Role of monocarboxylate transporter 4 in Alzheimer disease

Journal Pre-proof Role of monocarboxylate transporter 4 in Alzheimer disease Ping Hong, Xiaoyi Zhang, Shichao Gao, Peichang Wang

PII:

S0161-813X(19)30133-0

DOI:

https://doi.org/10.1016/j.neuro.2019.11.006

Reference:

NEUTOX 2554

To appear in:

Neurotoxicology

Received Date:

10 June 2019

Revised Date:

9 November 2019

Accepted Date:

11 November 2019

Please cite this article as: Hong P, Zhang X, Gao S, Wang P, Role of monocarboxylate transporter 4 in Alzheimer disease, Neurotoxicology (2019), doi: https://doi.org/10.1016/j.neuro.2019.11.006

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Role of monocarboxylate transporter 4 in Alzheimer disease

Ping Hong a, 1, Xiaoyi Zhang b, 1, Shichao Gao a, Peichang Wang a, * a

Clinical Laboratory of Xuanwu Hospital, Capital Medical University, Beijing 100053, China

b

Department of Administration Management, Weifang People's Hospital, Weifang, Shandong

Province 261000, China Corresponding author. E-mail addresses: [email protected]

1

The two authors contributed equally to this work.

ro of

*

Highlights

MCT4 expression was elevated in the cerebrospinal fluid of patients with mild cognitive impairment. The APP/PS1 mice began to show cognitive decline at 3 months of age and MCT4 in the hippocampus of 2- and 3-month old APP/PS1 mice was higher than that of C57 mice. This change is similar to that in people with mild cognitive impairment. Overexpression of cytoplasmic MCT4 increased the expression of Aβ42, γ-secretase, and CD147 in the co-culture system; in addition, the growth ability of primary neurons decreased significantly, extracellular lactic acid increased, and neuronal apoptosis increased. In AD model mice, siMCT4 injection improved cognitive ability, reduced neuronal apoptosis, and reduced γ-secretase expression.

ur

na



lP



re

-p



Jo

Abstract The pathological process of Alzheimer disease (AD) is closely related to energy metabolism disorders. In the nervous system, monocarboxylate transporter 4 (MCT4) is expressed in the glial cell membrane and is responsible for transporting intracellular lactic acid. In this study, we found that MCT4 expression was elevated in the cerebrospinal fluid of patients with mild cognitive impairment. Two- and three-month-old APPswe/PS1dE9 (APP/PS1) mice and C57 mice were studied. The APP/PS1 mice began to show cognitive decline at 3 months of age and MCT4 in the hippocampus of 2- and 3-month old APP/PS1 mice was higher than that of C57 mice. This change is similar to that in people with mild cognitive impairment. Subsequently, MCT4 overexpression/siRNA lentiviral particles were used to establish stable primary astrocytes. Overexpression and knockdown of MCT4 had no significant effect on glial cell apoptosis.

Jo

ur

na

lP

re

-p

ro of

Transfected astrocytes were co-cultured with neurons. Overexpression of cytoplasmic MCT4 increased the expression of Aβ42, γ-secretase, and CD147 in the co-culture system; in addition, the growth ability of primary neurons decreased significantly, extracellular lactic acid increased, and neuronal apoptosis increased. In AD model mice, siMCT4 injection improved cognitive ability, reduced neuronal apoptosis, and reduced γ-secretase expression. Taken together, these results suggest that MCT4 is involved in energy metabolism during early pathological processes in AD, and suppression of MCT4 represents a new potential neuroprotective factor for AD. Keywords: Alzheimer disease, Monocarboxylate transporter 4, Mitochondria, Lactic acid, Apoptosis 1. Introduction Alzheimer disease (AD) is a neurodegenerative disease that is clinically characterized by progressive cognitive impairment and personality changes that severely affect the quality of life of patients. Elucidating the pathogenesis of AD and formulating appropriate treatment measures has become an important issue to be solved in AD research. Normal energy metabolism is important for the brain in carrying out its various physiological activities. Energy metabolism disorders affect the normal function of neurons and are associated with memory disorders in AD (Dimitrios and Mattson, 2011). Current research confirms that lactic acid is not only an energy substrate used preferentially by neurons, it also plays an important role in memory formation (Bouzier Sore et al. , 2003). Monocarboxylate transporters (MCTs) are a type of transmembrane transporter of monocarboxylic acids, such as lactic acid. There are three subtypes of MCT in the brain: MCT1, MCT2, and MCT4; these synergistically transport lactic acid between astrocytes and neurons (Machler et al. , 2016). In the early stages of this experiment, we noted that MCT4 is elevated in the cerebrospinal fluid (CSF) of patients with mild cognitive impairment. There is currently no report of the correlation between MCT4 and AD. We investigated the potential of MCT4 as an AD drug via in vitro and in vivo experiments. Moreover, we found that its anti-AD mechanism may be mediated by mitochondrial energy metabolism. 2 Materials and Methods 2.1 Study population A total of 16 patients with dementia (10 women, 6 men; mean age 71.05 ± 8.13 years), who were admitted to Xuanwu Hospital of Capital Medical University (Beijing, China) between December 2016 and December 2017, were enrolled in the present study. These patients were diagnosed with AD in the stage of mild cognitive impairment (MCI), based on the revised diagnostic criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) (Hogervorst et al. , 2003), published by the National Institute on Aging and Alzheimer's Association (NIA-AA) in April 2011. In addition, a total of 16 age- and sex-matched control participants (10 women, 6 men, mean age 69.64 ± 7.45 years) were included in the study. The design of this study was approved by the ethics committee of Xuanwu Hospital, Capital Medical University, and all patients and controls or their legally authorized representatives provided individual informed consent. The study was conducted in accordance with the tenets of the Declaration of Helsinki. 2.2 Protein chip detection A 5-mL volume of CSF was obtained from patients and control participants by lumbar

