Phasuketal. Mol Brain (2021) 14:42
Enhanced contextual fear memory
inperoxiredoxin 6 knockout mice isassociated
withhyperactivation ofMAPK signaling
Sarayut Phasuk1,2, Tanita Pairojana1, Pavithra Suresh1, Chee‑Hing Yang3, Sittiruk Roytrakul4, Shun‑Ping Huang6,
Chien‑Chang Chen5, Narawut Pakaprot2, Supin Chompoopong7, Sutisa Nudmamud‑Thanoi8,9
and Ingrid Y. Liu1*
Fear dysregulation is one of the symptoms found in post‑traumatic stress disorder (PTSD) patients. The functional
abnormality of the hippocampus is known to be implicated in the development of such pathology. Peroxiredoxin 6
(PRDX6) belongs to the peroxiredoxin family. This antioxidant enzyme is expressed throughout the brain, including
the hippocampus. Recent evidence reveals that PRDX6 plays an important role in redox regulation and the modula‑
tion of several signaling molecules involved in fear regulation. Thus, we hypothesized that PRDX6 plays a role in the
regulation of fear memory. We subjected a systemic Prdx6 knockout (Prdx6−/−) mice to trace fear conditioning and
observed enhanced fear response after training. Intraventricular injection of lentivirus‑carried mouse Prdx6 into the
3rd ventricle reduced the enhanced fear response in these knockout mice. Proteomic analysis followed by valida‑
tion of western blot analysis revealed that several proteins in the MAPK pathway, such as NTRK2, AKT, and phospho‑
ERK1/2, cPLA2 were signiﬁcantly upregulated in the hippocampus of Prdx6−/− mice during the retrieval stage of
contextual fear memory. The distribution of PRDX6 found in the astrocytes was also observed throughout the hip‑
pocampus. This study identiﬁes PRDX6 as a participant in the regulation of fear response. It suggests that PRDX6 and
related molecules may have important implications for understanding fear‑dysregulation associated disorders like
Keywords: Peroxiredoxin 6, Trace fear conditioning, Fear memory, Posttraumatic stress disorder, MAPK signaling
© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat iveco
mmons .org/publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Fear acquisition and expression to threatening stimuli are
innate responses to avoid dangers or predators to ensure
safety and survival [1, 2]. Several pieces of evidence sug-
gest that brain regions, including the amygdala, medial
prefrontal cortex, and hippocampus, are required for an
appropriate level of fear response [3–5]. Dysregulation of
these brain regions leads to an excessive fear response in
post-traumatic stress disorder (PTSD) . e underly-
ing molecular mechanism is still unclear. Peroxiredoxin
6 (PRDX6) is a multifunctional enzyme belonging to the
peroxiredoxin superfamily . Among the peroxiredoxin
superfamily, PRDX6 is the only member that displays
multiple functions, including the glutathione peroxidase
(GPx), acidic calcium-independent phospholipase A2
(aiPLA2), and lysophosphatidylcholine acyltransferase
(LPCAT) activities [7, 8]. ese activities determine their
roles in various organs under diﬀerent physiological and
1 Institute of Medical Sciences, Tzu Chi University, Hualien, Taiwan
Full list of author information is available at the end of the article
Page 2 of 17
Phasuketal. Mol Brain (2021) 14:42
pathobiological conditions [9, 10]. Although PRDX6
is expressed in various brain regions associated with
fear regulation, including the hippocampus [11, 12] and
expressed in all cell types with high expression level in
the astrocytes [11, 13, 14], its function regarding cogni-
tion, particularly fear memory regulation has not yet
been identiﬁed. Previous ﬁndings conﬁrmed the asso-
ciation between enhanced fear memory and decreased
overall enzymatic activity of GPx in the hippocampus,
suggesting that GPx-PRDX6 may be involved in the regu-
lation of fear response . Besides, activation of PLA2 is
required to acquire and retrieve emotional memory ,
indicating that aiPLA2-PRDX6 may also have a similar
function. All the evidence mentioned above led us to
hypothesize that PRDX6 may play an important role in
regulating fear memory.
Trace fear conditioning (TFC) is a behavioral paradigm
widely used to study associative fear memory . e
molecular mechanisms underlying fear memory pro-
cesses are commonly approached with a fear condition-
ing paradigm, which shares similar mechanisms across
species [18, 19]. is task causes fear memory formation
by triggering a series of molecular and cellular changes
to strengthen synaptic plasticity in emotion-related brain
regions, including the hippocampus and the amygdala
. Tyrosine kinase receptor B (TrkB) and its down-
stream targets such as extracellular signal-regulated pro-
tein kinases 1 and 2 (ERK1/2) and protein kinase B (AKT)
 are involved in the mediation of synaptic plasticity
for fear memory formation. Interestingly, PRDX6 can
modulate both ERK1/2 and AKT  expression, sup-
porting our hypothesis that PRDX6 may participate in
the neurobiological process of fear memory.
