later, two identical objects were placed in the box 5 cm apart. Mice were then allowed to explore the maze for 10 min and were considered to be exploring an object if their nose was within 1 cm of the object. The following day, one object was replaced with a novel object and the experiment was repeated. Time spent by each mouse exploring the objects was recorded. The percentage recognition index was computed using this formula ((Exploration time for either familiar or novel object/Total time spent exploring both objects) × 100) (Abd-Elrahman et al., 2017, 2020). Mice were excluded from analysis if the exploration time for similar objects on day 1 of the experiment was <10 s.
Open field
Spontaneous locomotor activity was assessed in an open field box (45 cm × 45 cm × 45 cm). Mice were placed in the corner of an opaque white box illuminated at 250-300 lux; 10 min was given to the mice to explore the box, and the activity was monitored by an overhead-mounted camera fed to a computer in a separate room. Analysis was done using the Noldus EthoVision 11 software to calculate the percentage time mobile, and frequency of entries to the center (25 cm × 25 cm) during the 10 min test period (Abd-Elrahman et al., 2017).
Rotarod
Mice motor coordination was assessed using accelerating rotarod (IITC Life Science). On the first day, mice were habituated on a still rod for 3 min followed by four-trial training sessions on accelerating protocol (from 4 to 45 rpm in 300 s) with 10 min intervals between each trial. The following day, mice were tested using the same training paradigm, and the times required for the mice to fall were recorded. Average latency to fall values from the four trials of the second day were used for analysis (Li et al., 2021, 2022).
Horizontal ladder
A horizontal ladder was used to assess forelimb and hindlimb coordination. The ladder is composed of two clear Plexiglas walls (69.5 cm × 15 cm) containing 121 irregularly spaced metal rungs (0.15 cm in diameter). Mice were then trained for two trials by placing them at the start of the ladder and guiding them toward the end if necessary. Mouse nests were placed at the end of the ladder to serve as an incentive to cross the ladder. Mice were tested for three trials that were video-recorded, and the numbers of successful/missed steps were then quantified. Percent error represents the percentage of missed steps of the total number of steps required to cross the ladder (Abd-Elrahman et al., 2017; Li et al., 2021).
Forelimb grip strength
Mice were allowed to firmly grip on a rectangular grid bar of the Chatillon DFE II Grip Strength Meter (Columbus Instruments). Each mouse was then slowly pulled away at a speed of ∼2.5 cm/s horizontally from the bar until released. The value of maximal strength value was recorded. Each mouse underwent seven trials, with a period of 5-10 s interval in between each trial (Abd-Elrahman et al., 2017; Li et al., 2021).
Immunohistochemistry
Fifteen-month-old mice brains were immunostained using the peroxidase-based protocol. Formalin-fixed, paraffin-embedded brain tissues were coronally sectioned through the striatum and the cortex. Immunostaining was performed on 5 μm sections using the Leica Bond system using standard IHC protocol (Protocol F) (Li et al., 2022). A modification of Protocol F was used for rabbit antibodies, which eliminates the antibody post-primary step when used on mouse tissue. Sections stained with rabbit antibodies were pretreated using heat-mediated antigen retrieval with either EDTA buffer (pH 9.0, epitope retrieval solution 2; AR9640) or sodium citrate buffer (pH 6.0, epitope retrieval solution 1; AR9961) for 20 minutes. The sections were then incubated with neuronal nuclei (NeuN) antibody at 1:1500, or lba-1 antibody at 1:8000 for 30 minutes at room temperature and detected using an HRP-conjugated compact polymer system (Leica Biosystems, catalog #DS9800). For HTT immunostaining, brain sections were blocked with Rodent Block M (Biocare Medical) for 60 minutes and were incubated with mouse anti-huntingtin EM48 antibody at 1:100. Slides were then stained using DAB as the chromogen, counterstained with hematoxylin, mounted, and coverslipped. Sections were visualized with a Zeiss Axio Scan Z1 Slide Scanner with a 20× lens.