Jo

ur

na

lP

re

-p

ro of

puncture, shipped on dry ice, and stored in liquid nitrogen until required. Differentially expressed proteins were detected using human whole-genome expression chip technology containing 38,400 protein spots. Positively expressed proteins were further verified using a low-density protein chip. 2.3 AD transgenic mice APPswe/PS1dE9 (APP/PS1) mice were purchased from the Institute of Laboratory Animal Science, Chinese Academy of Medical Sciences & Comparative Medical Center. All animal experiments conformed to the National Institutes of Health guidelines. All animal protocols were approved by the ethics committee of Xuanwu Hospital. 2.4 Morris water maze (MWM) testing Behavioral tests in mice were performed in a silent, isolated room maintained within a constant temperature range (20–24°C). MWM testing was carried out to assess spatial learning and memory performance of the mice (Dai et al. , 2015). The escape latency to reach the hidden platform during the acquisition phase was determined over 5 days. Swimming trajectories were videotaped for subsequent analysis. 2.5 ELISA assay The levels of MCT4 in mouse hippocampal homogenates were measured using Ultrasensitive ELISA kits (Invitrogen, Camarillo, CA, USA), according to the manufacturer’s protocol. 2.6 Protein analysis Frozen hippocampi or cultured cells were homogenized in ice-cold RIPA buffer (Beyotime Biotechnology, Jiangsu, China) containing 0.5 mM phenylmethanesulfonyl fluoride (PMSF). Insoluble material was removed by centrifuging the homogenates at 16,500 g for 15 minutes at 4°C. Protein concentrations in the supernatants were quantified using a BCA protein assay kit (Beyotime Biotechnology, Jiangsu, China). Protein samples were boiled for 5 minutes with loading buffer, and proteins (30 μg per well) were separated by 15% SDS-polyacrylamide gel electrophoresis (Beyotime Biotechnology, Jiangsu, China) and then transferred onto nitrocellulose membranes (Millipore, Watford, UK). Membranes were blocked with 5% nonfat milk in PBST for 1 hour and then incubated at room temperature in PBST containing 3% BSA. Subsequently, the following primary antibodies were added: anti-MCT4 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-γ- secretase (Santa Cruz Biotechnology), anti-amyloid beta (Aβ42; Sigma-Aldrich, St. Louis, MO, USA), anti-β-actin (Santa Cruz Biotechnology), and anti-CD147 (Cell Signaling Technology, Danvers, MA, USA). Membranes were washed and incubated with horseradish peroxidase-conjugated anti-IgG antibody at room temperature for 1 hour. Bands were revealed by chemiluminescence, and band intensities were quantified using ImageJ software. 2.7 Cell culture Primary neuronal or glial cultures were prepared from the hippocampi of embryonic day 16 embryos of APP/PS1 mice. After mechanical dissociation, cells were cultured on plates and maintained in Gibco Neurobasal Medium (Gibco, Invitrogen, USA) containing 2% B27 supplement and 100 units/mL penicillin-streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) at 37°C in a humidified 5% CO2 atmosphere; primary neurons were subsequently obtained. 2.8 Cell transfection The RNA of primary astrocytes was extracted and reverse-transcribed into the first strand cDNA. The mRNA sequence of the MCT4 gene was searched in the gene bank, and the specific upstream and downstream primers of the target gene were designed. Overexpression oligonucleotide sequence, sense: 5-ATGGGAGGGGCTGTGGTTGA-3; antisense:

