close

Вход

Забыли?

вход по аккаунту

?

j.neuroscience.2018.08.003

код для вставкиСкачать
Accepted Manuscript
Research Article
The age-dependent elevation of miR-335-3p leads to reduced cholesterol and
impaired memory in brain
Obayed Raihan, Afrina Brishti, Md Rasel Molla, Wenbo Li, Qilun Zhang, Peng
Xu, Muhammad Imran Khan, Juan Zhang, Qiang Liu
PII:
DOI:
Reference:
S0306-4522(18)30541-4
https://doi.org/10.1016/j.neuroscience.2018.08.003
NSC 18596
To appear in:
Neuroscience
Received Date:
Revised Date:
Accepted Date:
28 April 2018
2 August 2018
6 August 2018
Please cite this article as: O. Raihan, A. Brishti, M.R. Molla, W. Li, Q. Zhang, P. Xu, M.I. Khan, J. Zhang, Q. Liu,
The age-dependent elevation of miR-335-3p leads to reduced cholesterol and impaired memory in brain,
Neuroscience (2018), doi: https://doi.org/10.1016/j.neuroscience.2018.08.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.
The age-dependent elevation of miR-335-3p leads to reduced cholesterol and impaired
memory in brain
Obayed Raihana,b,c,d,ǂ , Afrina Brishtia,b,c,d,ǂ, Md Rasel Mollaa,b,c,d, Wenbo Lia,b,c,d, Qilun
Zhanga,b,c,d, Peng Xue, Muhammad Imran Khana,b,c,d, Juan Zhanga,b,c,d* and Qiang
Liua,b,c,d,f*
a. The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, School of
Life Sciences, University of Science and Technology of China, Hefei 230001, China.
b. Hefei National Laboratory for Physical Sciences at the Microscale, University of Science
and Technology of China, Hefei 230026, China.
c. Neurodegenerative Disease Research Center, University of Science and Technology of
China, Hefei 230026, China.
d. CAS Key Laboratory of Brain Function and Disease, University of Science and
Technology of China, Hefei 230026, China.
e. The Liaocheng People’s Hospital, Liaocheng 252000, China
f. National Synchrotron Radiation Laboratory, University of Science and Technology of
China, Hefei, 230029, China.
ǂ These authors contributed equally for this work.
* Correspondence should be addressed to Juan Zhang (zj2014@ustc.edu.cn) or Qiang Liu
(liuq2012@ustc.edu.cn)
1
ABSTRACT
MiR-335-3p, a neuron-enriched microRNA, has been reported to be involved in aging and age
related neurological diseases. However, the role of miR-335-3p in cholesterol metabolism of
astrocytes, and whether it affects neuronal functions, particularly during aging process, largely
remains unknown. In this study, we uncover that miR-335-3p is significantly increased in aged
cultured astrocytes and aged hippocampal brains, accompanied by decreased cellular cholesterol
and diminished expression of HMGCR and HMGCS1, both step-limiting enzymes in cholesterol
synthesis pathway. We also demonstrate that miR-335-3p suppresses HMGCS1 posttranscriptionally by directly binding to its 3’UTR, and HMGCR through binding mediated by
SFRS2. More importantly, aged mice with miR-335-3p deficiency in hippocampal brains exhibit
improved learning and memory, accompanied by enhanced levels of PSD95. We further reveal
that the level change of PSD95 is resulted from altered cholesterol metabolism. Our findings
provide a novel insight into the regulatory role of miR-335-3p in cholesterol metabolism in
astrocytes, and consequently cognitive functions during aging.
Key words: miRNA, astrocytes, cholesterol, aging, memory.
2
INTRODUCTION
MicroRNAs (miRNAs), a class of small non-coding transcripts, are considered to be key
regulators of gene regulation at post-transcriptional level. MiRNAs have been also shown to be
implicated in various pathological diseases, especially neurological diseases such as Alzheimer’s
disease, Huntington’s, autism, multiple sclerosis and Parkinson’s disease (Wang et al., 2014).
Aging is a major risk factor for most neurodegenerative diseases, and numerous miRNAs are
differentially expressed during aging, particularly in aged brains (Smith-Vikos, 2012).
Brain is the most cholesterol-rich organ, it requires more cholesterol than any other part of the
body due to its higher energy demands (Pfrieger, 2003). Cholesterol is not only a major
constituent of the cell membrane, but also an essential component for synapse and dendrite
formation in brain. Cholesterol in brain is primarily produced locally, given that blood brain
barrier (BBB) prevents uptake of lipoproteins from the circulation. Neurons have very limited
capacity of cholesterol synthesis, as they constitutively rely on astrocytes to deliver cholesterol
via lipoproteins. Both 3-hydroxy-3-methylglutaryl-CoA synthase-1 (HMGCS1) and 3-hydroxy3-methylglutaryl-CoA reductase (HMGCR) are the key rate-limiting enzymes in cholesterol
synthesis pathway. HMGCS1 converts acetyl-CoA to (HMG)-CoA, and HMGCR catalyzes
(HMG)-CoA toward mevalonic acid, that is a key step in cholesterol biosynthesis. Indeed,
disruption of cholesterol synthesis in the brain have been associated with aging (de Chaves et al.,
2008) and numerous neurodegenerative diseases (Zhang et al., 2015).
MicroRNAs are shown to be involved in cholesterol regulation and homeostasis (Sacco and
Adeli, 2012; Vickers et al., 2014). Several miRNAs have been reported for their regulatory roles
in cholesterol metabolism (Rotllan and Fernandez-Hernando, 2012), and some of these miRNA
such as miR33 is particularly involved in brain cholesterol metabolism and implicated in related
diseases (Jaouen and Gascon, 2016).
Mi-335 has been reportedly associated with lipid metabolism in liver and white adipose tissues
(Nakanishi et al., 2009b). The differential expression of miR-335-3p has been associated with
aging (Kabir et al., 2016; Bai et al., 2011; Tomé et al., 2014), as well as with neurodegenerative
3
diseases (Cogswell et al., 2008; Denk et al., 2015; Lau et al., 2013) . However, the role of miR335-3p in brain cholesterol metabolism, particularly during aging, largely remains unknown.
In the present study, we discover that miR-335-3p level is significantly enhanced in aged
cultured astrocytes, as well as in hippocampal brains. Gain of function of miR-335-3p leads to a
reduction of cholesterol levels by suppressing HMGCR and HMGCS1 levels posttranscriptionally, both are rate-limiting enzymes in cholesterol synthesis pathway. More
importantly, knockdown of miR-335-3p in hippocampal brain of aged mice results in restored
levels of cholesterol, PSD95 and consequently improved learning and memory. Our findings
reveal that miR-335-3p serves as a novel regulator in cholesterol metabolism in the brain,
providing a potential therapeutic target for age related neurological diseases.
EXPERIMENTAL PROCEDURES
Reagents
anti-HMGCS1 antibody (WB, 1:750, Proteintech); anti-HMGCR antibody (WB, 1:500,
Elabscience); anti-SFRS2 antibody (WB, 1:1000, Proteintech); anti-GAPDH antibody (WB,
1:5000, Proteintech); anti-PSD95 antibody (WB, 1:750, Proteintech); anti-GFAP antibody (IF,
1:200, Sigma); Atorvastatin (Wuhan Yuancheng Co, China); Cholesterol (Invitrogen), anti-rabbit
Alexa Fluor 488/594 (1:400, Molecular probe); AceQ qPCR SYBR Green Master Mix
(Vazyme).