We performed behavioral, cellular, and molecular stud-
ies in the Prdx6 knockout (Prdx6−/−) mice in the pre-
sent study. We ﬁrst identiﬁed the function of PRDX6
by employing Prdx6−/− mice to trace fear conditioning
(TFC) and found that this knockout strain exhibited
enhanced contextual fear memory. We further conﬁrmed
with a gain-of-function study by injecting lentivirus-car-
rying mouse PRDX6 (mPRDX6) into the lateral ventricle
of Prdx6−/− mice, which mitigated their enhanced con-
textual fear memory. We also investigated their general
behaviors using open ﬁeld, three-chambers tests, marble
burying, and elevated plus-maze. Proteomic and immu-
noblotting analyses were also performed in this study to
understand the molecular mechanism better.
All experiments on animals were approved by the Insti-
tutional Animal Care and Use Committee of Tzu Chi
University, Taiwan (approval #104099), and complied
with the Taiwan Ministry of Science and Technology
guidelines for animals’ ethical treatment. Twelve- to
14-week-old wild-type (C57BL/6J) and Prdx6−/− mice
were originally generated by Wang X. and colleaguesand
provided by Dr. Shun-Ping Huang at Tzu Chi University,
Taiwan . All mice were maintained in the Laboratory
Animal Center of Tzu Chi University and were housed
with adlibitum access to food and water under a constant
12-h light/dark cycle. Heterozygous knockout mice with
one male and two females were crossed to reproduce
Prdx6−/− mice and their wild-type littermates. Genotyp-
ing (Additional ﬁle1: Fig. S1a) was conducted to conﬁrm
the absence ofthe Prdx6 genein knock-out mice before
every behavioral test. After the completion of trace fear
conditioning, qRT-PCR (Additional ﬁle1: Fig. S1b) and
immunoblotting (Additional ﬁle 1: Fig. S1c) were car-
ried out to visualize mRNA and protein level of PRDX6
in Prdx6−/− mice. Moreover, we also recorded the mor-
phology and bodyweight of the Prdx6−/− mice. We found
that both morphology (Additional ﬁle 1: Fig. S1d) and
body weight (Additional ﬁle 1: Fig. S1e) (t19 = − 1.426,
p = 0.170) of the Prdx6−/− mice appeared to be normal.
Trace fear conditioning (TFC)
Trace fear conditioning was modiﬁed from the protocol
used in our previous study . e conditioned cham-
ber (17 cm (W) × 17 cm (L) × 25 cm (H)) illuminated
with a white 30-lx light under the top-view camera was
used in this study. After three days of habituation, mice
were placed into the chamber for 2min as a baseline and
were then trained with three pairs of tone (CS) and elec-
tric foot shock (US) with an inter-trial interval of 1min.
One pair of CS-US consisted of a 20s of tone (6000Hz,
85dB) followed by 1s electric foot shock (2mA) with a
10 straining interval. e mice were maintained in the
conditioned chamber for a total of 9min. To test their
contextual fear memory retention, the mice were re-
exposed to a conditioned chamber for 6 min without
giving any tone and footshock after 24h of the training
session. One hour later, themice were tested with cue
fear memory by exposing them to 6 min of tone only
after 1min of habituation in an unconditioned context.
e freezing behavior, deﬁned as no movement except
breathing, was analyzed using tracking software (EthoVi-
sion XT 15, Noldus Information Technology). e freez-
ing time was converted to freezing percentage using the
total freezing time/total test time
Page 3 of 17
Phasuketal. Mol Brain (2021) 14:42
Open eld test
An open chamber (50cm (W) × 50cm (L) × 50 cm (H))
was used to test the locomotor functionand anxiety-like
behavior of the mice under the light-on condition .
e camera hung on top recorded the animals’ locomo-
tor activity within 10 min of the test. eir locomotor
activity (distance traveled and moving speed) and time
spent in the center and outer area were measured and
analyzed by the tracking software (EthoVision XT 15,
Noldus Information Technology).
is task was composed of three trials with 10min of
exploration time for each. e intertrial interval was
20 min. During the habituation trial, the experimental
mice were placed into the middle compartment. Mice
freely explored all three compartments that contained
empty cups at the end of the left and right compartments.
For the second trial, a sex- and age-matched stranger
mouse (S1) was kept inside the cup in the right com-
partment. e experimental mice were then allowed to
explore all compartments. For the third trial, another
stranger mouse (S2) was placed in the cup located in
the left compartment. e experimental mice were
again placed in the middle compartment and allowed to
explore the chamber. e time spent interacting with the
empty cups or stranger mice was analyzed by tracking
software (EthoVision XT 15, Noldus Information Tech-
nology). We followed the protocol described in a previ-
ous study .
Marble burying test
e protocol was described in a previous study .
Brieﬂy, the cage (30cm × 27cm × 26cm) was ﬁlled with
5cm autoclaved bedding containing 20 marbles arranged
centrally 4 by 5 and was kept in a soundproof box with
10lx. Mice were placed and then ﬁlmed for 30min. e
number of unburied marbles was counted after 25min.