Jo

ur

na

lP

re

-p

ro of

5-CCCGGAAACAAGCGTTTGA-3 and specific shRNA oligonucleotide sequence targeting the MCT4 coding region, sense: 5-CCCACGUCUACAUGUACGU-3; antisense: 5-ACGUACAUGUAGACGUGGG-3. The mRNA secondary structure was analyzed using RNA structure software and BLAST homology analysis was performed. Two pairs of specific oligonucleotide strands were synthesized, and the abovementioned synthesized MCT4 overexpression/siRNA sequence was inserted into the lentiviral vector (Santa Cruz Biotechnology, CA, USA); the recombinant plasmid and helper plasmid were simultaneously transfected into HEK293 cells. After 48 hours, the supernatant was collected and concentrated. Primary cultured astrocytes were transfected with lentiviral particle-containing solution, and the transfection efficiency was measured using real-time PCR and western blot techniques. Stably transfected cells were screened by flow cytometry. 2.9 Cell viability assay The transfected astrocytes were co-cultured with primary neurons in Corning Transwell membrane-nested 96-well plates (Corning, NY, USA). The empty vector and untransfected astrocytes were cultured as a control group, and an equal amount of whole medium was added to each group. After incubation at 37°C, 5% CO2 for 24 hours, 20 μL of MTT (Beyotime Biotechnology, Jiangsu, China) solution was added to each well of 96-well plates, and after incubating at 37°C for 4 hours, the supernatant in each well was carefully aspirated. Supernatants were then discarded and 150 mL of DMSO was added. After 10 minutes of low-speed shaking (100 rpm), the absorbance at 490 nm in each well was measured using a Bio-Rad 680 microplate reader (Bio-Rad, Hercules, CA, USA). The above steps were repeated every 24 hours. The viability curve was plotted with time as the horizontal axis and absorbance as the vertical axis. All experiments were performed in biological triplicate. 2.10 Apoptosis analysis To determine the percentage of apoptotic neurons, primary neurons were harvested after 24 hours of co-culture, and then stained with FITC-labeled Annexin V and propidium iodide (BD Pharmingen, San Diego, CA, USA). Stained cells were immediately analyzed on a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA). 2.11 Mitochondrial function test About 2 × 105 primary neurons were co-cultured with empty vector or untransfected astrocytes, and then resuspended in 0.5 mL cell culture medium; 0.5 mL JC-1 staining solution (Beyotime Biotechnology, Jiangsu, China) was added and the solution was mixed by inverting culture flasks several times. Neurons were incubated in a cell culture incubator at 37°C for 20 minutes. During the incubation period, an appropriate amount of JC-1 staining buffer (Beyotime Biotechnology, Jiangsu, China) was prepared by adding 4 mL of distilled water per 1 mL staining buffer (5X), which was placed in an ice bath and then centrifuged at 37°C for 4 minutes. The supernatant was discarded and cells were washed twice with JC-1 staining buffer (1X), resuspended by adding 1 mL of staining buffer (1X), centrifuged at 600 g for 4 minutes, and then pelleted and the supernatant discarded. The cells were resuspended in 1 mL JC-1 buffer (1X), centrifuged at 600 g for 4 minutes, and precipitated; the supernatant was then discarded and cells resuspended in an appropriate amount of JC-1 staining buffer (1X). Cells were visualized by laser confocal microscopy (Carl Zeiss, Jena, Germany) and subsequently analyzed by flow cytometry (BD Biosciences, San Jose, USA). 2.12 Lactic acid assay