Six coronal sections per mouse spanning different regions of the striatum were analyzed. For each section, two 300 μm × 300 μm ROIs in the striatum were used to quantify the number of mHTT puncta, Iba1, and NeuN-labeled cells using the ImageJ Cell Counter tool (National Institutes of Health) (Schneider et al., 2012; Abd-Elrahman et al., 2017; Li et al., 2021, 2022). The estimation of nuclear HTT aggregates was done using a custom pipeline on CellProfiler software (Carpenter et al., 2006). First, the DAPI-stained nuclei were identified within the ROIs, and the background staining was masked. This was followed by counting EM48 puncta within the unmasked nuclei. The percentage of EM48-positive nuclei was calculated using the following formula: ((no. of EM48 puncta/Total number of nuclei) × 100).
Immunoblotting
Striata were dissected from mice brains and were lysed in ice-cold lysis buffer (1% Triton® X-100, 25 mm HEPES, 300 mm NaCl, 1.5 mm MgCl2, 0.2 mm EDTA) containing protease inhibitors cocktail (100 μm [4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride], 80 nm aprotinin, 2 μm leupeptin, 5 μm bestatin, 1.5 μm E-64, and 1 μm pepstatin A) and phosphatase inhibitors (500 μm Na3VO4 and 10 mm NaF). Tissue debris was pelleted and removed by centrifugation twice at 14,500 rpm at 4°C for 10 min. Supernatants were then collected, and protein concentrations were measured using the DC protein assay kit (Bio-Rad). Homogenates were diluted to a protein concentration of 1 µg/µl in a mix of lysis buffer and β-mercaptoethanol-containing 3× loading buffer (187.5 mm Tris-HCl, 30% glycerol, 6% SDS, 0.006% bromophenol blue), and then boiled for 10 min at 90°C. Aliquots containing 40 µg of total protein were resolved by electrophoresis on 7.5% SDS-polyacrylamide gels and transferred onto nitrocellulose membranes. Blots were blocked for 1 h at room temperature in TBS, pH 7.6, containing 0.05% Tween 20 (TBST) and 5% nonfat dry milk for 1 h at room temperature. Blots were incubated overnight at 4°C with primary antibodies diluted (1:1000) in TBST containing 1% nonfat dry milk. Blots were then washed 3 times with TBST the next day and incubated with anti-rabbit/mouse secondary antibodies (1:5000) diluted in TBST containing 1% nonfat dry milk for 1 h at room temperature. Blots were washed again in TBST, and bands were visualized by incubation with 1:1 Clarity Western Peroxide and Luminol/Enhancer solutions. Imaging of the blots was done using the Bio-Rad Chemidoc Gel imaging system. Band densities were quantified using Image Lab software and normalized to the loading control or total protein for the protein/phosphoprotein of interest as indicated in each figure legend.
Experimental design and statistical analysis
The mean ± SEM for each independent experiment is shown in the figure legends. GraphPad Prism software version 9 was used to analyze data for statistical significance. Data analysis was done using either unpaired Student's t test or two-way ANOVA followed by Tukey's or Dunnett's post hoc tests to determine the source of significant interactions. Details of statistical tests are indicated in the figure legends. p < 0.05 was considered statistically significant. Data analysis for behavioral experiments was conducted in groups comprising at least 10 mice per group, and immunohistochemical and immunoblotting analyses were conducted in groups of 5 mice per group. Sample sizes described above were shown to provide sufficient statistical power (0.80) to detect effects at the 0.05 level based on previous studies (Hamilton et al., 2016; Abd-Elrahman et al., 2020, 2022).
Results
VGLUT3 deletion improves memory deficits in zQ175 mice
HD is associated with cognitive impairments and dementia, which accompany chorea symptoms in patients (Paulsen, 2011). Similarly, zQ175 HD mice exhibit a marked decline in memory functions when assessed using different testing paradigms (Carty et al., 2015; Abd-Elrahman et al., 2017; Piiponniemi et al., 2018). Therefore, we investigated the impact of VGLUT3 deletion on short-term memory functions using the novel object recognition test. At 6 and 9 months of age, male and female zQ175, VGLUT3–/–, and zQ175:VGLUT3–/– mice exhibited intact memory function as depicted by their ability to discriminate between novel and familiar objects that were comparable to age- and sex-matched WT mice (Fig. 1A,B). At 12 and 15 months of age, the cognitive function of both male and female WT and VGLUT3–/– mice remained intact, but zQ175 mice of both sexes were not able to discriminate between novel and familiar objects (Fig. 1C,D). In contrast, zQ175:VGLUT3–/– mice of both sexes showed improved recognition indices for the novel objects at 12 and 15 months of age (Fig. 1C,D). Together, this suggests that the deletion of VGLUT3 can rescue novelty memory deficits in HD mice.