Elevated‑plus maze test
e elevated-plus maze is used to assess the anxiety-
related behavior in rodents . e apparatus consists
of a "plus"-shaped maze at 60cm height above ground
with two oppositely positioned closed arms and two
oppositely positioned open arms and a center region. e
experiment was conducted during day time under the
same light intensity (~ 130lx) as provided in the animal
housing room. e mice were placed in the center region
facing one of the closed arms and allowed to explore
the maze freely for 10min. We used a video camera and
tracking system (EthoVision XT 15, Noldus Informa-
tion Technology) to record and analyze their anxiety-like
Lentiviral vector preparation
Total RNA was isolated from the mouse hippocampus
and converted to cDNA using oligo (dT) 18 primers.
e cDNA was then ampliﬁed using a speciﬁc forward
primer (5′-CTA GCT AGC ATG CCC GGA GGG TTG
CTT C-3′ containing a NheI site) and reverse primer
(5′-GC GAA TTC TTA AGG CTG GGG TGT ATA
ACG-3′containing an EcoRI site) . Full-length mouse
Prdx6 cDNA was puriﬁed by a PrestoTM Mini Plasmid
Kit (catalog #PHD300, Geneaid Biotech Ltd., Taiwan).
pLAS3wPpuro vectors containing EGFP and Prdx6
were designed for the production of lentiviral vectors.
HEK293T cell lines were used to produce lentiviruses
containing either EGFP or Prdx6 gene. Harvested lenti-
virus was concentrated using PEG-it (™) virus precipita-
tion solution (System Biosciences, CA) and processed for
Stereotaxic surgery andintracerebroventricular injections
oflentivirus containing mouse PRDX6
e procedures for stereotaxic injection were performed
according to our previous study with slight modiﬁcation
. e mice were anesthetized by intraperitoneal (IP)
injection of ketamine/xylazine mixture (0.45ml/25 g of
body weight) and then ﬁxed on the stereotaxic frame
(Stoelting, US). e lentivirus containing either EGFP
or mouse PRDX6 was dissolved in sterile 1× phos-
phate-buﬀered saline (PBS) to obtain the ﬁnal titer of
7 × 105 in 2 µl volume. e lentiviral vectors were then
unilaterally injected into the right lateral third ventricle
with the following brain coordinates: anterior–posterior
(AP), − 0.5mm; medial–lateral (ML), 1mm (from the
bregma): and DV, 2.33mm (from the skull surface). A
10-µl Hamilton syringe with a 26 G needle was placed on
the microinfusion pump (KD Scientiﬁc Inc. MA, USA)
and connected via polyethylene—28 mm I.D. tubing to
the internal cannula. We injected the lentiviruses with
a ﬂow rate of 0.5µl/min over 4min. e cannulas were
placed for another 5min to allow diﬀusion before remov-
ing them. Following surgery, mice were given pain killers
(meloxicam, Achefree, Taiwan) and allowed to recover
for 4weeks before the behavioral tests.
Liquid chromatography–tandem mass spectrometry (LC/
After completing a contextual test, protein samples were
collected from the whole hippocampi of Prdx6+/+ and
Prdx6−/− mice. Protein samples from 3 mice were pooled
Page 4 of 17
Phasuketal. Mol Brain (2021) 14:42
together for each group and measured the protein con-
centration using Lowry assay . For in-solution diges-
tion, 5µg of protein were used for each group of mice.
e samples were treated with 10mM ammonium bicar-
bonate and the disulﬁde bonds were reduced with 5mM
dithiothreitol (DTT) in 10mM ammonium bicarbonate
at 60 °C for 1 h. Samples were subsequently alkylated
with 15 mM Iodoacetamide (IAA) in 10 mM ammo-
nium bicarbonate for 45min in the dark at room tem-
perature. Protein digestion was done by incubating the
samples with 50ng/µl of sequencing grade trypsin (1:20
trypsin:protein) (Promega, Germany) o/n at 37°C. Before
the injection into the LC–MS/MS, the samples were pro-
tonated with 0.1% formic acid.
e tryptic peptides from the digested samples were
injected into an Ultimate3000 Nano/Capillary LC Sys-
tem (ermo Scientiﬁc, UK) coupled to a Hybrid quad-
rupole Q-Tof impact II™ (Bruker Daltonics) equipped
with a Nano-captive spray ion source. e peptides were
enriched on a µ-Precolumn 300µm i.d.×5mm C18 Pep-
map 100, 5µm, 100 A (ermo Scientiﬁc, UK), separated
on a 75μm I.D. × 15cm and packed with Acclaim Pep-
Map RSLC C18, two μm, 100Å, nanoViper (ermo Sci-
entiﬁc, UK). Solvent A and B containing 0.1% formic acid
in water and 0.1% formic acid in 80% acetonitrile were
supplied on the analytical column. A gradient of 5–55%
solvent B was used to elute the peptides at a constant
ﬂow rate of 0.30μl/min for 30min. Electrospray ioniza-
tion was carried out at 1.6 kV using the CaptiveSpray.
Mass spectra (MS) and MS/MS spectra were obtained
in the positive-ion mode over the range (m/z) 150–2200
(Compass 1.9 software, Bruker Daltonics). We performed
the LC–MS analysis of each sample in triplicate.
