A partial reduction of VDAC1 enhances mitophagy, autophagy, synaptic activities in a transgenic Tau mouse model

Abstract Alzheimer's disease (AD) is the most common cause of mental dementia in the aged population. AD is characterized by the progressive decline of memory and multiple cognitive functions, and changes in behavior and personality. Recent research has revealed age‐dependent increased levels of VDAC1 in postmortem AD brains and cerebral cortices of APP, APPxPS1, and 3xAD.Tg mice. Further, we found abnormal interaction between VDAC1 and P‐Tau in the AD brains, leading to mitochondrial structural and functional defects. Our current study aimed to understand the impact of a partial reduction of voltage‐dependent anion channel 1 (VDAC1) protein on mitophagy/autophagy, mitochondrial and synaptic activities, and behavior changes in transgenic TAU mice in Alzheimer's disease. To determine if a partial reduction of VDAC1 reduces mitochondrial and synaptic toxicities in transgenic Tau (P301L) mice, we crossed heterozygote VDAC1 knockout (VDAC1+/−) mice with TAU mice and generated double mutant (VDAC1+/−/TAU) mice. We assessed phenotypic behavior, protein levels of mitophagy, autophagy, synaptic, other key proteins, mitochondrial morphology, and dendritic spines in TAU mice relative to double mutant mice. Partial reduction of VDAC1 rescued the TAU‐induced behavioral impairments such as motor coordination and exploratory behavioral changes, and learning and spatial memory impairments in VDAC1+/−/TAU mice. Protein levels of mitophagy, autophagy, and synaptic proteins were significantly increased in double mutant mice compared with TAU mice. In addition, dendritic spines were significantly increased; the mitochondrial number was significantly reduced, and mitochondrial length was increased in double mutant mice. Based on these observations, we conclude that reduced VDAC1 is beneficial in symptomatic‐transgenic TAU mice.


| INTRODUC TI ON
Alzheimer's disease (AD) is a late-onset, neurodegenerative disease characterized by a progressive decline of memory and cognitive functions and changes in behavior and personality. AD results in the irreversible loss of neurons, particularly in the learning and memory regions of the brain (LaFerla et al., 2007;Selkoe, 2001). AD currently affects over 6 million people in the USA, with estimates suggesting this number will nearly triple by 2060 (Matthews et al., 2019;Vijayan et al., 2017;Vijayan & Reddy, 2016). Intracellular phosphorylated Tau (P-Tau) and neurofibrillary tangles (NFTs) are more definitive pathologic features and are tightly linked to cognitive decline in AD patients (Caspersen et al., 2005;Reddy & Beal, 2008;Selkoe, 2001).
Extracellular neuritic plaques are deposits of differently sized small peptides called β-amyloid (Aβ) derived via sequential proteolytic cleavages of the β-amyloid precursor protein (APP). Accumulation of Aβ has been demonstrated to occur within neurons with AD pathogenesis Ye et al., 2017;Ye & Cai, 2014).
Tau is a major microtubule-associated protein that plays a large role in the outgrowth of neuronal processes and the development of neuronal polarity (Reddy, 2011a). Tau is abundantly present in the central nervous system and is predominantly expressed in neuronal axons. The phosphorylation of Tau regulates microtubule binding and assembly (Wang & Liu, 2008). In contrast, pathological Tau becomes hyperphosphorylated, which destabilizes microtubules by decreased binding to microtubules, resulting in the aggregation of hyperphosphorylated Tau (Brandt et al., 2005;Pradeepkiran et al., 2019;Reddy, 2011a;Reddy, 2011b). Tau can also self-aggregate into oligomers and more extensive inclusions in neurons, known as neurofibrillary tangles (Palop & Mucke, 2009). Mitochondrial dysfunction has been strongly associated with Tau pathology in AD in recent years. Overexpression of hyperphosphorylated and aggregated Tau damages the axonal transport, leading to abnormal mitochondrial distribution Cheng & Bai, 2018;Wang et al., 2009). Synapses are exposed to disease-modified protein Tau, which may cause the loss of synaptic contacts in AD neurons Du et al., 2010;Jadhav et al., 2015).
Mitochondria are intracellular organelles with key roles covering cellular metabolism and are the primary source of adenosine triphosphate (ATP) generated via oxidative phosphorylation (Camara et al., 2010;Perez Ortiz & Swerdlow, 2019). There is extensive literature supporting the role of mitochondrial dysfunction and oxidative damage in the pathogenesis of AD (Gowda et al., 2021;Han, Jeong, Sheshadri, Su, & Cai, 2020;Lin & Beal, 2006;Reddy & Beal, 2008;Swerdlow, 2018). Mitochondrial dysfunction also plays a key role in other neurodegenerative diseases such as Parkinson's disease, multiple sclerosis, Huntington's disease, and amyotrophic lateral sclerosis (ALS) (Camara et al., 2017;. Mitochondria are comprised of two bio-lipid membranes: the inner membrane and the outer membrane. The inner membrane covers the mitochondrial matrix and electron transport chain, and the outer membrane is highly porous, allowing low-molecular-weight substances between the cytosol and the intermembrane space Reddy, 2008). The outer membrane consists of a voltage-dependent anion channel (VDAC) that provides the major pathway for transmembrane fluxes of ions and metabolites across the outer mitochondrial membrane (Colombini, 2012;Reddy, 2013).
The functions of VDACs are several-fold, including maintaining synaptic plasticity, maintaining mitochondrial shape, regulating hexokinases (HKs) and mitochondrial interactions, and regulating apoptosis signaling (Kerner et al., 2012). VDAC1 is a crucial protein in mitochondria-mediated apoptosis. VDAC1 forms oligomers and promotes apoptosis when the protein is overexpressed, even without any apoptotic stimulus (Reddy, 2013;Shoshan-Barmatz et al., 2018).
In addition, several recent studies revealed that VDAC proteins and their binding partners are modified post-translationally due to VDAC hyperphosphorylation and are involved in the dysfunction of VDAC (Lemasters et al., 2012). However, the causal factors of VDAC1 phosphorylation in AD are not completely understood.