Cell culture
Neuro 2a and HEK293T cells were originally from ATCC and cultured with Dulbercco’s
modified Eagle’s medium (DMEM, Gibco) plus 10% fetal bovine serum (FBS, Gibco) and 1%
penicillin/streptomycin (P/S, Invitrogen) at 37 under 5% CO2. All cell lines were tested
negative for mycoplasma contamination.
Neuron-astrocyte co-cultures
4
Neuron-astrocyte co–cultures were performed as described previously, with modifications (van
Deijk et al., 2017). Cortical brains were harvested from postnatal day 1 (P1) wild-type (WT)
C57B/6J mice, and digested in PBS containing 0.05% trypsin at 37°C for 10 minutes. Astrocytes
were seeded on a poly-D-lysine (Sigma) coated 6-well plate and maintained in DMEM (Gibco)
plus 10% FBS (Gibco), 1% sodium pyruvates (Invitrogen) and 1% penicillin/streptomycin
(Invitrogen), in CO2 incubator at 37°C. When cells reached confluent, astrocyte culture medium
was replaced with neurobasal medium (Gibco) supplemented with B-27 (Gibco), 1%
penicillin/streptomycin (Invitrogen). Cortical neurons were prepared from E17 wild-type mice
according to previously published (Liu et al., 2011), and plated onto the astrocyte monolayer.
Neuron-astrocyte co-cultures (7 DIV) were subjected to lentiviral transduction, followed by
atorvastatin (10 µM/L) or cholesterol (50 µM) treatment.
Plasmid construction
miR-335 precursor sequence was PCR amplified from N2a cell genomic DNA and sub-cloned
into EcoRI/BamHI restriction sites of the pCDH-EF1 lentivector (Addgenes). A short RNA
hairpins (miRZip), that competitively and specifically targets and inhibits miR-335, was subcloned into BamHI/EcoRI restriction sites of pGreenPuro hairpin backbone lentivector (System
Biosciences). H1 promoter and miR-335 antisense was PCR amplified from pGreenPuro vector
and inserted into Eco53KI/HindIII sites of pAAV-EF1a-DIO-Gcamp6s vector (Addgenes),
whereas EF1a promoter was replaced with CamkIIa promoter followed by zsGreen sequence for
neuronal visualization. Full-length mouse SFRS2 was PCR amplified from mouse astrocyte
derived cDNAs and inserted into the EcoRI/BamHI restriction sites of the pCDH-EF1 lentivector
for overexpression. SFRS2 shRNA sequences were annealed and inserted into the BamHI/EcoRI
restriction sites of the H1 promoter driven pGreenPuro backbone lentivector.
Lentiviral production
Lentiviral plasmid, together with packaging (pHR’8.2deltaR) and envelope plasmid (pCMVVSV-G) were transfected into 293T cells. Virus containing medium was harvested 24 h after the
transfection and subjected to ultracentrifuge for virus precipitations. Viral titer was determined
by qPCR.
5
AAV production and stereotaxic injection
293T cells were co-transfected with AAV target plasmid, together with pHelper and helper2/9 in
a ratio of 2:1:1. Transfection was carried out using PEI (Sigma) as a transfection reagent. Viral
titer was determined by qPCR through serial dilutions. Male WT C57BL/6J mice were
stereotaxically injected with adenovirus containing miR-335-3p RNAi or control AAV (0.5 µl,
1×1012 TU/ml ) into the bilateral CA1 area of hippocampus with an air pressure injector system
(KDS). The coordinates used for stereotaxic injections was AP -2.3, ML 2.0, DV -1.5 and AP 2.3, ML -2.0, DV -1.5.
RNA immunoprecipitation
Primary cultured astrocytes (12 DIV) were harvested and lysed in PBS plus 1% Triton-X100
(Sigma), followed by ultra-sonication for 2 min (Ningbo Scientz Biotechnology, China). The
homogenates were incubated on ice for 15 min and then centrifuged at 13,000 rpm for 10 min,
protein concentration of supernatant was determined by BCA Protein Assay (GeneStar). Equal
amount of lysates were incubated with protein A/G beads (BioTools) pre-conjugated with SFRS2
antibody or control IgG. Total RNA was extracted from beads and subjected to qPCR analyses.
RNA extraction, Reverse transcription and qPCR
Total RNA was extracted from cells or tissues using TRIzol (Invitrogen) following the
manufacturer’s instructions. Micro RNA detection was performed as previously described
(Zhang J, 2008), polyA tail was added to 3’ end of RNA using poly A polymerase (NEB). Tailed
RNA was reverse transcribed using HiScript® II Reverse Transcriptase (Vazyme) in the
presence of an anchor RT primer: cgactcgatccagtctcagggtccgaggtattcgatcgagtcgcacttttttttttttv.
Quantitative PCR was performed with AceQTM qPCR SYBR Green Master Mix (Vazyme) on
LightCycler 96 system (Roche) according to standard procedures. The real-time value for each
sample was averaged and compared using the CT method, where the amount of target RNA
(2−∆∆CT) was normalized to an endogenous reference (∆CT) and related to the amount of target
gene in tissues or cells, which was set as the calibrator at 1.0.
Western blot and Densitometry analysis
6
Cells or tissues were lysed in 1% Triton-X 100 containing PBS, plus cocktail protease inhibitors
(BioTool), then subjected to sonication for 2 min and incubation on ice for 15 min. Lysates were
then centrifuged at 14,000 rpm for 10 min, protein level of supernatant was determined using
BCA Protein Assay Kit (GeneStar). Equal amount of proteins were subjected to SDS-PAGE.
The immunoreactive bands were visualized by enhanced chemiluminescence (Advansta) and
detected by ChemiScope (CLiNX). The densitometric analyses of immunoreactive bands were
performed using Image J (National Institutes of Health, Bethesda, MD, USA).
Cholesterol assay
Total cholesterol level was determined using Amplex® Red Enzyme Assays (Invitrogen)
following manufacturer’s instructions.
Luciferase reporter assay
Full-length WT or mutant 3ˊUTR of SFRS2 and HMGCS1 were PCR amplified and inserted into
NheI/XbaI restriction sites of pmirGlo dual luciferase vector (Promega) according to
manufacturer’s protocol. N2a cells were transfected with control or miR-335-3p RNAi, 24 h
later, each luciferase construct was seperately introduced into these cells. Cells were harvested
48 h after the initial transfection, and cell lysates were incubated with luciferase substrate
(Promega). Luciferase activity was measured and data were presented as relative fold change
over control.
Animals
Male C57B/6J mice were purchased from Model Animal Research Center of Nanjing University
and were housed under controlled temperature and light condition (12 hours light/dark). The
mice were kept in standard laboratory mice cage and given ad libitum access to food and water. 2
month (2M) and 24 month (24M) old mice were used to represent young (2M) and aged (24M)
mice respectively. 15 month old (15M) mice were used for AAV mediated shRNA or RNAi
delivery into hippocampal brain. Protocols involving mice were approved by the Institutional
Animal Care and Use Committee at the University of Science and Technology of China.
7
Behavioral assay
Novel object recognition test (NOR)
Object location recognition, also known as the recognition memory test, was performed
according to previously established protocol to assess working memory (Li et al., 2018). During
habituation phase, the mice were first kept in an open acrylic box (30 cm × 20 cm × 30 cm) for
10 min to familiarize with the test apparatus. During the training phase, mice were placed in the
middle of the test box containing two identical cubes, and they were allowed to explore for 10
min. After the training phase, the mice were returned back to their home cages for 1 h. The mice
were again placed in the test chamber, where one cube was replaced with ball, followed by 5 min
exploration. Novel and familial object exploration time were recorded, and cognitive index (CI)
was calculated as CI=[(novel object exploration time)/ (familial object exploration time + novel
object exploration time) × 100].
