Upregulation of Wnt2b exerts neuroprotective effect by alleviating mitochondrial dysfunction in Alzheimer's disease

Abstract Aims This study investigated the relationship between plasma Wnt2b levels and Alzheimer's disease (AD), and explored the effect of Wnt2b on mitochondrial dysfunction in AD. Methods Healthy and AD subjects, AD transgenic mice, and in vitro models were used to investigate the roles of Wnt2b in abnormalities in canonical Wnt signaling and mitochondria in AD. RT‐qPCR, immunoblotting, and immunofluorescence analysis were performed to assay canonical Wnt signaling. Mitochondrial structure was analyzed by electron microscopy. Flow cytometry was used to examine the intracellular calcium and neuronal apoptosis. Results Plasma Wnt2b levels were lower in AD patients and positively correlated with cognitive performance. Similarly, Wnt2b was reduced in the hippocampus of AD mice and in vitro models. Next, Wnt2b overexpression and recombinant Wnt2b were used to endogenously and exogenously upregulate Wnt2b levels. Upregulation of Wnt2b could effectively prevent downregulation of canonical Wnt signaling, mitochondrial dysfunction in in vitro AD models. Subsequently, intracellular calcium overload and neuronal damage were ameliorated. Conclusions Our study highlights that Wnt2b decline is associated with cognitive impairment in AD, and upregulation of Wnt2b can exert neuroprotective effects in AD, particularly in ameliorating mitochondrial dysfunction.


| INTRODUC TI ON
Alzheimer disease (AD) is the most common neurodegenerative disorder leading to dementia in the elderly worldwide. AD is neuropathologically characterized by the presence of extracellular βamyloid plaques and intracellular neurofibrillary tangles composed of hyperphosphorylated tau protein. 1 Wnt signaling participates in multiple aspects of cellular function, and contributes to synapse formation, synaptic activity, and neurogenesis in the brain, [2][3][4] and dysfunction of Wnt signaling may aggravate the pathogenesis and progression of AD. 5 We have been focusing on the research and development of blood and cerebrospinal fluid (CSF) markers that can be used for early diagnosis of AD, and we have found a variety of proteins that were significantly different in the blood and CSF of AD patients and cognitively normal subjects. 6,7 Recently, we found that Wnt2b, a member of WNT family, in the plasma of AD patients are lower than in healthy controls. Wnt2b has been studied in cancers or tumor-related diseases, 8 while Wnt2b is one of the most important components that can induce the particular developing pattern of the fetal hippocampus and related to neurogenesis, such as retinal ganglion cells. 9,10 Interestingly, Wnt2b is also characterized as a mitochondria-expressed protein which shuttles between mitochondria and the nucleus. 11,12 However, Wnt2b has not been studied in AD, particularly since it is expressed in mitochondria, unlike other Wnt family members. The above evidence makes us intend to further explore the roles of Wnt2b in AD, especially on mitochondrial dysfunction. As known, mitochondrial dysfunction contributes to the development and progression of the pathological process that underlies AD, especially at the early stage. 13,14 Some mitochondriatargeting drugs have shown significant effect in improving pathological change and cognitive impairment in AD model. 15 Therefore, we hypothesized that upregulation of Wnt2b may improve mitochondrial dysfunction in AD.
In the present study, we were unable to detect CSF Wnt2b levels due to lack of appropriate CSF Wnt2b kit and inadequate detection techniques unfortunately, so human samples were only tested in plasma. We consider Wnt2b in plasma mainly from peripheral and cell secretion in blood. There are also some Wnt2b in plasma that may be neurogenic, and molecules in the plasma might be derived from central nervous system and these molecules can enter the periphery through exosomes and other ways. 16 Many studies have suggested that plasma biomarkers can reflect the progress and situation of neurological disorders to a certain extent. 17 Moreover, plasma is a more readily available and simple sample than CSF and brain tissue and has better translational potential. In this study, we also analyzed the correlation between plasma Wnt2b and cognitive function. In neurological disorders, some molecules in plasma, including proteins related to neurogenesis or synaptic plasticity, often have similar changes to those in the brain. 18 Therefore, we want to explore whether the changes in Wnt2b in the brain are similar to those in the plasma of AD patients. Considering the difficulty of detecting in the brain of AD patients, we examined the levels of Wnt2b protein in the brain of AD mice at different months of age and analyzed whether the changes were similar to those of plasma Wnt2b. We next investigate the effect of upregulating Wnt2b on mitochondrial dysfunction and neuronal damage in in vitro AD model. Our findings will provide some basis for the protection of Wnt2b in mitochondrial function in AD and highlight its potential for clinical application and drug development for AD.