VDAC1-phosphorylated Tau complexes block mitochondrial
pores, interrupt the flux of metabolites between mitochondrial membranes and cytoplasm, and impair the gating of the VDAC channel, leading to mitochondrial dysfunction and neuronal damage in AD (Reddy, 2013). Previously, we studied tissues from human postmortem AD brains and AD mouse brains and found an abnormal interaction between phosphorylated Tau and VDAC1, suggesting a direct link between VDAC1 and phosphorylated Tau, mitochondrial dysfunction, and neuronal damage in AD .
GSK3β phosphorylates VDAC1 on threonine 51, resulting in the detachment of hexokinase from VDAC1 (Pastorino et al., 2005). The binding of hexokinases with VDAC1 allows the direct access of hexokinases to mitochondrial ATP in the glycolytic pathway. Also, hexokinase inhibits apoptosis by binding to VDAC and preventing the release of cytochrome c (Abu-Hamad et al., 2008;Azoulay-Zohar et al., 2004).

| Reduced VDAC1 ameliorates TAU-induced behavioral deficits
Studies of the relationship between behavioral impairments and mice that overexpress human mutant TAU (P301L) suggest that mutant tau promotes the formation of phosphorylated Tau and neurofibrillary tangles, mediating age-dependent adverse effects on memory (Lewis et al., 2000). To determine whether the reduced expression of VDAC1 ameliorates behavioral impairments in double mutant mice, using 4 widely used behavioral tests, namely the rotarod, open field, Y-maze, and Morris Water Maze (MWM), we assessed motor coordination, locomotion, exploration abilities, spatial learning, and memory abilities. The behavior study scheme is illustrated in Figures 1 and 2. We used 6-month-old WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice for the above said behavioral tests.
On an accelerating rotarod test, TAU mice spent significantly less time than that WT mice (p < 0.0001, Figure 1a,b), validating the TAU-induced motor deficits in mice. Relative to VDAC1 +/− mice, TAU mice spent less time on the rod and reached a lower maximum rate (p < 0.0001, Figure 1a,b), suggesting impairments in motor learning and coordination. Strikingly, the average latency to fall was increased in the VDAC1 +/− /TAU compared with that of TAU mice (p < 0.0001, Figure 1a,b).
Similar to the lack of motor coordination observed in rotarod, the total distance traveled (p = 0.0003, Figure 1c Figure 2g) compared with TAU mice. In VDAC1 +/− /TAU mice (p < 0.0001), the mean latency time for finding a platform was significantly reduced compared with TAU mice. VDAC1 +/− /TAU mice spent more time in the NW quadrant than TAU mice (p < 0.0001).
Since the VDAC1 +/− /TAU mice spent more time on the rotarod test, traveled more distance in the open field test, increased the percentage of spontaneous alteration in the Y-maze test, and spent more time on the NW quadrant of the Morris Water Maze, we, therefore, can infer that reduced VDAC1 expression rescued the TAU-induced motor, locomotion, and spatial memory impairments in VDAC1 +/− /TAU mice.