Morris Water Maze test
The Morris water maze test was performed according to published protocol (Li et al., 2018). The
whole assessment was divided into two phases, training phase of consecutive five days followed
by a probe test on day six. The test was performed in a circular pool with four equal quadrants
(NE, NW, SE and SW), that was filled with opaque water (30 cm deep) at a constant temperature
of 21-22°C. A 12 cm platform was placed in the southwest (SW) quadrant, equidistant from the
center and wall. The mice were given four trials (60 sec each) per day for 5 days during training
phase, and on day six, the hidden platform was removed and mice were placed from quadrant,
opposite to that where the platform had been, and mice were allowed to swim freely for 90 sec.
The probe trial was recorded with a digital video camera placed above the center of the pool and
the behavioral parameters were analyzed using Ethovision XT 11 software (Noldus).
Immunofluorescence staining
Primary cultured astrocytes on cover slip were washed 3 times with PBS, fixed in 4% PFA for 15
min, followed by permeabilization with 0.2% Triton X-100 containing PBS. Cells were
subjected to blocking with 2% BSA in PBST for 30 min. Brain tissues sections mounted on glass
8
slides were washed 3 times in TBS, permeabilized with PBS buffer containing 0.25% Triton X100 and blocked in 5% BSA in PBS. After blocking, both cells and tissue sections were
incubated with anti-GFAP antibody at 4°C overnight, followed by Alexa-conjugated secondary
antibody (anti-rabbit Alexa Fluor 488 for astrocyte and anti-rabbit Alexa Fluor 594 for brain
sections). Fluorescence signals were detected with CCD camera mounted onto U-CB5S
microscope system (Olympus).
Statistical Analysis
Statistical comparisons were performed in GraphPad Prism 5.0, using Student’s t-test or two-way
ANOVA when appropriate. Values were represented as the mean ± standard error of the mean,
and P values <0.05 (indicated by asterisk) were considered significant.
9
RESULTS
MiR-335-3p level is increased during aging in vitro and in vivo, with a corresponding decrease
of cholesterol level.
To examine the age-associated differential expression of miR-335 in astrocytes, we assessed its
levels in young (7D) and aged (35D) primary cultured astrocytes using qPCR. We found that
levels of miR-335-3p, but not miR-335-5p, were significantly enhanced in aged cultured
astrocytes (Fig. 1A). We also observed a similar increase of miR-335-3p in hippocampal brain of
aged mice (24M), when compared to young controls (2M) (Fig. 1B). Cholesterol metabolism is
believed to be altered during natural course of aging, we therefore examined the cholesterol
levels of young and aged astrocytes and brains. We found that cholesterol levels were
significantly decreased both in aged cultured astrocytes (35D) (Fig. 1C) and in hippocampal
brains of aged mice (24M) (Fig. 1D). To determine how cholesterol metabolism was altered, we
measured the levels of HMGCR and HMGCS1, both are essential enzymes in cholesterol
synthesis pathway. We observed a significant decrease of both HMGCR and HMGCS1 levels in
aged cultured astrocytes (35D) (Fig. 1E) and in hippocampal brains from aged mice (24M) (Fig.
1G). The densitometric analyses also supported these notions (Fig. 1F, H). Postsynaptic density
protein 95 (PSD95) is exclusively located in the post synaptic density of neurons and plays an
important role in synaptic plasticity and long-term potentiation. In our study, we found that
PSD95 levels were substantially decreased in the hippocampus of aged mice (24M) (Fig. 1G, H),
indicating that neuronal function was declined in aged mouse. Taken together, these findings
demonstrate that miR-335-3p level is up-regulated during aging, and it may associate with the
disruption of cholesterol metabolism in aged brains.
MiR-335-3p negatively regulates HMGCS1 and HMGCR levels in cholesterol synthesis
pathway, consequently cholesterol level.
We next investigated whether miR-335-3p is involved in cholesterol metabolism. Using
lentiviral delivered antisense miRNA to knockdown miR-335-3p (Fig. 2A), we observed a
substantial elevation of cellular cholesterol in aged cultured astrocytes (35DIV) (Fig. 2D).
HMGCR and HMGCS1 are known as rate-limiting enzymes in cholesterol biosynthesis, whereas
10
SFRS2, a member of SR family proteins, is a multifunctional RNA binding protein and recently
reported to have a regulatory role in cholesterol synthesis and metabolism (Cheng et al., 2016).
Given that the seed region of miR-335-3p base-pairs to binding sites on 3’UTR of HMGCR,
HMGCS1 and SFRS2, as predicted by TargetScan (www.targetscan.org), we accordingly
assessed the levels of HMGCR, HMGCS1 and SFRS2, and found that all of them were
significantly increased in the absence of miR-335-3p (Fig. 2B, C). Moreover, levels of glial
fibrilar acidic protein (GFAP), the astrocyte specific marker, have showed gradual increase
during aging progression (Gross et al., 1991; Nichols et al., 1993). In our study, we found that
levels of GFAP were significantly decreased in cultured astrocytes with miR-335-3p deficiency,
as indicated by immunostaining of GFAP (Fig. 2E). Conversely, to investigate the gain of
function of miR-335-3p, we introduced miR-335-3p into cultured astrocytes using lentiviral
delivery technology (Fig. 2F). We observed that cellular cholesterol was significantly decreased
in miR-335-3p administered cultured astrocytes (Fig. 2I). Levels of HMGCS1, SFRS2 and
HMGCR were also found significantly decreased, as indicated by Western blot and subsequent
densitometric analyses (Fig. 2G, H). Moreover, GFAP was significantly activated in cultured
astrocytes in the presence of miR-335-3p (Fig. 2J). Taken together, these findings demonstrate
that miR-335-3p modulates cholesterol metabolism by altering levels of HMGCS1, SFRS2 and
HMGCR in astrocytes.
MiR-335-3p mediated cholesterol loss results in reduction of PSD95 in neuron and
astrocyte mixed cultures.
Emerging evidence has demonstrated that astrocytes, a major population of glia cells in the brain,
play essential roles in neuronal functions. To determine whether miR-335-3p affects neuronal
function by mediating cholesterol metabolism in astrocytes, we seeded neurons onto the
monolayer of astrocytes with miR-335-3p overexpression to create neuron-astrocyte mixed
cultures (Fig. 3A). Consistent with previous findings, we observed a significant decrease of
cellular cholesterol in miR-335-3p administered mixed cultures (Fig. 3D). In addition, we also
found that levels of HMGCS1, HMGCR and SFRS2 were significantly decreased in those
cultures (Fig. 3B). More importantly, PSD95 levels were found decreased substantially,
11
indicating that neuronal synapse functions were deeply affected (Fig. 3B). The densitometric
analyses also supported those conclusions (Fig. 3C). Taken together, these findings show that
miR-335-3p regulates neuronal synapse plasticity by modulating cholesterol metabolism in
astrocytes.
MiR-335-3p directly binds to 3’UTR of HMGCS1 and SFRS2, and represses their
expressions.