Bioinformatics anddata analysis
e MS data were quantiﬁed with MaxQuant 188.8.131.52
using Andromeda search engine to correlate MS/MS
spectra to the Uniprot Mus Musculus database .
Using MaxQuant’s standard settings, label-free quan-
titation was performed. We used trypsin as a digesting
enzyme, carbamidomethylation of cysteine as a ﬁxed
modiﬁcation, and the oxidation of methionine and acety-
lation of the protein N-terminus as variable modiﬁca-
tions. We set two miss cleavages as the maximum and a
0.6 Dalton as the main search’s mass tolerance. At least
one unique peptide with a minimum of 7 amino acids
was used for further analysis [30, 31].
e log2 fold change > 1.2 was a cut oﬀ for diﬀerential
expression proteins (DEPs) [31, 32]. e list of diﬀerential
expression proteins (DEPs) was then inputted to Venn
diagrams . e list of up-and down-regulated pro-
teins was then inputted in Panther software for protein
classiﬁcation . Enrichr software was used to analyze
enrichment terms from gene ontology (GO) biologi-
cal processes (https ://amp.pharm .mssm.edu/Enric hr/).
e functional interaction networks between DEPs and
memory-associated molecules were analyzed using the
Search Tool for the Retrieval of Interacting Genes/Pro-
teins (STRING) database version 11 (http://strin g-db.
org/cgi/input .pl). e MultiExperiment Viewer (MeV)
software  was used to produce a heatmap for up-and
down-regulated proteins extracted from the GO term
Detection ofoxidative stress levels inthehippocampus
To measure reactive oxygen species (ROS) levels in the
hippocampus, mice were sacriﬁced, and the brains were
isolated 20 min after the contextual test. e proce-
dure was conducted according to a previous study with
minor modiﬁcations . Brieﬂy, the ﬁxed brains were
sectioned by cryostat with 20µm thickness. Hippocam-
pal sections were then immersed in 1 μmol/l dihydro-
ethidium (DHE) in PBS solution at room temperature
for 5 min. e stained sections were washed with 1×
PBS three times and cover-slipped. DHE is oxidized by
superoxide anion to form ethidium binding to DNA in
the nucleus and emits red ﬂuorescence. e images were
viewed and taken under a ﬂuorescent microscope (Nikon
model #ECLIPSE Ni-E, Japan) with an excitation/emis-
sion wavelength of 380/420nm.
For immunohistochemistry, mice were anesthetized and
transcardially perfused using 0.9% saline and 4% para-
formaldehyde. Brains were exercised immediately and
postﬁxed with 4% PFA for another 2 days. After that,
the brains were washed with 1× PBS three times and
then stored in 30% sucrose at 4°C. After the dehydration
period, the brains were embedded in an optimal cutting
compound (Sakura Finetek USA, Inc., USA) and stored
at − 80 °C until sectioning. Cryopreserved brains were
sectioned at 20 µm using cryostat. Brain sections were
washed with a washing buﬀer (1× PBS containing 0.3%
Triton X-100) and treated with apermeating buﬀer (1%
Triton X-100 and 2% Tween 20 in 1× PBS) for 30min.
Sections were further blocked with a blocking buﬀer
containing 1% normal goat serum, 0.25% Triton X-100
dissolved in 1× PBS for 1h. Subsequently, samples were
double-stained with polyclonal rabbit anti-GFAP (1:200,
Abcam, UK) and monoclonal mouse anti-PRDX6 (1:150,
Bethyl laboratories, Inc, USA). e samples were then
washed with washing buﬀer and incubated in secondary
antibody: Alexa Fluor 546 anti-mouse and Alexa Fluor
488 anti-rabbit IgG (1:200, ermoFisher Scientiﬁc,
Page 5 of 17
Phasuketal. Mol Brain (2021) 14:42
USA) for 1h, followed by washes with PBS, and coun-
terstained with DAPI (1:10,000) for 5min. e images
were obtained by either ﬂuorescent microscope (Nikon
model# ECLIPSE Ni-E, Japan) or confocal microscope
(Nikon model#C2+, Japan).
Western blot analysis
e mice were sacriﬁced immediately after the comple-
tion of acute immobilization stress. Under trace fear
conditioning, hippocampal proteins were extracted at
3h after training and 20 min after the contextual test.
After decapitation, the whole hippocampi were isolated
and homogenized in ice-cold RIPA lysis buﬀer 1× (Mil-
lipore, USA) containing protease and phosphatase inhibi-
tors. e protein samples were kept on ice for 30min
before centrifuging at 13,000rpm for 15min at 4°C. e
supernatants were collected for further experiments.