VGLUT3 loss improves novelty memory impairments in zQ175 Huntington's disease mice. Data are mean ± SEM of the recognition index for exploring a novel object versus a familiar object in novel object recognition test at (A) 6, (B) 9, (C) 12, and (D) 15 months of age in male and female WT and heterozygous zQ175 HD mice in the presence or absence of VGLUT3 expression (n = 10-13 mice per group). *p < 0.05 (two-way ANOVA followed by Tukey's post hoc test for multiple comparisons).
VGLUT3 deletion mitigates impairments in motor coordination and grip strength in zQ175 mice
Reports from our laboratory and others demonstrated that zQ175 mice exhibited notable deficits in limb coordination, grip strength, and locomotor activity starting from 8 to 9 months of age (Menalled et al., 2012; Abd-Elrahman et al., 2017; Deng et al., 2021; Li et al., 2021). Here, we assessed limb coordination using rotarod and horizontal ladder tests, and forelimb grip force using the grip strength test. In the rotarod test, from 6 to 15 months of age, both male and female VGLUT3–/– exhibited similar latencies to fall compared with age- and sex-matched WT mice (Fig. 2A). However, we detected an age-dependent deterioration in the rotarod performance for both male and female zQ175 mice. For male mice, there was a significant reduction in latency to fall at 12 and 15 months of age compared with WT male mice, whereas, for female zQ175 mice, significant reductions in latency to fall values were observed at 6, 9, 12, and 15 months of age compared with WT female mice (Fig. 2A). Double mutant mice (zQ175:VGLUT3–/–) demonstrated improved rotarod performance in both male and female mice at all ages to levels that were indistinguishable from age- and sex-matched WTs (Fig. 2A). We then used the horizontal ladder test to assess limb coordination by counting the number of missed steps while crossing a horizontal ladder with irregularly spaced steps. VGLUT3–/– mice of both sexes exhibited comparable percentages of missed steps at all ages compared with age- and sex-matched WT mice (Fig. 2B). In contrast, male and female zQ175 mice exhibited an age-dependent increase in the percentage of missed steps that were significantly different at 9, 12, and 15 months of age compared with age- and sex-matched WT mice (Fig. 2B). Deletion of VGLUT3 on a zQ175 background restored limb coordination of both male and female mice in the horizontal ladder task to values that were not significantly different from age- and sex-matched WT mice (Fig. 2B). Grip strength was not affected by VGLUT3 deletion in both sexes, and the values of forelimb grip force were comparable at all ages between VGLUT3–/– and age- and sex-matched WT mice (Fig. 2C). Starting from 9 months of age, both male and female zQ175 mice exhibited significantly weakened grip strength, which was notably improved at all ages by VGLUT3 deletion to values that were indistinguishable from age- and sex-matched WT mice (Fig. 2C). Collectively, our data indicate that VGLUT3 deletion does not impair motor functions in healthy mice but rescues deficits in limb coordination and grip force in zQ175 HD mice.
Loss of VGLUT3 rescues motor coordination deficits in zQ175 mice. (A) Latency to fall from accelerating rotarod, (B) percentage of errors made on the horizontal ladder, and (C) forepaw grip strength (gram-force [gf]) of male and female WT and heterozygous zQ175 HD mice in the presence or absence of VGLUT3 tested at 6, 9, 12, and 15 months of age (n = 10-13 mice per group). *p < 0.05, significant difference compared with WT mice (two-way ANOVA followed by Dunnett's test for multiple comparisons).