Given that miR-335-3p harbors binding sites to 3’UTR of SFRS2 and HMGCS1 respectively, as
indicated in schematic (Fig. 4A), we next evaluated whether miR-335-3p directly binds to the
3’UTR of SFRS2 and HMGCS1 mRNA through these sequences. We generated luciferase
reporter constructs containing full-length 3’UTR of HMGCS1 or SFRS2, along with their
respective mutant 3’UTR, as indicated in schematic (Fig. 4A) . Constructs containing wild type
or mutant 3’UTR were introduced into control or miR-335-3p administered N2a cells. Luciferase
assay demonstrated that wild type, but not mutant 3’UTR, showed decreased luciferase activities
in miR-335-3p administered cells, when compared to controls (Fig. 4B). More importantly,
mRNA levels of HMGCS1 and SFRS2 were significantly decreased in miR-335-3p
overexpressed cells, as a result of direct binding of miR-335-3p to HMGCS1 and SFRS2 (Fig.
4C). Conversely, miR-335-3p knockdown cells introduced with wild type 3’UTR, showed
significantly increased luciferase activities, whereas miR-335-3p knockdown cells introduced
with mutant 3’UTR showed similar luciferase activities to controls (Fig. 4D). In addition, mRNA
levels of both HMGCS1 and SFRS2 were increased in the absence of miR-335-3p (Fig. 4E).
Similarly, mRNA levels of HMGCS1 and SFRS2 were significantly enhanced in miR-335-3p
deficient cells. In summary, miR-335-3p directly binds to 3’UTR of HMGCS1 and SFRS2, and
suppresses their expressions.
SFRS2 binds to 3’UTR of HMGCR and stabilizes its mRNA level.
As a RNA binding protein, SFRS2 is involved in regulation of splicing, translation and stability
of target mRNA (McFarlane et al., 2015; Qian et al., 2011) through binding of its RNA
recognition motif (RRM) and target mRNA. Moreover, HMGCR mRNA is stabilized by RNA
binding protein (Yu et al., 2013). Given that SFRS2 contains RNA recognition domain, it could
12
base pair with 3’UTR of HMGCR mRNA by direct binding (Fig. 5A). We next investigated
whether miR-335-3p regulates HMGCR levels through SFRS2. Cultured astrocyte lysates were
subjected to immunoprecipitation by SFRS2 antibody, following by qPCR analyses of SFRS2bound RNA. We observed that SFRS2 was successfully immunoprecipitated and detected in
Western blots (Fig. 5A), and HMGCR mRNA was significantly enriched by SFRS2 antibody,
whereas GAPDH showed no significant enrichment, as compared to control IgG (Fig. 5B). To
further confirm that HMGCR is indeed the regulatory target of SFRS2, we assessed HMGCR
levels in SFRS2 administered or knockdown cultured astrocytes. Our findings revealed that both
mRNA and protein levels of HMGCR were significantly decreased in the absence of SFRS2
(Fig. 5C, G), and increased in the presence of SFRS2 (Fig. 5E, H), suggesting that SFRS2
positively regulates HMGCR level. The densitometric analyses also supported these notions
(Fig. 5D, F). Altogether, SFRS2 binds to HMGCR mRNA and alters its levels.
Deficiency of miR-335-3p in hippocampal brain leads to increased cellular cholesterol and
improved spatial learning and memory.
To investigate the role of miR-335-3p in vivo, we knockdown miR-335-3p in the hippocampus
of 15-month old mice (15M) by using AAV delivered miR-335-3p antisense. Three month after
AAV injection, mice were subjected to behavioral tests to assess their learning and memory. In
the Morris water maze test, mice with miR-335-3p deficiency in the hippocampus located the
hidden platform with significantly less travel time during the training phase (n=6 for each group)
(Fig. 6E). During the probe trial, these mice spent considerably less time reaching the target
quadrant (Fig. 6F) and crossed this target quadrant more often, when compared to controls (Fig.
6G). In the novel object recognition test, mice lacking miR-335-3p in hippocampal brains spent
significantly longer time on object exploration in a novel context than controls (Fig. 6H).
Hippocampal brains were then isolated from control and miR-335-3p deficient mice, and
subsequently subjected to RNA, protein and cholesterol assays. We first verified miR-335-3p
levels in hippocampal brains, indicating a successful application of AAV (Fig. 6A). Cholesterol
assay revealed that cholesterol levels were significantly increased in the absence of miR-335-3p
(Fig. 6B). Moreover, protein levels of HMGCS1, HMGCR, SFRS2 and PSD95 were
significantly elevated in miR-335-3p deficient hippocampal brains (Fig. 6C). This was also
13
evident by densitometric analyses (Fig. 6D). Consistent with previous immunostaining of GFAP
in cultured cells, significant astrocyte deactivation was observed upon knockdown of miR-3353p in the hippocampus of mouse brain (Fig. 6I). In summary, these findings demonstrate that
depletion of miR-335-3p improves learning and memory in aged mice, primarily by modulating
levels of HMGCS1, HMGCR, SFRS2 and subsequently cholesterol metabolism.
Atorvastatin reduces the elevated PSD95 levels led by miR-335-3p deficiency, whereas
supplementation of cholesterol restores the decreased PSD95 levels caused by miR-335-3p
overexpression.
To further investigate whether the regulatory role of miR-335-3p in neuronal synapse functions
is cholesterol dependent, we next either depleted cholesterol in cells with miR-335-3p depletion,
or supplied cholesterol to cells with miR-335-3p administration. Neuron-astrocyte co-cultures
were subjected to miR-335-3p knockdown, followed by treatment of HMGCR inhibitor
atorvastatin to deplete cholesterol. We found that atorvastatin reversed the enhanced PSD95
level led by miR-335-3p deficiency to the similar level of controls (Fig. 7A, B). Conversely,
when cholesterol was supplied to miR-335-3p administered neuron-astrocyte co-cultures, PSD95
levels were restored to the similar levels of controls (Fig. 7C, D). These results suggest that
functions of miR-335-3p in neuronal synapses are highly dependent on cholesterol metabolism.
DISCUSSION
Astrocytes are the most abundant glial subtype in the brain. The level of GFAP, an astrocyte
marker, increases progressively during natural aging course, in addition, astrocytes also become
persistently activated upon aging (Goss et al., 1991; Nichols et al., 1993). These observations are
believed coming from the enhanced load of oxidative damage and inflammatory processes, that
occurs throughout the brain during aging (Morgan et al., 1997). The gene expression pattern of
aging astrocytes partially overlap with reactive astrocytes, that were induced by injuries,
indicating a mild to moderate reactive state occurs in aging astrocytes (Boisvert et al., 2018).
Accordingly, the increase of GFAP is considered as one of the early responses to metabolic and
oxidative stress in aged brains (Bellaver et al., 2017; Clarke et al., 2018). In this study, we
14
identify miR-335, whose levels are substantially elevated in aged cultured astrocytes and
hippocampal brains. Administration of miR-335 in either cultured astrocytes or hippocampal
brains leads to a significant activation of astrocytes.
Aging brains gradually exhibit cognitive decline, accompanied by reduced neuronal synaptic
functions. Astrocytes also serve as key regulators in structural and functional integrity of neurons
over the lifespan (Halassa et al., 2014). Enhanced GFAP during aging is regarded to be
associated with declined synaptic functions (Lefrancois et al., 1997). Astrocytes are the primary
source of cholesterol synthesis in the brain, and neurons constitutively rely on astrocytes for
cholesterol supplying. Lacking of cholesterol production in astrocytes may lead to decreased
synaptic functions, indicating astrocyte-derived cholesterol is essential for neuronal synapse
function (Mauch et al., 2001; van Deijk et al., 2017). Interestingly, levels of astrocytes
homeostasis genes mostly remain unaltered during aging, however, the capacity of cholesterol
synthesis were significantly down-regulated in aging astrocytes (Boisvert et al., 2018), this could
explain the general decreased cholesterol level in aging brain. Moreover, the regulation of
cholesterol metabolism in astrocytes and how it affects neuronal synaptic functions mostly
remains unknown.