K E Y W O R D S
Alzheimer's disease, mitochondrial function, neuronal damage, Wnt2b were generated as previously described. 19,20 The study protocol was approved by the Ethics Committee of Capital Medical University (No. AEEI-2017-004). These mice were housed under standard conditions. The brains of PS1 V97L transgenic mice and wild-type mice were extracted and the hippocampus were then isolated and homogenized after being lysed with RIPA lysis buffer with protease inhibitor and a phosphatase inhibitor mixture (Applygen Technology) for 30 min. The protein concentrations of all of the samples were determined using the BCA assay kit (Applygen Technology). The protein concentration of the samples was normalized and then the samples were denatured at 98°C for 10 min. 21

| Preparation of Aβ Os
Oligomerized Aβ 1-42 was prepared as previously reported. 22 Briefly, Aβ 1-42 was dissolved at 1 mg/mL in HFIP, and the HFIP was removed in a SpeedVac vacuum concentrator after 1 h. The peptide film was resolved at 1 mM in DMSO (Sigma-Aldrich). The solution was diluted to 100 μM with DMEM. Then, the solution was incubated at 4°C overnight followed by centrifuging at 14,000 × g for 10 min at 4°C to remove any insoluble aggregates. 23

| Immunoblotting
The western blot assays were based on our previous studies. 21,24 The samples were subjected to SDS-polyacrylamide gel electrophoresis for about 40 min at 80 V in stacking gel and about 50 min at 120 V in resolving gel. Proteins were transferred electrophoretically to PVDF membranes (Millipore) at 250 mA for 1-2 h. Membranes were washed with TBST (Tris-Buffered Saline plus 0.05% Tween-20, pH 7.4) and then incubated in blocking buffer (5% skimmed milk in TBST) for 1 h at room temperature. The membranes were incubated overnight at 4°C on a shaker with primary antibody in TBST plus 5% skimmed milk. After four 5-min washes in TBST buffer, the membranes were incubated for 1 h at room temperature on a shaker with horseradish peroxidase (HRP)-conjugated secondary antibody diluted in blocking buffer. The membranes then underwent four 5min washes with TBST and were visualized using the ECL chemiluminescence detection kit (Millipore). The immunoblots were quantified with the Gel Doct EZ system (Bio-Rad). Band intensities were normalized to β-actin and were analyzed with Quantity One software (version 4.4.0, BioRad).

| RT-qPCR assay
Total RNA was extracted from HT22 cells using the RNAsimple Total RNA kit (Tiangen). After DNAse digestion reverse transcription of total RNA was performed using FastKing RT Kit (Tiangen). 2x TB Green Premix Ex Taq II (Takara) was used for real-time quantitative polymerase chain reaction (RT-qPCR) of the respective genes.
Cycling conditions were as follows: initial denaturation at 95°C for

| Immunofluorescent staining
The mouse was perfused with 0.01 mol/L PBS and 4% paraformaldehyde (pH 7.4), followed by post-fixation overnight at 4°C. The cultured cells were washed with PBST three times, fixed with 4% paraformaldehyde for 20 min, and followed by permeabilization with 0.1% Triton X-100. After blocked with 10% donkey serum (Solarbio), cells were incubated with primary antibody. Primary antibody incubation was performed on a shaker overnight at 4°C, followed by secondary antibody incubation for 1 h at room temperature. The following antibodies were used at the indicated concentrations: anti-Wnt2b antibody (1:100), donkey anti-rabbit IgG Alexa Fluor Plus 488 secondary antibody (1:500). DAPI reagent (Beyotime) was used for the nucleus staining. Images were taken using a fluorescence microscope (Olympus) and processed with ImageJ (https:// imagej.nih.gov/ij/).

| Mitochondrial structure and function
Mitochondrial structure was analyzed by electron microscopy. Then, images were obtained using a digital video camera (Tecnai G20 TWIN, FEI company), and the images were transferred to a computer. The intracellular ATP levels were analyzed by ATP assay kit (Beyotime Institute of Biotechnology). The mitochondrial membrane potential was detected by JC-1 fluorescent dye kit (Beijing Solarbio Science & Technology Co., Ltd). Cells were subjected to JC-1 fluorescent dye for 20 min in the dark. Fluorescence microscope was used to capture images, and the ratio of red fluorescence intensity to green fluorescence intensity was calculated to analyze the change in mitochondrial membrane potential. For ROS detection, cells were subjected to 1 μM DCFH-DA (Beijing Solarbio Science & Technology Co., Ltd) at 37°C for 20 min. Fluorescence microscope was then used to capture images (Ex: 488 nm, Em: 525). Fluorescence of the control group was defined as 100% and fluorescence of other groups was normalized to that of control group.