VGLUT3 deletion does not improve anxiogenic locomotor behavior in zQ175 mice
Hypokinesia and anxiety are among the characteristic behavioral abnormalities observed in zQ175 HD mice (Menalled et al., 2012; Abd-Elrahman et al., 2017). Thus, we assessed the impact of VGLUT3 loss on anxiogenic locomotor behaviors in both male and female WT and zQ175 in the open field test. Both the male and female groups of zQ175, VGLUT3–/–, and zQ175:VGLUT3–/– exhibited a progressive decline in the percent time mobile from 6 to 15 months of age compared with age- and sex-matched WT mice (Fig. 3A). However, female WT mice only exhibited a statistically significant difference in mobility from each of the other mouse groups at 15 months of age (Fig. 3A). VGLUT3 deletion in both sexes of zQ175 mice did not reverse this decline in the percent time mobile in any of the age groups tested (Fig. 3A). Interestingly, male VGLUT3–/–, zQ175, and zQ175:VGLUT3–/– mice displayed a persistent decrease in the frequency to enter the center of the open field from 9 to 15 months. On the other hand, in female mice, the frequency of entries to the center of the open field was only different compared with age- and sex-matched WT mice at 15 months of age (Fig. 3B). Our findings indicate that VGLUT3 ablation results in anxiety-like behavior in healthy mice; therefore, it is not capable of rescuing the anxiogenic behavior in HD mice.
VGLUT3 loss in mice impairs anxiety-mediated locomotor performance in the open field test. (A) Percent time mobile and (B) frequency of entries to the center of open field box in male and female WT and heterozygous zQ175 HD mice in the presence or absence of VGLUT3 tested at 6, 9, 12, and 15 months of age (n = 10-13 mice per group). *p < 0.05, significant difference compared with zQ175 or VGLUT3–/– mice (two-way ANOVA followed by Dunnett's test for multiple comparisons).
VGLUT3 deletion rescues striatal neuronal loss and augments ERK1/2 and Akt activation in zQ175 mice
zQ175 and Q175FDN HD mice were previously shown to exhibit atrophic changes in both striatal and cortical regions, that correlated with HD phenotypic deficits (Heikkinen et al., 2012; Southwell et al., 2016). Here, we investigated the extent of neuronal loss in our mice by immunostaining for neuronal nuclei marker, NeuN (Mullen et al., 1992). As previously described (Li et al., 2021, 2022), a significant decrease in the number of NeuN-labeled neurons was noted in the striatum of 15-month-old male and female zQ175 mice compared with age- and sex-matched WTs (Fig. 4A,B). Interestingly, VGLUT3 deletion in zQ175 mice prevented the loss of NeuN-labeled neurons in the striatum compared with age- and sex-matched zQ175 mice (Fig. 4A,B). Mutant HTT was previously reported to disrupt extracellular signal-regulated protein kinases (ERK1/2) and protein kinase B (Akt) signaling pathways, both are vital for neural cell development and survival (Apostol et al., 2006; Abd-Elrahman et al., 2017; Abd-Elrahman and Ferguson, 2019; Rai et al., 2019). Therefore, we assessed whether ERK1/2 and Akt phosphorylation were altered by VGLUT3 deletion in the striatum of 15-month-old WT and zQ175 mice. We found that genetic VGLUT3 deletion in WT mice increased the levels of pERK1/2 and pAkt in male, but not female striatum (Fig. 5A,B). In addition, pERK1/2 was increased in female, but not male zQ175 mouse striatum compared with sex-matched WT controls (Fig. 5A,B). On the other hand, pAkt levels were increased in the striata of both male and female zQ175 mice compared with sex-matched WT controls (Fig. 5A,B). The genetic deletion of VGLUT3 in zQ175 mice resulted in further increases in pERK1/2 and pAkt levels compared with sex-matched zQ175 and WT mice (Fig. 5A,B). In contrast, we found that autophagy initiation was disrupted in zQ175 and zQ175:VGLUT3–/– mice of both sexes as depicted by elevations in the inhibitory phosphorylation of autophagosome initiation marker, Unc-51-like kinase activity (ULK1) (Fig. 6A,B), (Nixon, 2013). Thus, augmented activation in ERK1/2 and Akt signaling, but not autophagy, may be associated with the prevention of neuronal loss in zQ175:VGLUT3–/– mice.