Cholesterol is an essential component in physiological and pathological conditions. Disorders of
cholesterol metabolism has been linked to multiple neurological diseases, such as Alzheimer’s
disease (Yoon et al., 2016), Autism spectrum disorders (ASDs) (Cartocci et al., 2018). HMGCR
is the key and rate-limiting enzyme of cholesterol biosynthesis process, also one of the most
intensively studied proteins. In addition to short-term regulation, achieved by altering enzyme
activities through phosphorylation/dephosphorylation cycles, HMGCR is also subjected to longterm regulation through transcriptional and post-transcriptional regulations (Xu et al., 2005). The
regulation of HMGCR is well characterized in peripheral tissue such as liver, however, little is
known in CNS. Cholesterol biosynthesis was modulated differently in various brain regions,
implying cholesterol homoeostasis is precisely maintained to meet the regional needs (Segatto et
al., 2012; Segatto et al., 2013). In our study, we focus on hippocampal brain, a brain area critical
for learning and memory. It is also one of the most vulnerable brain regions to damages.
MicroRNAs (miRNAs) are small endogenous non-coding RNAs (20-22 nucleotide long). Its
typical function is to repress translation or promote degradation of target mRNA, by binding to
15
the specific region of 3' untranslated region of target mRNAs, controlling a variety of biological
processes (Bartel et al., 2009). Mounting research has showed that miRNAs play regulatory roles
in cholesterol/lipid metabolism of peripheral tissues (Fernandez-Hernando et al., 2011), some
evidence has also demonstrated that miRNAs serve as key regulators in brain cholesterol
metabolism. However, most the studies focus on lipid transporting protein like ApoE and
ABCA1, little is known for cholesterol synthesis genes (Yoon et al., 2016). Here, in our study,
we demonstrate that miR-335-3p post-transcriptionally inhibits HMGCS1 and HMGCR levels in
cholesterol biosynthesis pathway, consequently reduces cholesterol levels.
Levels of miR-335-3p have been found significantly increased in ASD by high-throughput
sequencing (Hicks et al., 2016), and other studies have also reported that altered brain cholesterol
metabolism is also observed in ASD rats (Cartocci et al., 2018). All the evidence leads us to
study the regulatory role of miR-335 in cholesterol metabolism. More importantly, miR-335 is
tightly related to cholesterol metabolism in liver (Nakanishi et al., 2009), however, the role of
miR-335 in brain cholesterol metabolism has not been established yet.
Atorvastatin treated mice exhibited impaired cognitive functions, accompanied by significant
decrease of post-synaptic density protein-95 (PSD95) (Schilling et al., 2014), indicating that
cholesterol metabolism is directly linked to synapse functions. In this study, we found that miR335-3p decreases PSD95 levels in neuron-astrocyte co-cultures, as well as in hippocampal brains.
Reduction of miR-335 in hippocampal brains partially restores the declined cognitive functions
in aged mice. Our findings elucidate a novel function of miR-335-3p in cholesterol mediated
regulation of synaptic function and memory in aging.
The observed decrease of PSD95 led by miR-335 is very likely to be an indirect effect, given that
miR-335 contains no direct binding sites to PSD95 mRNA, therefore we speculated that the
regulation is probably through other targets, such as cholesterol. To test this hypothesis, we
supplemented cholesterol to miR-335-3p administered neuron-astrocyte co-cultures and observed
that PSD95 levels were successfully restored. Conversely, depletion of cholesterol in miR-335
deficient co-cultures led to reduction of PSD95 to control levels, indicating that cholesterol
levels were tightly linked to PSD95 levels. Consistently, mounting evidence has suggested that
cholesterol in general has a beneficial effect on aging or neurodegenerative diseases (Ferris et al.,
2017; Martin et al., 2014; Silverman and Schmeidler, 2018).
16
Our results elucidate a novel function of miR-335-3p in cholesterol metabolism, and
subsequently cholesterol mediated regulation of synaptic function and memory in aging. Our
study provides a novel way to understand cholesterol metabolism in aging.
CONCLUSIONS
MiR-335-3p has been implicated in aging and age-related neurological diseases. However, the
role of miR-335-3p in disrupted cholesterol metabolism during aging remains largely unknown.
In our current study, we demonstrate that miR-335-3p suppresses cholesterol by inhibiting
expressions of HMGCS1 and HMGCR in astrocytes. This, in turn, affects neuronal synapse
functions. Reduction of miR-335-3p in aged hippocampal brains leads to rescued cognitive
impairment and neuronal synaptic functions during aging.
ACKNOWLEDGEMENTS
We thank Dr. Riaz Khan, Dr. Ihtisham Bukhari, Md Shafayat Hossain and Elsayed Metwally for
their critical reading on this manuscript. This research was supported by the National Natural
Science Foundation of China grants 81422015, 91332111, 31371087 (Q.L.) and 81601221 (J.Z.),
Key Research Program of Frontier Sciences, CAS, QYZDB-SSW-SMC035 (Q.L.), The Strategic
Priority Research Program, CAS, XDPB10 (Q.L.), CAS-TWAS President’s Fellowship Program.
AUTHOR CONTRIBUTIONS
Q.L. supervised the project; O.R., and Q.L. conceived and designed the study; O.R. and A.B.
performed experiments; O.R., M.R.M., QL.Z, WB.L, J.Z and A.B. analyzed data; O.R., J.Z. and
Q.L. wrote the manuscript.
17
REFERENCES
Bai, X.-Y., Ma, Y., Ding, R., Fu, B., Shi, S., and Chen, X.-M. (2011). miR-335 and miR-34a
Promote Renal Senescence by Suppressing Mitochondrial Antioxidative Enzymes. Journal
of the American Society of Nephrology : JASN 22, 1252-1261.
Bartel, D.P. (2009). MicroRNA Target Recognition and Regulatory Functions. Cell 136, 215233.
Bellaver, B., Souza, D.G., Souza, D.O., and Quincozes-Santos, A. (2017). Hippocampal
astrocyte cultures from adult and aged rats reproduce changes in glial functionality
observed in the aging brain. Molecular neurobiology 54, 2969-2985.
Boisvert, M.M., Erikson, G.A., Shokhirev, M.N., and Allen, N.J. (2018). The Aging Astrocyte
Transcriptome from Multiple Regions of the Mouse Brain. Cell reports 22, 269-285.
Cartocci, V., Catallo, M., Tempestilli, M., Segatto, M., Pfrieger, F.W., Bronzuoli, M.R., Scuderi,
C.,
Servadio,
M.,
Trezza,
V.,
and
Pallottini,
V.
(2018).
Altered
Brain
Cholesterol/Isoprenoid Metabolism in a Rat Model of Autism Spectrum Disorders.
Neuroscience 372, 27-37.
Cheng, Y., Luo, C., Wu, W., Xie, Z., Fu, X., and Feng, Y. (2016). Liver-specific deletion of
SRSF2 caused acute liver failure and early death in mice. Molecular and cellular biology
36, 1628-1638.