Apoptosis was measured using Annexin V-FITC Apoptosis Detection
Kit (C1062, Beyotime) according to the manufacturer's guidelines.

| Human subjects and plasma
Wnt2b detection 40  For the detection of Wnt2b protein, the plasma was directly detected with stock suspension, and its concentration was presented as pg/mL through a standard curve.

| Statistical analysis
Data were presented as the mean ± standard error of mean (SEM) or mean ± standard deviation (SD) or number (%). The biochemical measurements were performed with the experimenter blind to the experimental groups. Statistical analysis of data were performed using PRISM (version 8) (GraphPad Software) and/or SPSS for Windows (version 21.0) (SPSS Inc.). The difference between groups was performed using χ 2 test for categorical variables. All continuous data were subject to test for normality. Variables subject to normal distribution were analyzed using independent samples t-test, or oneway analysis of variance (ANOVA) followed by Tukey's post hoc test as appropriate. Variables that do not exhibit a normal distribution were analyzed using Mann-Whitney U test or Kruskal-Wallis test.
The multiple logistic regression was performed to detect the influence of Wnt2b on AD after adjusting for age, gender, APOE ε4, and education with SPSS software. The association between qualitative variables was evaluated by Spearman correlation. Values of p < 0.05 were considered statistically significant.

| Plasma Wnt2b was decreased in AD patients and correlated with cognitive function
We have previously reported a variety of proteins that have significant changes in the blood or cerebrospinal fluid (CSF) of AD patients. 6 Figure 1A), and demographic data as shown in Table 1. After other confounders including age, gender, education, and APOE ε4 were adjusted, there were still differences between NC and AD subjects in plasma Wnt2b level (Table S1). Further correlation of Wnt2b protein and cognitive performance (MoCA) was analyzed, and plasma Wnt2b was positively correlated with MoCA scores ( Figure 1B).

| Wnt2b expression was declined in in vivo and in vitro AD model
In order to figure out Wnt2b expression and levels in the brain, we used PS1 V97L transgenic mice that an AD transgenic model mice constructed and reported by our team. PS1 V97L mice began to appear Aβ oligomers in the brain at the age of 6-month and showed cognitive impairment at 9-month-old. 19

| Overexpression of WNT2B improved mitochondrial dysfunction and neuronal damage in HT22 cells in the presence of AβOs
In addition to being involved in canonical Wnt signaling, Wnt2b is characterized as a mitochondrial protein. 11,28 We found that Wnt2b was both expressed in the mitochondrial and cytoplasm fraction of mouse hippocampal HT22 cells ( Figure S2), and mitochondrial dysfunction is an important pathological feature of AD.
Therefore, we explored improving Wnt2b decline on mitochondrial dysfunction and downregulation of canonical Wnt signaling in AD. pcDNA3.1 constructs with WNT2B overexpression were used to upregulate the expression of Wnt2b in hippocampal cells, and we used the optimal transfection condition ( Figure S3). We found that overexpression of WNT2B could significantly reverse increase in pβ-catenin /β-catenin and decrease in p-GSK3β/GSK3β and BDNF level after AβOs exposure for 24 h (Figure 3A,B). Mitochondrial morphologic was analyzed using electron microscope, we found that the number of mitochondria was decreased and mitochondria showed swelling, disordered ridges, and partial vacuolization in the presence of AβOs, while more mitochondria with arranged ridges were seen in the WNT2B-overexpressed group ( Figure 3C). Overexpression of WNT2B could effectively ameliorate disruption of mitochondrial membrane potential ( Figure 3D, Figure S4), increase in intracellular ROS level ( Figure 3E, Figure S4) and decrease in intracellular ATP level ( Figure 3F) induced by AβOs. Overexpression of WNT2B also normalized the increased expression of mitochondrial stress genes CLPX, and OMA1 ( Figure 3G).
Mitochondria play a vital role in calcium homeostasis, 29 and mitochondrial dysfunction-induced intracellular calcium imbalance leads to neuronal death and is implicated in AD. 30 Next, we used flow cytometry to detect the changes in intracellular Ca 2+ and neuronal apoptosis, and we found that overexpression of WNT2B could F I G U R E 1 Wnt2b was decreased in plasma of AD patients and in brain of AD mice. (A) Plasma Wnt2b levels in AD patients were lower than that in NC groups. (B) Plasma Wnt2b protein was positively correlated with cognitive function (r 2 = 0.258, p = 0.021). (C) Representative western blots and quantification of fold changes in Wnt2b, β-Actin expression in the hippocampus of 6-month-old mice. (D) Representative western blots and quantification of fold changes in Wnt2b, β-Actin expression in the hippocampus of 9-month-old mice. Data values are expressed as mean ± SD or mean ± SEM. The difference between groups was performed using χ 2 test. The multiple logistic regression was performed to detect the influence of Wnt2b on AD after adjusting for age, gender, APOE ε4, and education with SPSS software. The association between qualitative variables was evaluated by Spearman correlation. The difference between WT and AD mice was performed using independent samples t-test. *p < 0.05 in comparison to WT mice, **p < 0.01 in comparison to WT mice. AD, Alzheimer's Disease; NC, normal cognitive; WT, wild-type prevented the increase of intracellular Ca 2+ (Figure 4H,I) and neuronal apoptosis ( Figure 4J,K) in HT22 cells induced by AβOs.