| Reduced expression of VDAC1 induces mitophagy and autophagy in VDAC1 +/− /TAU mice
Recent studies on VDAC1 revealed that age and P-Tau induced increased synaptic and mitochondrial damage, particularly abnormal regulation of mitophagy and autophagy in the disease process Morton et al., 2021;Reddy & Oliver, 2019). Currently, it is unclear how reduced VDAC1 protects against defective autophagy and mitophagy.
To address these issues, we crossed VDAC1 heterozygote knockout (VDAC1 +/− ) mice with transgenic TAU (P301L strain) mice and generated double mutant (VDAC1 +/− / TAU) mice and studied the protective effects of a partial reduction of VDAC1 on mitophagy and autophagy. We performed i) immunoblotting analysis of mitophagy and autophagy proteins from cortical tissues and ii) immunofluorescence analysis in the hippocampal sections from 6-month-old WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice. As a result, significantly decreased levels of mitophagy proteins (PARKIN and PINK1) and increased levels of BNIP3L were found in TAU mice (Figure 3a-d,

F I G U R E 1
Rotarod and open field, behavioral assessment of 6-month-old WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice. Rotarod test: (a) latency to fall of various cohorts, namely WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice as assessed by the rotarod test. VDAC1 +/− /TAU mice improve motor learning and coordination. (b) TAU mice spent less time on the rod (lower latency to fall), indicating impaired motor learning and coordination compared to the WT mice (****p < 0.0001). Open field test: TAU mice exhibited reduced locomotor and exploratory activity than VDAC1 +/− /TAU mice (*p < 0.05), as evidenced by reduced total distance traveled and average speed. (c) Quantification of total distance traveled, (d) average speed, (e) the number of center entries, and (f) time spent in the center area by all indicated cohorts assessed by open field test. (g) Shows representative trajectory maps (time spent in the center) of all mentioned cohorts as analyzed by an open field test. N = 10 per group. Bars represent mean ± SEM. ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, one-way ANOVA followed by Turkey's test for multiple comparisons Figure S1). In addition, decreased levels of autophagy proteins LC3BI, ATG5, Beclin1, and P62 (Figure 4a-d, Figure S2) were found in TAU mice compared with WT mice. At the same time, PARKIN, PINK1, and autophagy proteins were significantly increased in VDAC1 +/− , VDAC1 +/− /TAU compared with TAU mice. These results suggested that partial reduction of VDAC1 expression induces mitophagy and autophagy in VDAC1 +/− /TAU mice.

| VDAC1 interaction with phosphorylated Tau, HK1, and HK2
VDAC1/ Hexokinase interactions link glycolysis and oxidative phosphorylation (Rodrigues-Ferreira et al., 2012). HK1 binding of VDAC1 is thought to increase the catalytic efficiency of both processes by facilitating mitochondrial ATP release from VDAC for glucose phosphorylation and by channeling ADP into mitochondria for oxidative phosphorylation (Jackson et al., 2015).
VDAC1 interacts with phosphorylated Tau, leading to blocking the pores of mitochondria and mitochondrial transport in AD neurons . In the present study, we determined whether phosphorylated Tau interacts with VDAC1, we conducted a double-labeling analysis of VDAC1 and phosphorylated tau, VDAC1, and HK1, VDAC1, and HK2 using hippocampal sections from the brains of WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice.

| Impact of VDAC1 +/− on synapse numbers in the hippocampal and cortical tissues
It is well-established that spine density is critical for synaptic function and cognitive behavior in AD patients and AD mice . Therefore, we also examined the impact of VDAC1 +/− on synapse organization and numbers in both hippocampal and cortical tissues of WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice.
The synapse location and organization in the red arrow showed the synaptic cleft. As shown in Figure 7f

| Reduced expression of VDAC1 increases the dendritic spine density
We quantified dendritic length and the number of spines using Golgi-cox staining in the hippocampus and cortex of 6-month-old WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice to assess the effects of VDAC1 +/− on dendritic length and spines.