Clarke, L.E., Liddelow, S.A., Chakraborty, C., Münch, A.E., Heiman, M., and Barres, B.A.
(2018). Normal aging induces A1-like astrocyte reactivity. Proceedings of the National
Academy of Sciences 115, E1896-E1905.
Cogswell, J.P., Ward, J., Taylor, I.A., Waters, M., Shi, Y., Cannon, B., Kelnar, K., Kemppainen,
J., Brown, D., Chen, C., et al. (2008). Identification of miRNA changes in Alzheimer's
disease brain and CSF yields putative biomarkers and insights into disease pathways.
Journal of Alzheimer's Disease 14.
de Chaves, E.P., and Narayanaswami, V. (2008). Apolipoprotein E and cholesterol in aging and
disease in the brain. Future lipidology 3, 505-530.
18
Denk, J., Boelmans, K., Siegismund, C., Lassner, D., Arlt, S., and Jahn, H. (2015). MicroRNA
Profiling of CSF Reveals Potential Biomarkers to Detect Alzheimer`s Disease. PLOS ONE
10, e0126423.
Fernández-Hernando, C., Suárez, Y., Rayner, K.J., and Moore, K.J. (2011). MicroRNAs in lipid
metabolism. Current opinion in lipidology 22, 86-92.
Ferris, H.A., Perry, R.J., Moreira, G.V., Shulman, G.I., Horton, J.D., and Kahn, C.R. (2017).
Loss of astrocyte cholesterol synthesis disrupts neuronal function and alters whole-body
metabolism. Proceedings of the National Academy of Sciences 114, 1189.
Goss, J.R., Finch, C.E., and Morgan, D.G. (1991). Age-related changes in glial fibrillary acidic
protein mRNA in the mouse brain. Neurobiology of Aging 12, 165-170.
Halassa, M.M., D'Ascenzo, M., Boccaccio, A., and Fellin, T. (2014). Chapter 12 - Astrocytic
Regulation of Synapses, Neuronal Networks, and Behavior A2 - Faingold, Carl L. In
Neuronal Networks in Brain Function, CNS Disorders, and Therapeutics, H. Blumenfeld,
ed. (San Diego: Academic Press), pp. 157-165.
Hicks, S.D., and Middleton, F.A. (2016). A Comparative Review of microRNA Expression
Patterns in Autism Spectrum Disorder. Frontiers in Psychiatry 7, 176.
Jaouen, F., and Gascon, E. (2016). Understanding the Role of miR-33 in Brain Lipid
Metabolism: Implications for Alzheimer's Disease. Journal of Neuroscience 36, 25582560.
Kabir, T.D., Leigh, R.J., Tasena, H., Mellone, M., Coletta, R.D., Parkinson, E.K., Prime, S.S.,
Thomas, G.J., Paterson, I.C., Zhou, D., et al. (2016). A miR-335/COX-2/PTEN axis
regulates the secretory phenotype of senescent cancer-associated fibroblasts. Aging
(Albany NY) 8, 1608-1624.
Lau, P., Bossers, K., Janky, R.s., Salta, E., Frigerio, C.S., Barbash, S., Rothman, R., Sierksma,
A.S.R., Thathiah, A., Greenberg, D., et al. (2013). Alteration of the microRNA network
during the progression of Alzheimer's disease. EMBO Molecular Medicine 5, 1613-1634.
Lefrançois, T., Fages, C., Peschanski, M., and Tardy, M. (1997). Neuritic Outgrowth Associated
with Astroglial Phenotypic Changes Induced by Antisense Glial Fibrillary Acidic Protein
(GFAP) mRNA in Injured Neuron–Astrocyte Cocultures. The Journal of Neuroscience 17,
4121.
19
Li, D., Zhang, J., Wang, M., Li, X., Gong, H., Tang, H., Chen, L., Wan, L., and Liu, Q. (2018).
Activity dependent LoNA regulates translation by coordinating rRNA transcription and
methylation. Nature Communications 9, 1726.
Liu, Q., Zhang, J., Zerbinatti, C., Zhan, Y., Kolber, B.J., Herz, J., Muglia, L.J., and Bu, G.
(2011). Lipoprotein Receptor LRP1 Regulates Leptin Signaling and Energy Homeostasis
in the Adult Central Nervous System. PLOS Biology 9, e1000575.
Martin, M.G., Ahmed, T., Korovaichuk, A., Venero, C., Menchón, S.A., Salas, I., Munck, S.,
Herreras, O., Balschun, D., and Dotti, C.G. (2014). Constitutive hippocampal cholesterol
loss underlies poor cognition in old rodents. EMBO molecular medicine, e201303711.
Mauch, D.H., Nägler, K., Schumacher, S., Göritz, C., Müller, E.-C., Otto, A., and Pfrieger, F.W.
(2001). CNS Synaptogenesis Promoted by Glia-Derived Cholesterol. Science 294, 1354.
McFarlane, M., MacDonald, A.I., Stevenson, A., and Graham, S.V. (2015). Human
papillomavirus 16 oncoprotein expression is controlled by the cellular splicing factor
SRSF2 (SC35). Journal of virology 89, 5276-5287.
Morgan, T.E., Rozovsky, I., Goldsmith, S.K., Stone, D.J., Yoshida, T., and Finch, C.E. (1997).
Increased Transcription of the Astrocyte Gene GFAP During Middle-Age is Attenuated by
Food Restriction: Implications for the Role of Oxidative Stress. Free Radical Biology and
Medicine 23, 524-528.
Nakanishi, N., Nakagawa, Y., Tokushige, N., Aoki, N., Matsuzaka, T., Ishii, K., Yahagi, N.,
Kobayashi, K., Yatoh, S., and Takahashi, A. (2009). The up-regulation of microRNA-335
is associated with lipid metabolism in liver and white adipose tissue of genetically obese
mice. Biochemical and biophysical research communications 385, 492-496.
Nichols, N.R., Day, J.R., Laping, N.J., Johnson, S.A., and Finch, C.E. (1993). GFAP mRNA
increases with age in rat and human brain. Neurobiology of Aging 14, 421-429.
Pfrieger, F.W. (2003). Role of cholesterol in synapse formation and function. Biochimica et
Biophysica Acta (BBA)-Biomembranes 1610, 271-280.
Qian, W., Iqbal, K., Grundke-Iqbal, I., Gong, C.-X., and Liu, F. (2011). Splicing factor SC35
promotes tau expression through stabilization of its mRNA. FEBS letters 585, 875-880.
Rotllan, N., Fern, #xe1, and ndez-Hernando, C. (2012). MicroRNA Regulation of Cholesterol
Metabolism. Cholesterol 2012, 8.
20
Sacco, J., and Adeli, K. (2012). MicroRNAs: emerging roles in lipid and lipoprotein metabolism.
Current Opinion in Lipidology 23, 220-225.
Schilling, J.M., Cui, W., Godoy, J.C., Risbrough, V.B., Niesman, I.R., Roth, D.M., Patel, P.M.,
Drummond, J.C., Patel, H.H., Zemljic-Harpf, A.E., et al. (2014). Long-term atorvastatin
treatment leads to alterations in behavior, cognition, and hippocampal biochemistry.
Behavioural brain research 267, 6-11.
Segatto, M., Di Giovanni, A., Marino, M., and Pallottini, V. (2012a). Analysis of the protein
network of cholesterol homeostasis in different brain regions: An age and sex dependent
perspective. Journal of Cellular Physiology 228, 1561-1567.