For non-reducing SDS-PAGE, protein (30 or 45 μg)
samples were boiled at 95°C in 1× sample buﬀer with-
out reducing agent for 10min, and samples were cooled
for 5 min. Similar to non-reducing conditions, adding
reducing agents into protein samples were included to
study total PRDX6 and other proteins of interest under
reducing condition. e samples were loaded and run
on 8% or 10% SDS-PAGE at 80V in stacking gel and
120V in resolving gel. e separated proteins were then
transferred to a PVDF membrane (0.2 and 0.4μm pore
size) at 30V overnight. e blots were incubated with
anti-NTRK2 (1:1000; Abcam, UK), anti-cPLA2 (1:1000;
Santa Cruz, USA), anti-pERK1/2 (1:1000; Cell Signaling,
Danvers, MA), anti-total ERK1/2 (1:1000; Cell Signal-
ing, Danvers, MA), anti-PSD95 (1:2000; ermoFisher,
USA), anti-PRDX6 (1:2000; Abcam, UK) or anti-β-actin
antibody (1:10,000; Sigma-Aldrich) in TBST containing
0.1% BSA (ermoFisher, USA) overnight at 4°C room
on a shaker. e next day, the blots were incubated with
horseradish peroxidase-conjugated secondary antibody
goat anti-mouse IgG (Cell signaling, Danvers, MA) for
cPLA2, PSD95, PRDX6, and β-actin and goat anti-rabbit
(Santa Cruz Biotechnology, Santa Cruz, CA, USA) for
total NTRK2 or TrkB, pERK1/2 and ERK1/2 with the
dilution of 1: 10,000 in blocking buﬀer for 1h at room
temperature. A list of antibodies used in this study was
provided in Additional ﬁle2: TableS5. After three washes
for 5-min in the TBST buﬀer, the membranes were devel-
oped using ECL (Western lightning® Plus ECL, Perki-
nElmer Inc, MA, USA) and detected under the UVP
Biospectrum 810 imaging system. e band intensities
were quantiﬁed using ImageJ 1.52a (National Institutes of
Based on previous studies [37–39], we decided to use a
sample size from 3 to 20 per group with enough power to
see a statistically signiﬁcant diﬀerence. Statistical analy-
sis was performed using SPSS (version 25, IBM Corpora-
tion), and the graphs were made using GraphPad Prism
version 8. After assessing the normality using the Shap-
iro–Wilk test, Student’s t-tests were conducted compared
to two independent groups with a normal distribution.
In contrast, data that is not normally distributed were
assessed by Mann–Withney U-test. One-way ANOVA
followed by Bonferroni’s post hoc analysis was used for
multiple comparisons. For learning ability of TFC and
social interaction of three-chamber test, the results were
analyzed as mixed-design repeated-measures ANOVA
with trials as within-subjects factor and genotypes as a
between-subjects factor. e signiﬁcant interaction was
then followed up with the Bonferroni-corrected t-test
when a signiﬁcant F-value was determined. All data are
presented as mean ± SEM, with statistically signiﬁcant at
p < 0.05. Sample sizes are indicated in ﬁgure legends.
Prdx6−/− mice exhibited enhanced fear learning
To identify the function of PRDX6 in fear response,
Prdx6−/− mice underwent trace fear conditioning (TFC)
according to the protocol schemed in Fig. 1a. During
the ﬁrst three days, mice were placed in the condition-
ing chamber and acclimatized to the context for 15min
per day. On day 4, TFC was applied, followed by a con-
textual test 24 h later. Using mixed design repeated
ANOVA, there was no signiﬁcant eﬀect of the interaction
between the genotypes and trials on freezing percent-
age (F(2.254,58.606) = 1.042, p = 0.366, Fig. 1b) during TFC.
e two genotypes exhibited normal learning during the
training session, indicated by an increased freezing per-
centage from baseline to trial 3 as shown by the main
eﬀect of trials (F(2.254,58.606) = 125.868, p = 0.000, Fig .1b).
ere was a signiﬁcant eﬀect of genotypes on freezing
percentage during TFC (F(1,26) = 6.638, p = 0.016, Fig .1b).