Jo

ur

na

lP

re

-p

ro of

Astrocytes and neurons were co-cultured for 48 hours, and 100 μL of the co-culture system solution was collected. The lactate concentration was measured using a lactate assay kit (BioVision, Milpitas, CA, USA), according to the manufacturer’s instructions. 2.13 Stereotaxic administration APP/PS1 mice were anesthetized with 4% chloral hydrate (10 mL/ kg, intraperitoneally) and fixed in a stereotaxic apparatus. Each mouse was administered about 5 mL of transfection complex solution containing 0.2 mg siRNA or a scrambled sequence as a negative control (Santa Cruz, CA, USA) and 0.5 mL Invitrogen Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). The transfection complex solution was injected into the third ventricle using a 10 mL microsyringe, at a rate of 0.2 mL/minute. Sham-operated APP/PS1 mice received an equal volume of PBS. 2.14 Immunofluorescence Frozen sections were air-dried at room temperature for 15 minutes and then fixed in 4% paraformaldehyde for 10 minutes. Sections were then rinsed with PBST (0.25% Triton X-100) and blocked with 5% goat serum for 30 minutes at room temperature. After incubation with primary antibodies diluted in PBS for 2 hours at room temperature, sections were washed with PBS and incubated with secondary antibodies for 1 hour at room temperature. Sections were washed again and mounted using mounting medium containing DAPI (Vector Lab, Burlingame, CA, USA). Primary antibodies used were as follows: anti-γ-secretase (1:400; Cell Signaling Technology, MA, USA) and anti-CD147 (1:300; Santa Cruz, CA, USA). All immunofluorescence images are representative of at least three independent experiments. 2.15 TUNEL staining Neuronal apoptosis in the hippocampus was assessed using a TUNEL staining kit (Keygen Biotech, Nanjing, China), according to the manufacturer's instructions. Frozen embedded sections were incubated with TUNEL reaction mixture in a humidified atmosphere for 1 hour at 37°C in the dark. Mounting medium containing DAPI (Vector Lab, Burlingame, USA) was used to visualize the nuclei. Images were acquired with a confocal laser scanning microscope (Carl Zeiss, Jena, Germany). The apoptotic rate was calculated as (number of apoptotic cells per section/total number of cells per section) × 100%. Experiments were performed in triplicate. 2.16 Statistical analysis Statistical analysis was performed using GraphPad Prism software version 5.01 (GraphPad, San Diego, California, USA). All data are presented as mean ± standard deviation. The difference between two groups was analyzed using the Student t-test. The difference among more than two groups was analyzed using one-way analysis of variance (ANOVA) followed by Bonferroni’s post-hoc test. P-values < 0.05 were considered significant. 3. Results 3.1 Human protein chip technology for analysis of cerebrospinal fluid To identify protein changes in patients with early-stage AD, we examined the CSF of 16 patients with MCI using whole-genome chip technology. All enrolled patients developed dementia of Alzheimer type (DAT) after 1 year. Compared with the CSF of healthy controls, 13 proteins were differentially expressed in patients with MCI (Fig. 1A) including SLC16A4 (also known as MCT4), which is a functional protein in energy metabolism. The differential expression of MCT4 in patients with MCI and healthy controls was further confirmed by ELISA (Fig. 1B). 3.2 Behavior and protein analyses in APP/PS1 and C57 mice at 2 and 3 months old

Jo

ur

na

lP

re

-p

ro of

To determine whether there is also a change in MCT4 expression in the mouse model of early-stage AD, we first conducted Morris water maze (MWM) tests in 2- and 3-month-old APP/PS1 mice, with C57 mice of corresponding ages as controls. The passing times and platform quadrant residence times of APP/PS1 mice at 2 months of age were normal; however, passing times and platform quadrants residence times of mice aged 3 months were significantly lower than those of the control group (Fig. 1C, 1D, 1E). Further, APP/PS1 mice aged 3 months took significantly longer time than control mice to find the hidden platform (Supplementary Fig. 1A). However, no significant difference was observed in swimming speed among the four groups during the training trial (Supplementary Fig. 2B). These results suggest that APP/PS1 mice at 3 months of age begin to experience cognitive decline and have cognitive status that is similar to that of people with MCI. We then further observed whether MCT4 expression was also present in the brain tissue of APP/PS1 mice at 2 and 3 months of age. Western blot results showed that MCT4 in the hippocampus of mice at 2 and 3 months of age was higher than that in the control group (Fig. 1F, 1G). These results suggest that the change of MCT4 protein expression occurs earlier than behavioral changes in APP/PS1 mice. 3.3 Construction of MCT4 overexpression/siRNA lentiviral vector to establish stable astrocytes We transfected primary astrocytes with a solution containing MCT4 overexpression/siRNA lentiviral particles and measured the transfection efficiency using Western blot. Untransfected astrocytes were used as the normal control (NC) group, and we included an empty vector group (Vehicle group). There was no difference in MCT4 expression between the Vehicle and NC groups whereas expression in the MCT4-overexpression group (OE-MCT4 group) was significantly higher than that of the Vehicle group. Expression of the siRNA lentiviral vector group (siMCT4 group) was significantly lower than that of the Vehicle group, suggesting that stably transfected astrocytes were successfully established (Fig. 2A, 2B). To determine whether there is a change in the rate of apoptosis after establishment of stably transfected astrocytes, we tested the four groups of cells using flow cytometry. The results showed that there was no significant difference in the apoptosis rate among the four groups, indicating that MCT4 expression changes had no effect on apoptosis of astrocytes (Fig. 2C and 2D). 3.4 Aβ42, γ-secretase, and CD147 protein quantification after astrocyte and neuron co-culture To clarify the changes of AD-related proteins in the primary astrocyte–neuron co-culture system, we used Transwell membrane-nested 96-well plates to co-culture MCT4 overexpression/siRNA astrocytes with primary neurons. After culture, Aβ42, γ-secretase, and CD147 proteins were quantified by Western blot. An NC group and Vehicle group were included as defined in 1.3. The results showed that expression of Aβ42, γ-secretase, and CD147 was increased in the OE-MCT4 group, and expression of MCT2 was unchanged. The expression of Aβ42, γ-secretase, and CD147 was decreased and that of MCT4 was unchanged in the siMCT4 group (Fig. 3). These results indicate that the expression of MCT4 in astrocytes affected the expression of Aβ42, γ-secretase, and CD147 in the co-culture system but had no effect on MCT2. 3.5 Detection of extracellular matrix components, neuronal growth ability, and apoptosis after co-culture To determine whether MCT4 overexpression/knockdown in astrocytes has an effect on neuronal apoptosis, we analyzed neuronal apoptosis using flow cytometry. We observed no difference in the neuronal apoptosis rate between the Vehicle and NC groups after 48 hours of co-culture. The apoptosis rate in the OE-MCT4 group was significantly higher than that in the Vehicle group; the