Loss of VGLUT3 mitigates striatal neuronal loss in zQ175 mice. Representative images and quantification of neuronal nuclei (NeuN) immunostaining in the striatal regions of (A) male and (B) female WT and heterozygous zQ175 HD mice in the presence or absence of VGLUT3. Quantification of NeuN-labeled cells is presented as mean ± SEM and was done in 2 striatal (300 μm × 300 μm) regions from six brain slices per mouse (n = 5 for each group). Scale bar, 50 μm. *p < 0.05, significantly different from WT VGLUT3+/+ values (two-way ANOVA followed by Tukey's post hoc test). #p < 0.05, significantly different from zQ175:VGLUT3+/+ values (two-way ANOVA followed by Tukey's post hoc test).
Loss of VGLUT3 activates ERK1/2 and Akt signaling in the zQ175 mouse striatum. Representative Western blots and quantification of ERK1/2-pT202/Y204 and Akt-pS473 levels with the corresponding total protein in the striatum of (A) male and (B) female WT and heterozygous zQ175 HD mice in the presence or absence of VGLUT3. ERK1/2-pT202/Y204 was normalized to total ERK1/2, and Akt-pS473 was normalized to total Akt (n = 5 for each group). Values are expressed as a fraction of mean VGLUT3+/+ WT values. *p < 0.05, significantly different from WT VGLUT3+/+ values (two-way ANOVA followed by Tukey's post hoc test). #p < 0.05, significantly different from zQ175:VGLUT3+/+ values (two-way ANOVA followed by Tukey's post hoc test).
Autophagy initiation is suppressed in zQ175 and zQ175:VGLUT3–/– mouse striatum. Representative Western blots and quantification of ULK1-pS757 phosphorylation levels with the corresponding loading controls in the striatum of (A) male and (B) female WT and heterozygous zQ175 HD mice in the presence or absence of VGLUT3. ULK1-pS757 was normalized vinculin (n = 5 for each group). Values are expressed as a fraction of mean VGLUT3+/+ WT values. *p < 0.05, significantly different from WT VGLUT3+/+ values (two-way ANOVA followed by Tukey's post hoc test). #p < 0.05, significantly different from zQ175:VGLUT3+/+ values (two-way ANOVA followed by Tukey's post hoc test).
VGLUT3 deletion reduces mHTT nuclear aggregation with no impact on microgliosis in zQ175 mice
Aggregation of mHTT in the nuclei and the neuropil is among the pathologic features of HD in zQ175 mice (Gutekunst et al., 1999; Smith et al., 2014). In addition, mHTT aggregation evokes inflammatory responses in the basal ganglia that could further exacerbate HD pathology (Wilton and Stevens, 2020). For instance, microgliosis is evident in both HD patients and animal models of HD (Pavese et al., 2006; Savage et al., 2020). Here, we assessed whether the improvements in motor deficits and neuronal survival following VGLUT3 deletion were associated with a reduction in mHTT aggregation or microglial activation in 15-month-old male and female zQ175 mice. As we previously described (Li et al., 2021, 2022), both male and female zQ175 mice exhibited evident accumulation of mHTT aggregates and microgliosis in the dorsal striatum (Fig. 7). Deletion of VGLUT3 did not alter the total number of mHTT aggregates in either male or female zQ175 mice (Fig. 7A,B). However, the percentage of nuclear mHTT aggregates was reduced in the striata of zQ175:VGLUT3–/– mice compared with age- and sex-matched zQ175 mice (Fig. 7A,B). We then immunostained for Iba1, a pan marker for microglial cells, to assess microgliosis (Imai et al., 1996; Li et al., 2021). Significant elevations in the numbers of Iba1-positive cells were detected in the striata of both male and female zQ175 mice compared with age- and sex-matched WTs (Fig. 7C,D). However, the number of Iba1-positive cells was not affected by VGLUT3 deletion in the striata of either male or female zQ175 mice (Fig. 7C,D). Together, our findings suggest that the neuronal rescue in zQ175:VGLUT3–/– mice is not associated with reduced microgliosis, but it can be partially attributed to a reduction in nuclear mHTT aggregate levels.