Segatto, M., Trapani, L., Lecis, C., and Pallottini, V. (2012b). Regulation of cholesterol
biosynthetic pathway in different regions of the rat central nervous system. Acta
Physiologica 206, 62-71.
Silverman, J.M., and Schmeidler, J. (2018). Outcome age-based prediction of successful
cognitive aging by total cholesterol. Alzheimer's & Dementia.
Smith-Vikos, T., and Slack, F.J. (2012). MicroRNAs and their roles in aging. Journal of Cell
Science 125, 7-17.
Tomé, M., Sepúlveda, J.C., Delgado, M., Andrades, J.A., Campisi, J., González, M.A., and
Bernad, A. (2014). Mir-335 Correlates with Senescence/Aging in Human Mesenchymal
Stem Cells and Inhibits their Therapeutic Actions through Inhibition of AP-1 Activity.
Stem cells (Dayton, Ohio) 32, 2229-2244.
van Deijk, A.-L.F., Broersen, L.M., Verkuyl, J.M., Smit, A.B., and Verheijen, M.H. (2017). High
Content Analysis of Hippocampal Neuron-Astrocyte Co-cultures Shows a Positive Effect
of Fortasyn Connect on Neuronal Survival and Postsynaptic Maturation. Frontiers in
neuroscience 11, 440.
van Deijk Anne-Lieke, F., Camargo, N., Timmerman, J., Heistek, T., Brouwers Jos, F.,
Mogavero, F., Mansvelder Huibert, D., Smit August, B., and Verheijen Mark, H.G. (2017).
Astrocyte lipid metabolism is critical for synapse development and function in vivo. Glia
65, 670-682.
Vickers, K.C., Landstreet, S.R., Levin, M.G., Shoucri, B.M., Toth, C.L., Taylor, R.C.,
Palmisano, B.T., Tabet, F., Cui, H.L., and Rye, K.-A. (2014). MicroRNA-223 coordinates
21
cholesterol homeostasis. Proceedings of the National Academy of Sciences 111, 1451814523.
Wang, C., Ji, B., Cheng, B., Chen, J., and Bai, B. (2014). Neuroprotection of microRNA in
neurological disorders. Biomedical reports 2, 611-619.
Xu, F., Rychnovsky, S.D., Belani, J.D., Hobbs, H.H., Cohen, J.C., and Rawson, R.B. (2005).
Dual roles for cholesterol in mammalian cells. Proceedings of the National Academy of
Sciences of the United States of America 102, 14551.
Yoon, H., Flores, L.F., and Kim, J. (2016). MicroRNAs in brain cholesterol metabolism and their
implications for Alzheimer's disease. Biochimica et Biophysica Acta (BBA)-Molecular and
Cell Biology of Lipids 1861, 2139-2147.
Yu, C.-Y., Theusch, E., Lo, K., Mangravite, L.M., Naidoo, D., Kutilova, M., and Medina, M.W.
(2013). HNRNPA1 regulates HMGCR alternative splicing and modulates cellular
cholesterol metabolism. Human molecular genetics 23, 319-332.
Zhang, J., Du, Y.-y., Lin, Y.-f., Chen, Y.-t., Yang, L., Wang, H.-j., and Ma, D. (2008). The cell
growth suppressor, mir-126, targets IRS-1. Biochemical and Biophysical Research
Communications 377, 136-140.
Zhang, J., and Liu, Q. (2015). Cholesterol metabolism and homeostasis in the brain. Protein &
cell 6, 254-264.
22
FIGURE LEGENDS
Fig. 1. MiR-335-3p level is increased during aging in vitro and in vivo, with a corresponding
increase of cholesterol level. (A and B) Levels of miR-335-3p, but not miR-335-5p, were
significantly increased in aged cultured astrocytes and aged hippocampal brain. Levels of miR335-3p and miR-335-5p in young (7DIV) and aged (35DIV) primary culture astrocytes (A), in
young (2M) and aged (24M) hippocampal brain (B) were determined by qPCR analyses. Data
were normalized to U6 and presented as fold difference of aged over young. n=3 independent
culture per group for astrocytes and n=5 mice per age group. (C and D), Levels of cholesterol in
young (7DIV) and aged (35DIV) primary culture astrocytes (n=3) (C), in young (2M) and aged
(24M) hippocampal brain (n=3) (D). Cholesterol levels were normalized to protein levels and
plotted as the relative level of aged over young. (E-F) Levels of HMGCS1 and HMGCR in
young (7DIV) and aged (35DIV) primary culture astrocytes (n=3), were determined by Western
blot (E) and densitometric analysis (F). GAPDH was included as input controls. (G-H) Levels of
HMGCS1, HMGCR and PSD95 in young (2M) and aged (24M) hippocampal brain (n=3), were
determined by Western blot (G) and densitometric analysis (H). GAPDH was included as input
controls. *P<0.05; **P<0.01; ***P<0.001 by Student’s t test; error bars denote means ± SEM. In
this and subsequent figures, M denotes month, DIV denote day of in vitro culture.
Fig. 2. MiR-335-3p negatively regulates levels of HMGCS1 and HMGCR in cholesterol
synthesis pathway. (A) Levels of mir-335-3p in lentiviral mediated miR-335-3p deficient
astrocytes (35DIV) (n=3) were determined by qPCR. (B and C) Protein levels of HMGCS1,
HMGCR and SFRS2 in control and miR-335-3p deficient cells were measured by Western blot
(B) and densitometric analysis (C) (n=3). GAPDH was included as loading controls. (D)
Cholesterol levels were normalized to protein levels and plotted as the relative level of miR-3353p knock down over control cells (n=3). (E) Representative immunofluorescence images of
23
GFAP (green) in control and astrocytes lacking miR-335-3p. Scale bar, 20 µm. (F) Levels of
mir-335-3p in lentiviral mediated miR-335-3p overexpressed astrocytes (n=3) were determined
by qPCR. (G and H) Protein levels of HMGCS1, HMGCR and SFRS2 in control and miR-3353p administered cultured astrocytes were determined by Western blot (G) and densitometric
analysis (H) (n=3). GAPDH was included as loading controls. (I) Cholesterol levels were
normalized to protein levels and plotted as the relative level of miR-335-3p overexpression over
control cells (n=3). (J) Representative immunofluorescence images of GFAP (green) in control
and astrocytes with miR-335-3p overexpression. Scale bar, 20µm.
*
P<0.05; **P<0.01;
***P<0.001 by Student’s t test;; error bars denote means ± SEM.
Fig. 3. Loss of cholesterol level mediated by miR-335-3p results in reduction of PSD95 in
neuron and astrocyte mixed cultured cells. (A) Levels of mir-335-3p in lentiviral mediated
miR-335-3p administered neuron-astrocyte co-cultured cells , as determined by qPCR (n=3).
Data were presented as relative fold change of overexpression over controls. (B and C) Protein
levels of HMGCS1, SFRS2, HMGCR and PSD95 in control and miR-335-3p overexpressed
neuron-astrocyte co-cultured cells, as measured by Western blot (B) and densitometric analysis
(C) (n=3). GAPDH was included as input controls. (D) Cellular cholesterol levels in neuronastrocyte co-cultures, data were plotted as the relative level of miR-335-3p overexpression over
controls (n=3). *P<0.05; **P<0.01; ***P<0.001 by Student’s t test;; error bars denote means ±
SEM.