Bonferroni-corrected t-test revealed signiﬁcant diﬀer-
ence between the two genotypes at trials 2 (t26 = − 2.580,
p = 0.016, Fig.1b). ese results suggested that deﬁciency
of PRDX6 leads to fast acquisition of fear memory. No
signiﬁcant diﬀerence in total freezing percentage during
TFC training (t26 = − 1.302, p = 0.204, Fig. 1c) between
the Prdx6+/+ and Prdx6−/− mice. Interestingly, the
Prdx6−/− mice exhibited a signiﬁcantly higher freezing
response to conditioned context (t26 = − 2.985, p = 0.006,
Fig. 1d) and cue (t26 = − 2.956, p = 0.007, Fig. 1e) than
the Prdx6+/+ mice suggesting the impact of thePrdx6
Page 6 of 17
Phasuketal. Mol Brain (2021) 14:42
gene on the regulation of contextual and cued fear
Lentivirus containing mouse PRDX6 (LV‑mPRDX6)
attenuated contextual fear memory ofPrdx6−/− mice
To further conﬁrm the role of PRDX6 in the expres-
sion of fear memory, the gain-of-function study was
conducted by intracerebroventricularly injecting LV-
mPRDX6 into the lateral ventricle near the hippocampal
region of Prdx6−/− mice. e mice were then subjected
to TFC 4weeks after the injection (Fig.2a). Figure2band
c illustrate the site of injection and lentiviral construct,
respectively. ere was no eﬀect of group on learn-
ing ability as shown in Fig.2d (F(1,16) = 0.551, p = 0.469;
Prdx6−/− mice with LV-EGFP vs LV-mPRDX6). Both
groups displayed normal learning during training ses-
sions indicating an increased freezing percentage from
baseline to trial 3 as shown by the main eﬀect of trials
(F(3,48) = 26.691, p = 0.000, Fig. 2d). During the training
session, the total freezing percentage was similar between
the two groups (t16 = − 0.654, p = 0.522, Fig. 2e), indi-
cating that the injection of LV-mPRDX6 did not aﬀect
the learning ability of the Prdx6−/− mice. Importantly,
lentiviral injection of mPRDX6 successfully reduced
Fig. 1 Loss of the Prdx6 gene caused fast learning and enhanced fear memories. a General procedure for trace fear conditioning: Habituation of
mice in the chamber was performed for 3 consecutive days. The next day, mice were conditioned with three tone and shock pairs. Contextual fear
memories were tested 24 h later, followed by a tone test to evaluate cue fear memory (n = 14/group). b The learning curve for baseline and three
trials of TFC indicated both groups of mice learnt normally though signiﬁcant diﬀerences appeared in trial 2. c Total freezing percentage of Prdx6+/+
or Prdx6−/− mice during the training session. d Total freezing percentage during the contextual test of mice. e Total freezing percentage during the
tone test of mice. All data represent the mean ± the SEM. *p < 0.05. TFC trace fear conditioning
Page 7 of 17
Phasuketal. Mol Brain (2021) 14:42
enhanced contextual fear response of the Prdx6−/− mice
(t16 = 2.698, p = 0.016, Fig. 2f). However, re-expression of
mPRDX6 failed to rescue cue fear memory (t16 = − 0.700,
p = 0.494, Fig. 2g). Fluorescent images demonstrated the
expression of mPRDX6 (Fig.2h) and EGFP (Additional
ﬁle1: Fig. S2a, b) in three hippocampal regions, including
the CA1, CA3, and DG after the completion of the tone
test. We also detected the expression of mPRDX6 in the
Fig. 2 Intraventricular injection of mouse PRDX6 (mPRDX6) lentiviruses attenuated enhanced contextual fear memory of Prdx6−/− mice. a The
procedure for the overexpression study. Mice were injected with lentivirus and housed for 2 weeks before performing trace fear conditioning.
b Representative image of cannula tip position (red) in the right lateral ventricle. c The schematic lentivirus construct pLAS3w. Ppuro contains
either EGFP or mPRDX6. d The learning curve of baseline and after each tone‑shock pair (n = 10/group). e Total freezing percentage of Prdx6+/+
and Prdx6−/− mice during the training session. f Total freezing percentage during the contextual test of mice. g Total freezing percentage of
the mice during the tone test. h mPRDX6 expression across the diﬀerent subregions in the hippocampus of Prdx6−/− mice. All data represent the
mean ± the SEM. *p < 0.05. TFC trace fear conditioning, LV-mPRDX6 lentivirus containing mouse PRDX6
Page 8 of 17
Phasuketal. Mol Brain (2021) 14:42
amygdala and prefrontal cortex (Fig.2h and Additional
ﬁle1: Fig. S3a, b). ese results suggest that hippocampal
PRDX6 is involved in regulating fear expression, at least
for contextual fear memory.
Deletion ofthePrdx6 gene caused hyperlocomotion
withoutaecting social exploration andrecognition
e heatmap during 10min of exploration in the open
ﬁeld chamber was presented in Fig. 3a. e total dis-
tance traveled (t31 = − 2.191, p = 0.036, Fig. 3b) and
moving speed (t31 = − 2.197, p = 0.036, Fig. 3c) of the
Prdx6−/− mice were signiﬁcantly higher than those of
the Prdx6+/+ mice. An open ﬁeld test indicated that
Prdx6−/− mice exhibited hyperlocomotion compared
with Prdx6+/+ mice; hence higher freezing response
to context did not result from reduced locomotion. We
then assess object exploration, sociability and social
novelty behaviors of the Prdx6−/− mice using a three-
chamber apparatus . For the novel object exploration
test (trial 1), both genotypes demonstrated a signiﬁcant
preference for exploring empty cups, and no signiﬁcant
genotype eﬀect was observed (side: F(1.604,52.935) = 46.642,
p = 0.000; genotype: F(1,33) = 0.003, p = 0.958; geno-
type × side: F(1.604,52.935) = 0.794, p = 0.432, Fig. 3d). e
stranger mouse 1 (S1) was placed in the right compart-
ment within an inverted wire cup for the sociability test.