VGLUT3 deletion reduces mHTT nuclear accumulation with no impact on microgliosis in zQ175 mice. Representative images of mHTT staining using EM48 antibody in coronal whole-brain slice and magnified area of the striatum of (A) male and (B) female zQ175 and zQ175:VGLUT3–/– mice. Quantification as mean ± SEM of the total number of EM48-positive aggregates and percentage of EM48-positive nuclei was done in 2 striatal regions (300 μm × 300 μm) from six brain slices per mouse (n = 5 for each group). Scale bars: whole slice, 500 μm; magnified area of the striatum, 20 μm. Statistical significance was assessed by unpaired Student's t test. Representative images of microglial Iba1 immunostaining in striatal regions of (C) male and (D) female WT and heterozygous zQ175 HD mice in the presence or absence of VGLUT3. Iba1-positive cell quantification is presented as mean ± SEM and was done in two striatal regions (300 μm × 300 μm) from six brain slices per mouse (n = 5 for each group). Scale bar, 50 μm. *p < 0.05, significantly different from WT VGLUT3+/+ values (two-way ANOVA followed by Tukey's post hoc test).
Discussion
VGLUT3 regulation contributes to anxiety (Amilhon et al., 2010), drug addiction (Sakae et al., 2015), eating disorders (Favier et al., 2020), and motor dyskinesia in Parkinson's disease (Divito et al., 2015). Here, we provide evidence for VGLUT3's involvement in modifying HD progression in mice, where VGLUT3 deletion prevents short-term memory, motor decline, and striatal neuronal loss in both male and female zQ175 HD mice. VGLUT3 deletion augments ERK1/2 and Akt signaling and reduces nuclear mHTT localization. Together, this provides evidence that VGLUT3 signaling is a key modulator of the HD phenotype in zQ175 mice and could be a promising target for HD therapeutics.
Although HD is typically associated with motor dysfunctions, memory impairments are also evident in HD patients and animal models (Paulsen, 2011; Giralt et al., 2012; Piiponniemi et al., 2018; Li et al., 2022). Our study demonstrates that male and female zQ175 HD mice experience cognitive impairments from the age of 12 months. Recent studies have highlighted the role of VGLUT3 in hippocampal cognitive behaviors (del Pino et al., 2017; Fasano et al., 2017). Despite having higher contextual fear memory, VGLUT3–/– mice exhibit no deficits in working or spatial reference memory (Fazekas et al., 2019; de Almeida et al., 2023). Similarly, in this study, VGLUT3 deletion in WT mice of both sexes did not affect short-term memory up to 15 months of age. Additionally, VGLUT3 loss rescues short-term memory deficits in HD mice, suggesting that VGLUT3 may modulate memory circuits in HD. Recent research indicates that knocking out VGLUT3 decreases the cell-surface levels of NMDARs and mGluR5 in the hippocampus (Ibrahim et al., 2022). Since dysregulated signaling via both receptors is implicated in memory dysfunction (Wang and Reddy, 2017; Abd-Elrahman and Ferguson, 2022), these cell-surface reductions in NMDARs and mGluR5 may potentially delay the disruptions in synaptic plasticity responses in hippocampal neurons of HD mice (Parsons and Raymond, 2014). Indeed, much remains to be investigated regarding the mechanisms underlying these findings.