Jo

ur

na

lP

re

-p

ro of

siMCT4 group was not different from the Vehicle group (Fig. 4A, 4B). To further clarify the effect of MCT4 expression in astrocytes on neuronal growth, we assessed neuronal growth ability after co-culture of astrocytes and neurons for 6, 12, 24, 36, 48, and 72 hours. The growth ability of primary neurons in the OE-MCT4 group was significantly lower after 48 hours, which was statistically different from that in the Vehicle group. The growth ability of neurons in the siMCT4 group demonstrated an increasing trend compared with that in the Vehicle group, but this was not significant (Fig. 4C). We assessed the effect of MCT4 expression levels on the cell microenvironment pH and lactic acid content. After co-cultivation of astrocytes and neurons for 48 hours, pH was decreased and the lactic acid content was increased in the OE-MCT4 group; the lactic acid content was decreased and the pH was increased in the siMCT4 group (Fig. 4D, 4E). As lactic acid metabolism affects mitochondrial function, we further examined the effects of MCT4 expression on neuronal mitochondrial function using JC-1 staining. In the OE-MCT4 group, the red/green fluorescence ratio of primary neurons was significantly lower, indicating that the mitochondrial membrane potential was significantly decreased. The red/green fluorescence ratio of neurons was enhanced compared with the NC group, but this was not significant (Fig. 4F, 4G). These results suggest that changes of MCT4 expression in astrocytes can affect lactic acid metabolism, neuronal apoptosis rate, and mitochondrial function. 3.6 Behavioral changes, apoptosis, and protein expression in siMCT4-injected mice To clarify behavioral changes in the animal model after MCT4 expression intervention, we stereotactically injected siMCT4 particles into the bilateral third ventricle of APP/PS1 mice to inhibit MCT4 expression (knockdown of MCT4; KD-MCT4 group). The MWM test was used to assess the memory ability of mice after 2 weeks of injection. We found that the KD-MCT4 group had a shorter escape latency than the sham (injected with normal saline) and NC (injected with empty vector) groups (Supplementary Fig. 1B). Further, the passing times and platform quadrant residence times increased (Fig. 5A, 5B, 7C), suggesting an improvement in cognitive ability. The brain tissue of each group was stained with TUNEL to measure the apoptotic rate of neurons. The apoptosis rate of KD-MCT4 group was significantly lower than that of the sham and NC groups (Fig. 5D, 5E). CD147 and γ-secretase immunofluorescence staining was performed to detect the expression of these two proteins. CD147 and γ-secretase fluorescence in the KD-MCT4 group was lower than that in the sham and NC groups (Fig. 5F). 4. Discussion In this study, we used protein chip technology to analyze the CSF of patients with AD. The expression of MCT4 was significantly increased, i.e., MCT4 was mainly expressed in astrocytes. The expression of MCT2 in neurons was unchanged. After regulating MCT4 overexpression in astrocytes in vitro, neuronal proliferative capacity decreased and apoptosis increased. After knocking down expression of MCT4 in the hippocampus of AD model mice, the apoptosis rate of hippocampal neurons was significantly reduced, and the cognitive ability of mice was improved. These results suggest that MCT4 can be used as a potential target for AD treatment. At present, the pathological mechanism of AD remains unclear. The pathogenesis of AD is increasingly believed to involve multiple etiological mechanisms, and many factors could eventually lead to abnormal energy metabolism. The energy metabolism process is a key hub in the development of AD, with a switching role in the progression of the disease (Pedros et al. , 2016). Astrocytes mainly support neurons and participate in brain energy metabolism, which plays