Both genotypes demonstrated a signiﬁcant preference
for exploring stranger mouse 1 and no signiﬁcant geno-
type eﬀect was observed (side: F(2,66) = 28.869, p = 0.000;
genotype: F(1,33) = 0.232, p = 0.633; genotype × side:
F(2,66) = 0.118, p = 0.889, Fig. 3e). In the social novelty
preference test, the interaction duration with the novel
mouse (S2) appeared to be normal since the Prdx6−/−
mice spent similar time with the novel mouse compared
to wild-type group (F(1,20) = 0.000; p = 0.991, Fig. 3f ).
Both genotypes stayed with the novel mouse longer than
the familiar mouse (S1) (F(1.495,29.895) = 11.089; p = 0.001,
Fig.3f), representing the normal response of social nov-
elty. In each test, no signiﬁcant diﬀerence in locomo-
tor activity was recorded, measured by equal distance
traveled (t38 = − 1.056, p = 0.297, Fig. 3g) and moving
speed (t38 = 0.340, p = 0.736, Fig . 3h) between the two
Normal anxiety‑like behavior andhypervigilance
We next investigated anxiety response and hypervigilance
in Prdx6−/− mice using an open ﬁeld, elevated plus-maze,
and marble burying tests, respectively. We observed
equal time spent in the center (t31 = − 0.493, p = 0.632,
Fig. 3i) and outer zone (t31 = 0.235, p = 0.816, Fig. 3j)
in an open-ﬁeld chamber between the two genotypes.
e Prdx6−/− mice showed similar results in elevated
plus-maze as controls indicated by equal time spent in
open arms (U = 171, Z = − 0.263, p = 0.792, Fig. 3k) and
close arms (t36 = − 0.180, p = 0.858, Fig. 3l) indicating
normal anxiety-like behavior in Prdx6−/− mice. Perform-
ing the marble burying test, we observed no signiﬁcant
diﬀerence in the percentage of buried marbles between
the two genotypes (t20 = 0.378, p = 0.709, Fig. 3m). is
result demonstrated that deletion of the Prdx6 gene did
not cause hypervigilance.
Proteomic analysis fortotal hippocampal proteins
extracted duringthecontextual memory retrieval stage
To understand what hippocampal proteins are involved
in the retrieval process of contextual memory, we con-
ducted a proteomic analysis for total hippocampal pro-
teins collected during the retrieval stage of TFC (Fig.4a).
Liquid chromatography–tandem mass spectrometry
(LC/MS–MS) provided a total of 937 proteins that dif-
ferentially expressed in the hippocampus of Prdx6−/−
and Prdx6+/+ mice. e top 20 up- and down-regulated
diﬀerential expression proteins (DEPs) was provided in
Additional ﬁle2: TableS1. All proteins on the list from
both genotypes were plotted in Venn diagrams based on
their expressions (Fig.4b). ere were 11 proteins spe-
ciﬁcally expressed in Prdx6+/+ mice, 7 proteins expressed
only in Prdx6−/− mice, and 919 proteins expressed in
both genotypes. Using Panther software, the diﬀeren-
tial expression proteins (DEPs) were classiﬁed into three
gene ontologies (GO): molecular function, biological
process, and cellular component. According to molecu-
lar functions, the most overrepresented groups were
catalytic activity (40.50% up- and 25.23% down-regulated
proteins, Fig. 4c) and binding (35.10% up- and 47.85%
down-regulated proteins). In the GO biological process,
the main biological processes of DEPs were cellular pro-
cesses (27.13% up- and 25.93% down-regulated proteins,
Fig.4d) and metabolic processes (16.32% up- and 3.61%
down-regulated proteins). e analysis of cellular com-
ponents indicated that cell (the plasma membrane and
any external encapsulating structures; 23.78% up- and
23.83% down-regulated proteins, Fig. 4e) and cell part
(any constituent part of a cell; 23.78% up- and 23.83%
down-regulated proteins) were the main cellular compo-
nents of DEPs.
PRDX6 regulates fear memory retrieval viatheMAPK
Oxidative stress is known to be involved in the modula-
tion of fear memory . To measure the oxidative status
in the hippocampus of Prdx6−/− mice during memory
retrieval, we performed dihydroethidium (DHE) stain-
ing. e brains were collected 20min after the contextual
test (Additional ﬁle1: Fig. S4a). Additional ﬁle1: Figure
Page 9 of 17
Phasuketal. Mol Brain (2021) 14:42
Fig. 3 Increased locomotor function, but normal anxiety‑like behavior, exploration, sociability, and social novelty in Prdx6−/− mice. a Heatmaps
during 10 min of exploration in an open ﬁeld chamber. b Quantiﬁcation data of distance traveled for 10 min (n = 15–18/group, Student’s t‑test). c
The mean moving speed (cm/s) of the mice introduced in the open ﬁeld test. d Time spent on each side of the chamber containing empty wire
cups (novel object) (n = 17/group). e Time spent on each side of the chamber containing a stranger mouse 1 (S1) or empty wire cup. f Time spent
on each side of the chamber containing familiar mouse 1 (S1) or novel mouse (S2). g The mean distance traveled during three trials of the task. h
The mean moving speed during three trials of a task. i The mean percentage of center zone time (n = 15–18/group, Student’s t‑test). j The mean
percentage of outer zone time. k Percent time spent in open arms (n = 18–20/group). l Percent time spent in close arms. m Percent marbles buried
in the marble‑burying test. All data represent the mean ± the SEM. *p < 0.05
Page 10 of 17
Phasuketal. Mol Brain (2021) 14:42
Fig. 4 Three functional classiﬁcations of the proteins in the hippocampus of mice re‑exposed to conditioned chambers. a The schematic diagram
of trace fear conditioning and protein collection (n = 3/group). b Venn diagram deﬁning the diﬀerence in protein expressions between Prdx6−/−
and Prdx6+/+ mice. The 125 upregulated proteins and 130 down‑regulated proteins aﬀected by the contextual test were classiﬁed into three
functional classiﬁcations: c molecular function, d biological process, and e cellular component
Page 11 of 17
Phasuketal. Mol Brain (2021) 14:42
S4b shows the ethidium ﬂuorescence of DHE. Quantita-
tive analysis showed no signiﬁcant diﬀerence in the DHE-
positive density was found between genotypes in both
CA1 (t4 = − 0.508, p = 0.638, Additional ﬁle1: Fig. S4c)
and CA3 (t4 = − 0.060, p = 0.0.955, Additional ﬁle1: Fig.