VGLUT3-positive neurons are organized within glutamatergic and nonglutamatergic brain networks regulating different motor functions (Vigneault et al., 2015; Sakae et al., 2019; Ibrahim et al., 2020). We show that the deletion of VGLUT3 rescues the progressive decline of motor functions in zQ175 mice of both sexes. Although it is reported that zQ175 mice may perform better on the rotarod because of reduced body mass at older ages (Heikkinen et al., 2012; Deng et al., 2021), we have not observed this phenomenon in the present or previous studies (Abd-Elrahman et al., 2017; Li et al., 2022). This motor rescue corroborates current evidence showing that VGLUT3 deletion improves deficits in locomotor functions and dopamine signaling in mouse models of Parkinson's disease (Divito et al., 2015) and L-DOPA-induced dyskinesia (Gangarossa et al., 2016). Interestingly, deficits in dopaminergic and cholinergic neurotransmission are characteristic pathogenic features of HD that potentially underlie dyskinetic movements and cognitive decline in both HD patients and zQ175 mice (Bird, 1980; Richfield et al., 1991; Rothe et al., 2015; D'Souza and Waldvogel, 2016). Therefore, loss of VGLUT3 potentially mitigates the dopaminergic/cholinergic imbalance, leading to an overall enhancement of ambulatory activity in HD mice (Gras et al., 2008; Divito et al., 2015). This is evidenced by VGLUT3's ability to augment striatal dopamine release and alter dopamine D1 receptor densities in both the ventral striatum and the cortex (Divito et al., 2015; Sakae et al., 2015; Favier et al., 2020; Ibrahim et al., 2022). Furthermore, VGLUT3-expressing interneurons in the cortex can influence the activities of pyramidal neurons projecting to the striatum (Hioki et al., 2004; Somogyi et al., 2004). In early-stage HD, GABA-mediated inhibition of cortical neurons is impaired in patients (Schippling et al., 2009; Philpott et al., 2016). Thus, VGLUT3 ablation may have reduced striatal neuronal overactivation by augmenting GABAergic currents onto the pyramidal neurons of HD mice (Heimer et al., 2008; Fasano et al., 2017). Further investigation is required to confirm this hypothesis.
Similar to zQ175 mice, VGLUT3–/– and zQ175:VGLUT3–/– mice have reduced overall locomotor activity and anxiety-like behavior in the open field test. This may be attributed to enhanced neophobic responses displayed by VGLUT3–/– mice when placed in a new environment (Amilhon et al., 2010). This also reflects on the hypolocomotive behavior in VGLUT3–/– mice when measured during shorter periods of assessment (5-10 min), as opposed to relative hyperactivity when assessed over longer periods (>15 min) (Gras et al., 2008; Balázsfi et al., 2018). Overall, our observations provide novel evidence that VGLUT3 inhibition represents a promising therapeutic approach to slow and/or prevent HD motor phenotype in a sex-independent manner.
Activation of ERK1/2 and Akt is linked to glutamatergic transmission, and their disruption contributes to HD pathogenesis (Ribeiro et al., 2017). Deletion of VGLUT3 differentially alters glutamate receptor densities and transmission across the brain (Higley et al., 2011; Nelson et al., 2014; Ibrahim et al., 2022). For instance, knocking out VGLUT3 increases the densities of cell-surface mGluR5, total mGluR2/3, and AMPARs in the striatum of male mice (Ibrahim et al., 2022). Such increases in receptor densities may have resulted in the enhanced ERK1/2 and Akt phosphorylation seen in VGLUT3–/– male striatum (Perkinton et al., 1999; Mao et al., 2005; Di Liberto et al., 2011; Xi et al., 2011). However, female VGLUT3–/– mice elicit no changes in ERK1/2 and Akt signaling, and this discrepancy coincides with the sex specificity in mGluR5 and mGluR2/3 signaling profiles (Li et al., 2021, 2022; Abd-Elrahman and Ferguson, 2022), suggesting that mGluRs potentially mediate the observed sex-dependent changes in ERK1/2 and Akt activation in WT mice.