Jo

ur

na

lP

re

-p

ro of

an important role in the pathophysiological mechanism of AD (Bhat et al. , 2013). Changes in the astrocytes of patients with AD precede the formation of senile plaques and neurofibrillary tangles (Yeh et al. , 2011). In the present study, MWM test results in 3-month-old APP/PS1 mice were abnormal, but there was no significant change in APP/PS1 mice that were 2 months old; thus, the 2-month-old mice were in the pre-onset period of AD. MCT4 expression was abnormal in both 2and 3-month-old APP/PS1 mice, suggesting that change in the expression of MCT4 in astrocytes occurs earlier than cognitive decline. In astrocytes, lactic acid is transported to the intercellular substance via a monocarboxylate transporter, mainly MCT4. Increased lactic acid in the stroma enhances neuronal oxidative phosphorylation. At the same time, MCT2 on the surface of the neuronal membrane initiates the transport of lactic acid into the neurons, providing these with energy or exerting other biological functions. In this study, we found that the expression of MCT2 in the co-culture system did not change, and increased expression of MCT4 in astrocytes resulted in the accumulation of lactic acid in the stroma, leading to lactic acidosis. Lactic acid is found in the frontal cortex and caudate nucleus of patients with AD, where the lactic acid content is increased and the phosphorylated glutaminase and glutamate decarboxylase content are decreased (Yates et al. , 1990). On CSF examination in a canine model of AD, Pugliese et al. (Pugliese et al. , 2005) found that the concentrations of lactic acid, pyruvate, and potassium increase in parallel with the increase in cognitive impairment. In addition, acidosis may lead to mitochondrial dysfunction via the opening of the mitochondrial permeability transition pore (Graham et al. , 2004) and may cause aggregation of Aβ, according to results in rodent AD models (Atwood et al. , 1998, Michael et al. , 2006). Acidosis may play an important role in the development of AD. MCT4 requires the assistance of CD147 for normal biological functions. CD147 is a glycoprotein distributed on the surface of cell membranes [7]. It can serve as a chaperone protein to promote the anchorage and expression of MCT4 on the cell surface and can regulate the activity of MCT4 (Kirk et al. , 2000). One study found that MCT4 could not be accurately identified on the cell membrane in retinal cells of CD147-knockout mice (Philp et al. , 2003). CD147 can also be involved in the production of Aβ as a subunit of γ-secretase (Vetrivel et al. , 2008). In this study, we found that in the astrocyte–neuron co-culture system, changes in MCT4 expression can affect neuronal mitochondrial function, lactate metabolism, and the apoptosis rate. In addition, MCT4 expression was positively associated with expression of Aβ42, CD147, and γ-secretase. In the present study, after knocking down expression of MCT4 in the hippocampus of APP/PS1 mice, the expression of CD147 and γ-secretase was decreased, the apoptosis rate of neurons was decreased, and the cognitive ability of mice was improved. At present, there is no effective treatment for AD. Treatment regimens using Aβ as a therapeutic target do not improve cognitive function or delay the progression of disease (Graham et al. , 2017). In our study, intervention of MCT4 expression in the brain tissue of APP/PS1 mice could improve cognitive function, suggesting that the energy-regulated metabolism of astrocytes may have an important role in the pathogenesis of AD, offering the possibility of a new therapeutic target in AD. Conflict of interest We declare that we have no conflicts of interest. Acknowledgments This study was supported by grants from the National Natural Science Foundation of China (No. 81501841).