Fig. 4. MiR-335-3p directly binds to 3’UTR of HMGCS1 and SFRS2 and represses their
expression levels. (A) Schematic of miR-335-3p seed sequence, predicted binding sequences on
the 3’ UTR of HMGCS1 and SFRS2 mRNA (highlighted in blue), binding sequences containing
mutations (highlighted in red). (B) 3’UTR of HMGCS1, 3’UTR of HMGCS1 mutant, 3’UTR of
SFRS2 and 3’UTR of SFRS2 mutant variants were individually cloned into pmirGlo luciferase
vector and introduced into N2a cells with miR-335-3p overexpression. Binding affinities were
determined by luciferase assay. Luciferase activity values were presented as relative level to
controls with vector only (n=3). (C) mRNA levels of HMGCS1 and SFRS2 in the presence of
miR-335-3p, as determined by qPCR.. (D) 3’UTR of HMGCS1, 3’UTR of HMGCS1 mutant,
24
3’UTR of SFRS2 and 3’UTR of SFRS2 mutant in pmirGlo luciferase vector were introduced
into N2a cells with miR-335-3p deficiency. Binding affinities were determined by luciferase
assay. Luciferase activity values were presented as relative level to controls (n=3). (E) mRNA
levels of HMGCS1 and SFRS2 in the absence of miR-335-3p, as determined by qPCR (n=3).
*P<0.05; **P<0.01; ***P<0.001 by ANOVA or Student’s t test;; error bars denote means ±
SEM.
Fig. 5. SFRS2 binds to 3’UTR of HMGCR and stabilizes HMGCR mRNA. (A) Primary
astrocyte lysates were subjected to immunoprecipitation with an SFRS2 antibody or control IgG,
following by immunoblotting with SFRS2 antibody. SFRS2 RRM sites on 3’UTR of HMGCR
(down) (B) Total RNA was isolated and subjected to qPCR analyses. Data are shown as the
enrichment by SFRS11 antibody over control IgG (n=3). GAPDH was included as a negative
control. (C-F) Protein levels of SFRS2 and HMGCR in SFRS2 deficient (C-D) and SFRS2
overexpressed cells (E-F), as determined by Western blot (C, E) and densitometric analysis (D,
F) (n=3). GAPDH was used as input controls. (G-H) mRNA levels of SFRS2 and HMGCR in
SFRS2 deficient (G) and SFRS2 overexpressed cells (H), as determined by qPCR (n=3).
*P<0.05; **P<0.01; ***P<0.001 by Student’s t test;; error bars denote means ± SEM.
Fig. 6. Deficiency of miR-335-3p in hippocampal brain leads to increased cellular
cholesterol level and improved spatial learning and memory. (A-I) AAV mediated miR-3353p knockdown in hippocampal brain of aged mice (15M). (A) Levels of miR-335-3p in miR335-3p deficient mice, data presented as relative fold change of knockdown over controls, n=6
mice per group. (B) cellular cholesterol levels in hippocampal brains of injected mice, data were
plotted as the relative levels of miR-335-3p knockdown over controls (n=3). (C-D) Protein levels
of HMGCS1, SFRS2, HMGCR, and PSD95 in control and miR-335-3p knockdown hippocampal
brains, as determined by Western blot (C) and densitometric analyses (D) (n=3). GAPDH was
used as input controls. (E-G) In the Morris water maze test, escape latency during the training
phase of MWM (E), along with latency to reach the platform first (F) and the number of platform
crossing were recorded during the probe trail of MWM task (G) (n=6, per control and mir-3553p hippocampal knockdown mice). (H) Relative cognitive index in Novel Object Recognition
25
task was calculated and plotted as relative level of knockdown over controls. (n=6, per each
group). (I) Representative immunofluorescence images of GFAP (red) in control or and mir-3553p knockdown hippocampal brain. Scale bar, 20 µm. *P<0.05; **P<0.01; ***P<0.001 by
Student’s t test;; error bars denote means ± SEM.
Fig. 7. Atorvastatin reduces the elevated PSD95 levels led by miR-335-3p deficiency,
whereas supplementation of cholesterol restores the decreased PSD95 levels caused by
miR-335-3p overexpression. (A-B) Protein levels of PSD95 in control or miR-335-3p deficient
co-cultures with or without Atorvastatin (10µM) treatment, as determined by Western blot (A)
and densitometric analyses (B) (n=3). GAPDH was included as input controls. (C-D) Protein
levels of PSD95 in control or miR-335-3p overexpressed co-cultures with or without cholesterol
(50µM) treatment, as determined by Western blot (C) and densitometric analyses (D) (n=3).
GAPDH was included as input controls. *P<0.05; **P<0.01; ***P<0.001 by ANOVA or
Student’s t test;; error bars denote means ± SEM.
26
27
28
29
30
31
32
33
Table 1. qPCR detection primers
TARGET
PRIMER
SEQUENCE (5' to 3')
miR-335-5p
F
UCAAGAGCAAUAACGAAAAAUGU
miR-335-3p
F
UUUUUCAUUAUUGCUCCUGACC
Universal Reverse R
Primer for
miRNA detection
CCAGTCTCAGGGTCCGAGGTATTC
F
CTCGCTTCGGCAGCACA
R
AACGCTTCACGAATTTGCGT
F
AGGTCGGTGTGAACGGATTTG
R
GGGGTCGTTGATGGCAACA
F
AGAGCGAGTGCATTAGCAAAG
R
GATTGCCATTCCACGAGCTAT
F
CGGATCGTGAAGACATCAACTC
R
CGCCCAATGCAATCATAGGAA
F
GCGCTCCAGATCAACCTCC
R
CTTGGACTCTCGCTTCGACAC
F
TCTGTGCGAGAGGTAGCAGA
R
AAGCACTCCGTGAACTCCTG
U6
GAPDH
HMGCR
HMGCS1
SFRS2
PSD95
shRNA sequences (target sequences highlighted red)
)
SFRS2
miR-3353p
Top
strand
Bottom
strand
Top
strand
GATCCCCAAGTCCAGATCTGCCCGAACTTCCTGTCAGAT
TCGGGCAGATCT GGACTTGGTTTTTG
AATTCAAAAACCAAGTCCAGATCTGCCCGAATCTGACAG
GAAGTTCGGGCAGATCTGGACTTGGG
GATCCTTTTTCAGTATTGCTCCGGAACCTTCCTGTCAGAGGT
CAGGAGCAATAATGAAAAATTTTTG
34
Bottom
strand
AATTCAAAAATTTTTCATTATTGCTCCTGACCTCTGACAGGA
AGGTTCCGGAGCAATACTGAAAAAG
Primers for cloning
TARGET
pCDH-miR-3353p
pCDH-SFRS2
PRIMER
F
R
F
R
SEQUENCE (5' to 3')
CGGAATTCCTAGAATTGTGCCTGGTAGT
CGGGATCCCGATGGGCTAATAGTGATGG
CGGAATTCCGGCCACCCCGCTTCGCAGCCATGA
CGGGATCCCGCCGCTTGCCGATTCATCA
35
36
Highlights
MiR-335-3p is enriched in aged cultured astrocytes and hippocampal brain of aged
mice.
MiR-335-3p reduces cellular cholesterol by suppressing HMGCS1 and HMGCR, key
enzymes in cholesterol biosynthesis pathway.
MiR-335-3p deficiency leads to rescued PSD95 level and cognitive impairment in aged
mice, in a cholesterol dependent manner.
37
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 242 Кб
Теги
003, neuroscience, 2018
1/--страниц
Пожаловаться на содержимое документа