In zQ175 mice, however, ERK1/2 activation can be associated with increased caspase-3 activation and neuronal apoptosis (Abd-Elrahman et al., 2017). Likewise, increased Akt/mTOR activation is linked to suppressed autophagic machinery in zQ175 mice (Abd-Elrahman et al., 2017; Abd-Elrahman and Ferguson, 2019). Here, both ERK1/2 and Akt activities are upregulated in the striatum of zQ175 mice and are, unexpectedly, further augmented in zQ175:VGLUT3–/– striata. While ERK1/2 can engage pro-apoptotic pathways (Lu and Xu, 2006), we suspect that the augmented ERK1/2 phosphorylation in zQ175:VGLUT3–/– mice preferentially engages neuroprotective mechanisms. Indeed, we find that the rescue in neuronal loss in zQ175:VGLUT3–/– striatum was coupled with enhanced Akt activation, which is known to engage the pro-survival signaling arm of the ERK1/2 pathway (Mirza et al., 2004; Mendoza et al., 2011). Additionally, the increased ERK1/2 phosphorylation in VGLUT3–/– mice had no impact on striatal neurons' viability. Finally, such activation in Akt/mTOR signaling may play a key role in the improved motor functions that we observed in zQ175:VGLUT3–/– mice. This is supported by previous reports showing that suppression of striatal mTOR activity compromises motor learning skills in mice (Bergeron et al., 2014); and the unilateral injection of a constitutively active Akt downstream target, Ras homolog (Rheb), reverses striatal atrophy and improves mHTT-associated metabolic phenotypes in HD mice (Lee et al., 2015). Thus, our findings suggest that ERK1/2 and Akt signaling could mediate, at least partially, the rescue in striatal neuronal loss and motor deficits in zQ175:VGLUT3–/– mice. However, future experiments are needed to address the involvement of other signaling pathways in the observed rescue in these mice.
Despite the neuronal rescue in zQ175:VGLUT3–/– mice, the total mHTT burden is similar to that in zQ175 mice. This can be because of the suppression of ULK1-dependent autophagy in both strains, which is negatively regulated by Akt/mTOR and ERK1/2 signaling (Mendoza et al., 2011; Perluigi et al., 2015). One enticing observation was the reduction in nuclear mHTT levels, which potentially accounted for the neuronal rescue in zQ175:VGLUT3–/– mice. Cytotoxicity and accelerated HD neuropathology have been directly associated with mHTT translocation to the nuclei (Saudou et al., 1998; Yang, 2002; Gu et al., 2015). Interestingly, this reduction in nuclear mHTT could also be linked to augmented Akt activation in zQ175:VGLUT3–/– mice since Akt can phosphorylate HTT and reduces its nuclear fragments, ultimately protecting the cells from apoptosis (Humbert et al., 2002; Warby et al., 2009). Furthermore, phosphorylation of HTT can reduce NMDA-mediated excitotoxic insults in HD MSNs (Fan and Raymond, 2007; Metzler et al., 2010). Therefore, the resistance to excitotoxicity in zQ175:VGLUT3–/– mice may be because of decreased nuclear mHTT levels and lack of VGLUT3 inputs (Heikkinen et al., 2012; Goodliffe et al., 2018). This indeed warrants further investigations. On the other hand, microgliosis in the striatum of HD mice was not modified by VGLUT3 loss, and this is consistent with the lack of VGLUT3 expression in microglia and astrocytes (Schäfer et al., 2002; Li et al., 2013). Collectively, our findings suggest that reduced nuclear mHTT is a potential mediator for increased neuronal survival in zQ175:VGLUT3–/– mice.
In conclusion, we provide evidence that loss of VGLUT3 transmission rescues long-term motor and novelty memory deficits in both male and female zQ175 mice. Moreover, we report an evident rescue in striatal neuronal loss and reduction in mHTT nuclear burden in VGLUT3-null HD mice. Although toolkits for pharmacologic suppression of VGLUT3 are still awaited, our findings clearly highlight the importance of exploiting this pathway for HD therapeutic management.
Footnotes
This work was supported by Canadian Institutes of Health Research Grants PJT-148656, PJT-165967, and PJT-178060; and the Krembil Foundation to S.S.G.F. S.S.G.F. holds a Distinguished Chair in Neurodegeneration and held Tier I Canada Research Chair in Brain and Mind. We thank Tash-Lynn Colson and Shaunessy Hutchinson for breeding the animals and providing technical assistance; and Animal Behavior and Physiology Core and Louise Pelletier Histology Core facilities at the University of Ottawa for their assistance.
The authors declare no competing financial interests.
- Correspondence should be addressed to Stephen S.G. Ferguson at sferguso{at}uottawa.ca