Author contribution Author 1: Ping Hong Cell and animal experiments Collected the data Performed the analysis Wrote the paper Author 2: Xiaoyi Zhang Collected the data Contributed data or analysis tools Performed the analysis

ro of

Author 3: Shichao Gao Contributed data or analysis tools Performed the analysis

Jo

ur

na

lP

re

-p

Author 4: Peichang Wang Conceived and designed the analysis Performed the analysis Wrote the paper

References Atwood CS, Moir RD, Huang X, ., Scarpa RC, Bacarra NM, Romano DM, et al. Dramatic aggregation of Alzheimer abeta by Cu(II) is induced by conditions representing physiological acidosis. Journal of Biological Chemistry. 1998;273:12817. Bhat R, Crowe EP, Bitto A, Moh M, Katsetos CD, Garcia FU, et al. Astrocyte senescence as a component of Alzheimer's disease. Experimental Gerontology. 2013;48:695-6. Bouzier Sore AK, Voisin P, Canioni P, Magistretti PJ, Pellerin L. Lactate is a preferential oxidative energy substrate over glucose for neurons in culture. Journal of Cerebral Blood Flow & Metabolism. 2003;23:1298-306. Dai CL, Chen X, Kazim SF, Liu F, Gong CX, Grundke-Iqbal I, et al. Passive immunization targeting the N-terminal projection domain of tau decreases tau pathology and improves cognition in a transgenic mouse model of Alzheimer disease and tauopathies. Journal of neural transmission. 2015;122:607-17.

ro of

Dimitrios K, Mattson MP. Disrupted energy metabolism and neuronal circuit dysfunction in cognitive impairment and Alzheimer's disease. Lancet Neurology. 2011;10:187-98.

Graham RM, Frazier DP, Thompson JW, Shannon H, Huifang L, Wasserlauf BJ, et al. A unique pathway of cardiac myocyte death caused by hypoxia-acidosis. Journal of Experimental Biology. 2004;207:3189-200. Strategies. Annual Review of Medicine. 2017;68:413.

-p

Graham WV, Bonito-Oliva A, Sakmar TP. Update on Alzheimer's Disease Therapy and Prevention

Hogervorst E, ., Bandelow S, ., Combrinck M, ., Irani SR, Irani S, ., Smith AD. The validity and reliability

re

of 6 sets of clinical criteria to classify Alzheimer's disease and vascular dementia in cases confirmed post-mortem: added value of a decision tree approach. Dementia & Geriatric Cognitive Disorders. 2003;16:170-80.

lP

Kirk P, Wilson MC, Heddle C, Brown MH, Barclay AN, Halestrap AP. CD147 is tightly associated with lactate transporters MCT1 and MCT4 and facilitates their cell surface expression2000. Machler P, Wyss MT, Elsayed M, Stobart J, Gutierrez R, von Faber-Castell A, et al. In Vivo Evidence for a Lactate Gradient from Astrocytes to Neurons. Cell metabolism. 2016;23:94-102.

na

Michael P, Josef M, Christian H. Effects of acidosis on brain capillary endothelial cells and cholinergic neurons: relevance to vascular dementia and Alzheimer's disease. Neurological Research. 2006;28:657-64.

ur

Pedros I, Patraca I, Martinez N, Petrov D, Sureda FX, Auladell C, et al. Molecular links between early energy metabolism alterations and Alzheimer's disease. Frontiers in Bioscience. 2016;21:8. Philp NJ, Ochrietor JD, Rudoy C, Muramatsu T, Linser PJ. Loss of MCT1, MCT3, and MCT4 expression in

Jo

the retinal pigment epithelium and neural retina of the 5A11/basigin-null mouse. Investigative ophthalmology & visual science. 2003;44:1305-11. Pugliese M, Carrasco JC, Mas E, Mascort J, Mahy N. Severe cognitive impairment correlates with higher cerebrospinal fluid levels of lactate and pyruvate in a canine model of senile dementia. Prog Neuropsychopharmacol Biol Psychiatry. 2005;29:603-10. Vetrivel KS, Zhang X, Meckler X, Cheng H, Lee S, Ping G, et al. Evidence That CD147 Modulation of β-Amyloid (Aβ) Levels Is Mediated by Extracellular Degradation of Secreted Aβ. Journal of Biological Chemistry. 2008;283:19489. Yates CM, Butterworth J, Tennant MC, Gordon A. Enzyme activities in relation to pH and lactate in postmortem brain in Alzheimer-type and other dementias. Journal of neurochemistry.

1990;55:1624-30. Yeh CY, Vadhwana B, Verkhratsky A, Rodriguez JJ. Early astrocytic atrophy in the entorhinal cortex of a

Jo

ur

na

lP

re

-p

ro of

triple transgenic animal model of Alzheimer's disease. ASN neuro. 2011;3:271-9.