| Recombinant Wnt2b rescued mitochondrial dysfunction and neuronal damage in HT22 cells in the presence of Aβ Os
In addition to endogenously increasing Wnt2b expression in HT22 cells, we tested whether exogenous recombinant Wnt2b (rWnt2b) could ameliorate mitochondrial dysfunction and neuronal damage.

| DISCUSS ION
In the present study, we mainly found plasma Wnt2b levels in AD patients were decreased and positively correlated with cognitive function, and explored the pathological mechanism of Wnt2b in Alzheimer's disease in vitro and in vivo. Interestingly, we found ing the potential of Wnt2b in neurological disorders. 35 And our study found Wnt2b decline in AD was associated with cognitive impairment and underlying mechanism for the first time.
In this study, we found Wnt2b decline in the plasma of AD subjects, several studies have suggested that changes in blood Wnt protein can indeed reflect the occurrence and progression of some diseases. 36 Suzuki et al. have found that some serum peptides, including fragment of Wnt2b, were involved in the pathology of dementia with Lewy bodies (DLB) or Parkinson's disease. 37 Here, we found plasma Wnt2b levels were lower in AD patients than in agematched cognitively normal subjects and Wnt2b is positively correlated with cognitive function, which is consistent with the roles of Wnt signaling on cognition. 38 Our results provide some reference for Wnt2b in diagnosing neurological diseases including AD.
We tried to detect in cerebrospinal fluid, which is more diagnostic than plasma, but cannot be effectively and accurately detected due to insufficient technology and lack of appropriate detection kits. Considering that blood samples are clinically easy to obtain and detect, our study will also provide some help in the diagnosis of AD. We found Wnt2b was reduced in AD animal model and genetic modifications may also affect WNT2B. 45 However, we did not investigate the mechanism of WNT2B protein and mRNA decrease in AD in this study, and we will continue to explore it. The beneficial effects of WNT2B overexpression and recombinant Wnt2b verified the importance of Wnt2b in canonical Wnt signaling and mitochondrial dysfunction in AD. Several lines of evidences hint Wnt2b may play a role in mitochondrial function. 11,12 We found that Wnt2b was also expressed in mitochondria (Appendix S1 and S2), and aberrant mitochondrial structure and function in AD cell models were improved when WNT2B overexpressed or rWnt2b applied, thereby intracellular calcium imbalance and neuronal apoptosis induced by AβOs was prevented. As described in other studies, Wnt signaling is associated with mitochondrial function, particularly in some age-related disease. 15,46 Accumulation of damaged and dysfunctional mitochondria has been reported as an early important pathological feature of AD, and further contributes to disease progression. 47,48 Targeting mitochondria is a promising therapeutic strategy to improve neuronal damage and cognitive impairment in AD. 49 In the presence of Aβ, abnormal mitochondrial structure and function causes bioenergetic deficiency, intracellular calcium imbalance, and oxidative stress, thereby aggravating the effect of Aβ and tau pathologies, leading to synaptic dysfunction, cognitive impairment, and memory loss. 50   AD models and not validated in primary neuronal culture and animal models, which is a shortcoming of this study. We will continue to study the effects of Wnt2b on mitochondrial and synaptic dysfunction and cognitive impairment in primary neurons and animal models in depth. The study mainly detected the mitochondrial structure and function, and specific biological mechanism by which Wnt2b affects also needs to be further studied. What's more, although we detected differences in plasma Wnt2b level between the AD and cognitively normal groups, we will expand the sample size and enroll mild cognitive impairment subjects to further validate the relationship between plasma Wnt2b and cognitive function in AD. Next, we will explore the changes of Wnt2b in CSF and the correlation with plasma Wnt2b by optimizing the detection technology and appropriate kits. And we consider AD is a very complex neurodegenerative disease, and the cognitive diagnosis and assessment of AD should be comprehensive, rather than Wnt2b. If combined with other ADrelated biomarkers, it may be more effective and accurate for the diagnosis of AD.

| CON CLUS ION
In summary, the present study mainly found decrease of plasma Wnt2b levels in AD patients, and canonical Wnt signaling and mitochondrial dysfunction might be potential mechanisms for Wnt2b involved in neuronal damage and cognitive impairment in AD. Our findings may advance the understanding of the importance of Wnt family in AD, and we propose that Wnt2b might be a potential diagnostic indicator and therapeutic target for AD.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare that they have no conflicts 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 on request from the corresponding author.

CO N S E NT TO PA RTI CI PATE
All participants signed the ethics approval and consent to participate.