| DISCUSS ION
Mitochondrial dysfunction is a common pathological feature and contributes to neurodegeneration in AD (Reddy & Beal, 2008).
Recent reports suggested that mitochondrial fission/fusion, biogenesis, mitophagy, and autophagy are altered in postmortem AD brains and both in vitro and in vivo models of disease Fang et al., 2019;Kandimalla et al., 2016;Kandimalla et al., 2018;Kerr et al., 2017;Manczak et al., 2011;Reddy, 2014;Reddy et al., 2018;Santos et al., 2010;Su et al., 2010;Swerdlow et al., 2014;Wang et al., 2009). Other researchers and we previously reported that Aβ and P-Tau interact with many mitochondrial proteins (DRP1 and VDAC1), leading to mitochondrial dysfunction and the depletion of major mitophagy and autophagy proteins (Hirai et al., 2001;Ishihara et al., 2009;Kshirsagar et al., 2021;Manczak et al., 2010;Reddy et al., 2012;Reddy et al., 2018;Shoshan-Barmatz et al., 2018). In addition, our laboratory previously reported that reduced levels of VDAC1 may lead to decreased interaction between VDAC1 and APP, Aβ, and phosphorylated Tau and may allow mitochondrial pore opening and pore closure, ultimately leading to normal mitochondrial function and synaptic ATP and boosting synaptic and cognitive functions in AD . We also found that VDAC1 +/− mice showed improved mitochondrial function and synaptic activity and reduced expressions of several AD-related genes compared with VDAC1 +/+ mice . Further, we reported that reducing the human VDAC1 gene in an in vitro condition might enhance synaptic activity, improve mitochondrial maintenance and function, and protect against AD-related genes' toxicities ). In the current study, we have provided in vivo evidence that partial reduction of VDAC1 mediated mitophagy, autophagy, synaptic, reduced P-Tau pathology, and lessened memory impairment and anxiety symptoms in a murine model.
Our overall analysis of 6-month-old WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice revealed that reduced protein levels of all mitophagy, autophagy, synaptic, and other key proteins (HK1, HK2, AKT) and increased other key proteins (GSK3A, GSK3β, ANT1, pS422, VDAC1) in TAU mice compared to WT mice. In contrast, we observed the reverse trend in VDAC1 +/− and VDAC1 +/− /TAU mice compared with TAU (P301L) mice. Current findings of increased protein levels of mitochondrial and synaptic genes agree with our earlier observations .
Cognitive impairment and anxiety are extensively reported in most AD patients (Goncalves et al., 2020;Teri et al., 1999) and are associated with increased conversion rates from MCI to AD (Mah et al., 2015). Here, we show that 6-month-old TAU (P301L) mice in exploring the field than TAU mice. Similar to learning and motor coordination activities, VDAC1 +/− and VDAC1 +/− /TAU mice did well on spatial recognition and working memory, as evidenced by the Ymaze test. In addition, the percentage of spontaneous alteration was significantly higher in the VDAC1 +/− and VDAC1 +/− /TAU mice than in TAU mice. Most importantly, hippocampal-dependent learning and memory were increased dramatically in VDAC1 +/− and VDAC1 +/− / TAU mice relative to TAU mice. This data strongly suggests that reduced expression of VDAC1 has improved hippocampal-dependent learning and memory.
Mitophagy is the selective elimination of damaged mitochondria and is thus essential for mitochondrial quality control (Ashrafi & Schwarz, 2013). A recent paper has shown that VDAC1 is indeed one of the regulators of mitophagy (Ordureau et al., 2018). But others are still debating whether VDAC1 is a critical component for the PINK1-PARKIN pathway or VDAC1 is irrelevant to mitophagy (Ham et al., 2020). Here, we identify PINK1, PARKIN, p62, and the mito- Significantly decreased number of mitochondria in hippocampi (****p < 0.0001) and cortex (****p < 0.0001) of VDAC1 +/− /TAU mice relative to TAU mice, the mitochondrial length is significantly increased in hippocampal (****p < 0.0001) and cerebral cortical tissues (****p < 0.0001) in VDAC1 +/− / TAU mice. (f) Synaptic densities are sharply defined and contain electron-dense materials uniformly distributed in all indicated cohorts of the hippocampus and cerebral cortex. The arrowheads show the bowl-shaped structure of synapses and synaptic mitochondria with normal structure. Magnification ×22,000. (g) Synapse number in the hippocampus. (h) Synapse number in the cerebral cortex. The synapse numbers were found to be significantly increased in the hippocampi (****p < 0.0001) and cerebral cortical tissues (****p < 0.0001) of VDAC1 +/− /TAU mice relative to TAU mice. N = 5 per group. Results were expressed as mean ± SEM. ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, one-way ANOVA followed by Turkey's test for multiple comparisons (scale bar = 600 nm) TAU mice. Furthermore, we observed a mechanistic link between reduced VDAC1-dependent enhanced mitophagy and autophagy in AD for the first time.
Hexokinase is the key enzyme in glucose metabolism. The decreased glucose metabolism in TAU (P301L) mice reflects the abnormal expression and distribution of HK (Chiara et al., 2008;Pastorino & Hoek, 2008). HK1 and HK2 are mitochondrial hexokinase isotypes because they participate in glucose metabolism by binding to mitochondria. HK1 and HK2 protein levels were reduced in 3XTg AD mice (Han et al., 2021). Hexokinase isoforms bind to mitochondrial outer membranes in large part by interacting with the outer membrane VDAC1 (Pastorino & Hoek, 2008). GSK3β phosphorylates VDAC1 on threonine 51, resulting in the detachment of hexokinase from VDAC1. Given the association of GSK3 with phosphorylation in AD, VDAC1 is phosphorylated on the putative GSK3β epitope in AD, leading to the inability of hexokinases to interact with VDAC1, resulting in the dissociation of VDAC1 from hexokinases (Pastorino et al., 2005).