Jo

ur

na

lP

re

-p

ro of

Figure legends Fig. 1. Expression of monocarboxylate transporter 4 (MCT4) in patients with mild cognitive impairment (MCI) and transgenic (Tg) AD model mice. (A) Volcano map of cerebrospinal fluid detection in the MCI group and control group. (B) ELISA assay of MCT4 expression in the MCI group and control group (mean ± SD of independent experiments, n = 3, *p < 0.05). (C) Representative swimming trajectories of the probe trial in Morris water maze tests. (D) Passing times in the probe trial. (E) Percentage of time mice spent in the four quadrants during the probe trial (mean ± SD, n = 5 per group, *p < 0.05, #p > 0.05). (F) Western blot analysis of MCT4 in the hippocampus of C57/BL6 and Tg APP/PS1 mice at 2 and 3 months old (n = 5 per group). β-actin was used as a loading control. (G) Relative MCT4 levels in (F) (mean ± SD, n = 5 per group, *p < 0.05).

ro of -p re lP na ur

Fig. 2. Establishment of stably transfected astrocytes and apoptosis detection

Jo

(A) Western blot analysis of MCT4 protein in primary astrocytes transfected with lentivirus. β-actin was used as a loading control. (B) Relative MCT4 levels in (A) (mean ± SD, n = 3 per group, #p > 0.05, *p < 0.05). (C) Annexin V/PI staining of apoptotic astrocytes in the four cell groups. Representative images are shown. (D) Average percentages of apoptotic cells as determined by Annexin V/PI staining.

ro of -p

Jo

ur

na

lP

re

Fig. 3. Protein quantification after astrocyte and neuron co-culture. (A) Western blot analysis of four proteins in the co-culture system. β-actin was used as a loading control. (B) Relative MCT2 levels. (C) Relative amyloid beta (Aβ42) levels. (D) Relative γ-secretase levels. (E) Relative CD147 levels (mean ± SD, n = 3 per group, #p > 0.05, *p < 0.05).

Jo

ur

na

lP

re

-p

ro of

Fig. 4. Extracellular matrix components, neuronal growth, and apoptosis detection after co-culture. (A) Annexin V/PI staining of apoptotic neurons after co-culture. Representative images are shown. (B) Average percentages of apoptotic neurons as determined by Annexin V/PI staining (n = 3, #p > 0.05, *p < 0.05). (C) MTT assay of primary neurons after co-culture (n = 3, *p < 0.05 compared with Vehicle group). (D) Analysis of lactate release in extracellular matrix after co-culture (n = 3, #p > 0.05, *p < 0.05). (E) pH value in extracellular matrix after co-culture (n = 3, #p > 0.05, *p < 0.05). (F) JC-1 staining of neurons in each group. (G) Red/green fluorescence ratio in each group (n = 3, #p > 0.05, *p < 0.05).

ro of -p re lP na ur Jo

Fig. 5. Knockdown of MCT4 attenuates cognitive deficits in APP/PS1 mice. (A) Representative swimming trajectories of the probe trial in MWM tests. (B) Passing times in the probe trial (mean ± SD, n = 5 per group, #p > 0.05, *p < 0.05). (C) Percentage of time mice spent in the four quadrants during the probe trial of MWM tests (n = 5 per group, #p > 0.05 compared with control (NC) group). (D) TUNEL assay of apoptotic cells in the CA3 region of the hippocampus in three groups of APP/PS1 mice. Bar: 50 mm. (E) Apoptosis rate in (D) (*p < 0.05, #p > 0.05). (F)

Jo

ur

na

lP

re

-p

ro of

Immunofluorescence staining of CD147 and γ-secretase in the hippocampus of APP/PS1 mice (n = 5 per group). Bar: 50 mm.

Jo

ur

na

lP

re

-p

ro of

Supplementary Fig. 1. MWM testing to determine escape latency during the acquisition phase. (A) Escape latency of C57/BL6 and APP/PS1 mice aged 2 and 3 months during the acquisition phase (n = 5 per group, *p < 0.05). (B) Escape latency of the sham group, NC group, and KD-MCT4 group (n = 5 per group, *p < 0.05). Supplementary Fig. 2. Body weight and average swimming speed during the water maze training. (A) Body weight of C57/BL6 and APP/PS1 mice aged 2 and 3 months. (B) Average swimming speed of each group (n = 5 per group). Supplementary Fig. 3. Escape latency and passing times of three APP/PS1 mice groups treated for different durations. (A) Escape latency of three APP/PS1 mice groups after 5 days of treatment. (B) Escape latency of three APP/PS1 mice groups after 10 days of treatment. (C) Passing times in the probe trial of each group after 5 days of treatment. (D) Passing times of each group after 10 days of treatment (n = 5 per group). Supplementary Fig. 4. Representative image of the water tank divided into four equal quadrants.