Abnormalities in mitochondrial pore opening and closure may lead to defects in oxidative phosphorylation, mitochondrial dysfunction, and ultimately cell death . Several key mitochondrial proteins, including outer membrane protein VDAC1, inner membrane protein ANT, and matrix protein Cyclophilin D (CypD), are involved in mitochondrial pore opening and pore closure . We previously reported that Hexokinases 1 and 2 were significantly upregulated in the VDAC +/− mice . Further, free radical production and lipid peroxidation levels were reduced in the VDAC1 +/− mice, and cytochrome oxidase activity and ATP levels were elevated, indicating an enhanced mitochondrial function in the VDAC1 +/− mice .
Our present study found increased protein expression of HK1, HK2, and AKT and decreased GSK3A, GSK3β, and ANT1 in VDAC1 +/− heterozygote knockout and VDAC1 +/− /TAU double mutant mice. We also found reduced phosphorylated Tau (pS422) protein levels in 6-month-old VDAC1 +/− and age-matched double mutant VDAC1 +/− / TAU mice relative to TAU mice. Our study understands how partial reduction of VDAC1 impacts HK1 and HK2 and glycolic pathways in TAU mice in disease progression. Further studies are still needed to understand the mechanistic links.
We previously studied the RNA silencing of VDAC1 and assessed mitochondrial function in AD pathogenesis .
We reported increased mRNA expression of synaptic function and mitochondrial fission genes and reduced levels of mitochondrial fusion genes in RNA-silenced SHSY5Y cells for VDAC1 gene. In addition, RNA-silenced VDAC1 gene in SHSY5Y cells showed reduced H 2 O 2 production, lipid peroxidation, and fission-linked guanosine triphosphate (GTPase) activity, and increased cytochrome oxidase activity and ATP production . In the present study, increased expression of synaptic genes suggests that reduced VDAC1 is beneficial in the presence of Tau in double mutant (VDAC1 +/− /TAU) mice. These observations demonstrated that phosphorylated Tau interaction with VDAC1 increases mitochondrial fragmentation, ultimately leading to mitochondrial dysfunction and neuronal damage. Findings from our current study support our previous study, in which increased VDAC1 levels correlated with reduced synaptic and mitochondrial activity at different stages in disease progression . Taken together, these findings suggest that partial reduction of VDAC1 may be beneficial to the maintenance of mitophagy, autophagy, and synaptic activity. (h) The number of dendritic spines in the hippocampus. (i) Dendritic length in the hippocampus. Significantly increased dendritic number (****p < 0.0001) and the length of dendritic spines (****p < 0.0001) in VDAC1 +/− /TAU mice relative to TAU mice. N = 5 per group. Results were expressed as mean ± SEM. ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, one-way ANOVA followed by Turkey's test for multiple comparisons. Scale bars 1000 μm in (a, b & e), 400 μm in (c), 200 μm in (f) and 1 μm, 2 μm, 5 μm, 10 μm in (d & g) Dysfunction of mitochondria is correlated with disease progression in neurodegenerative diseases and is suggested to contribute to excessive neuron loss in AD (Chan, 2006;Cipolat et al., 2006;Wu et al., 2014). We found significant differences in the mitochondrial length and number in the hippocampal and cortical tissues of 6-month-old VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice relative to age-matched WT mice. There were increased numbers of large, abnormal-shaped mitochondria in the TAU mice. Furthermore, we also found disrupted cristae in the TAU mice. In addition, mitochondrial length was drastically decreased in the TAU mice compared with WT mice in both the hippocampus and cortex. Oxygen tension, oxidative stress, and autophagic activation are the factors that can modulate the mitochondrial shape (Gomes et al., 2011). On the contrary, increased mitochondrial number and decreased length in TAU (P301L) mice may be due to excessive mitochondrial fragmentation or ineffective degradation of damaged mitochondria after fission. A similar observation was reported in an earlier study of P301L mice (Kandimalla et al., 2016;Kandimalla et al., 2018). In VDAC1 +/− and Dendritic spines bear a strong potential for morphological plasticity, thus enabling neurons to modify their synaptic interconnections, the correlates of learning and memory (Hoffmann et al., 2013).
Quantitative dendritic spine analysis is crucial in AD research as there is principal evidence that the disease changes the number, structure, and function of these postsynaptic sites of excitatory synapses (John & Reddy, 2021;Kandimalla et al., 2018;Kartalou et al., 2020;Reddy & Beal, 2008). Our earlier findings stated that a reduced number of dendritic spines and synaptic proteins are extensively reported in mouse models of AD (Hegde et al., 2019;Kandimalla et al., 2016) and postmortem AD brains (Reddy et al., 2005;Reddy & Beal, 2008).
These observations prompted an investigation of VDAC1 +/− 's impact on dendritic spines in TAU (P301L) mice. As described above, we quantified dendritic length and a number of spines using Golgi-Cox staining in the hippocampus and cortex of 6-month WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice. We observed significantly increased dendritic length and number in VDAC1 +/− and VDAC1 +/− /TAU mice relative to TAU mice. Our study observations strongly indicate that the reduced expression of VDAC1 enhances both dendritic length and the number of dendritic spines in VDAC1 +/− /TAU mice. Our dendritic number and length observations positively correlated with improved behavior in VDAC1 +/− and VDAC1 +/− /TAU mice.
In conclusion, we have demonstrated that the partial reduction of VDAC1 improves (i) spatial learning and memory, (ii) mitophagy, autophagy, synaptic, and other key proteins, (iii) mitochondrial length and number, and (iv) synapse and dendritic spine morphology in VDAC1 +/− , and VDAC1 +/− /TAU mice. Our findings further suggest that the impaired mitophagy in TAU contributes to an accumulation of defective mitochondria with abnormal morphology, causing mitochondrial dysfunction and a deficiency in cellular energy supply.
Further, our current study observations provided protective effects of reduced VDAC1 against P-Tau-induced mitochondrial and synaptic toxicities in TAU (P301L) mice and provided new evidence to develop VDAC1 therapeutic strategies for AD. Ours is the first genetic crossing study to report the beneficial effects of reduced VDAC1 in AD.

| TEM of brain mitochondria
To determine the mitochondrial number and size, we performed transmission electron microscopy in hippocampal and cortical sections of 6-month-old WT, VDAC1 +/− , TAU, and VDAC1 +/− /TAU mice.
Animals were perfused using the standard method, and the brains were removed from the mice as described earlier (  Vijayan, Bose, & Reddy, 2021b). Briefly, dendrites within the CA1 subregion of the hippocampus and cerebral cortex were imaged using a 4×, 10×, 20×, and 100× objective using EVOS microscope-AMG (therm ofish er.com) and Olympus1X83. Approximately 20 neurons were randomly selected from each group and quantified with a double-blind, controlled design. In addition, ImageJ and Image-Pro Plus were used to evaluate the number of spines and the total dendritic length.

| Statistical analysis
Data were represented as mean ± standard error of the mean (SEM). Conclusions were drawn based on statistical analyses using GraphPad™ PRISM software (version 9.0; GraphPad Software). The one-way ANOVA was performed using Tukey's test for multiple comparisons. Group comparisons were considered significant when the p-value was less than 0.05 (p < 0.05).
For all other procedures, see Appendix S1. P.H.R is responsible for funding acquisition.

ACK N OWLED G M ENTS
The research presented in this article was supported by the National Institutes of Health (NIH) grants AG042178, AG047812, NS105473, AG060767, AG069333, AG066347, and R41 AG060836.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.