S4d) subregions of the hippocampus. e results demon-
strated that PRDX6 might not regulate fear response by
controlling cellular oxidation. Moreover, it led us to ques-
tion whether PRDX6 directly modulates the cellular sign-
aling cascade to control fear memory expression.
To delineate the molecular pathways responsible for
the enhanced fear response of the Prdx6−/− mice, Enri-
chr software was then conducted to identify the enriched
biological process of the DEPs (cut-oﬀ 1.2 fold change)
in Prdx6−/− mice. e protein phosphorylation (GO:
0006468) (p = 0.0056) was one of the signiﬁcant enrich-
ment terms from the GO biological process of DEPs in
Prdx6−/− mice (Additional ﬁle 2: Table S2; the top 25
enrichment terms from GO biological process). We next
extracted 15 DEPs from the GO term "protein phospho-
rylation" (GO: 0006468) (Additional ﬁle2: TableS3). e
8 proteins that were upregulated and 7 proteins that were
down-regulated in Prdx6−/− mice were input to STRING
software to obtain the networks of protein–protein inter-
action involved in memory processes (Fig. 5a, b). e
signiﬁcant nodes of DEPs were identiﬁed according to
the Kyoto Encyclopedia of Genes and Genomes (KEGG)
pathway database (Additional ﬁle2: TableS4). e DEPs
of Prdx6−/− mice were strongly associated with the neu-
rotrophin signaling pathway (false discovery rate or
FDR = 0.00012), Ras signaling pathway (FDR = 0.00073),
and MAPK signaling pathway (FDR = 0.0011). ese
pathways are well known to fear memory consolidation
and retrieval [42, 43]. To study the molecular changes that
participate in regulating contextual fear memory retrieval
of Prdx6−/− mice, we extracted DEPs from the MAPK
signaling pathway, including AKT2, CHUK, NTRK2, and
RPS6KA1. We created new networks, including MAPK1,
MAPK3, BDNF, cPLA2, and PRDX6, using STRING soft-
ware (Fig. 5c). During retrieval (Fig. 6a), western blot
analysis was performed to conﬁrm the expression of
the key proteins from the network, including NTRK2,
AKT, ERK1/2, and cPLA2, during retrieval of contextual
memory. Signiﬁcant upregulation of NTRK2 (or TrkB)
(t6 = − 2.798, p = 0.031, Fig. 6b), AKT (t6 = − 4.242,
p = 0.005, Fig. 6c), ERK1/2 phosphorylation (t5 = − 5.336,
p = 0.003, Fig. 6d) and cPLA2 (t6 = − 2.761, p = 0.033,
Fig.6e) were recorded in the hippocampus of Prdx6−/−
mice after a contextual test. Postsynaptic density protein
95 (PSD95), a postsynaptic marker, was also detected
and no signiﬁcant diﬀerence was recorded (t6 = − 1.843,
p = 0.115, Fig. 6f) in Prdx6−/− mice. ese results dem-
onstrated the correlation of the MAPK pathway with
PRDX6 to regulate fear memory retrieval.
Co‑localization ofPRDX6 withtheastrocytic marker, GFAP,
Since we focused on identifying the function of PRDX6
in the regulation of fear memory, we then confirmed
the distribution of PRDX6 in three brain regions
Fig. 5 Proteomic analysis revealed diﬀerential expression proteins (DEPs) MAPK and Ras signaling pathways in the hippocampus of Prdx6−/−
mice during retrieval of contextual fear memory. a Functional protein–protein interaction networks of 15 proteins related to GO term "protein
phosphorylation" (GO:0006468). The signiﬁcant nodes were labeled in red for Neurotrophin signaling pathway (FDR 0.00012), blue for the
Ras signaling pathway (FDR 0.00073), and green for the MAPK signaling pathway (FRD 0.0011). b Heatmap of 15 proteins in GO term "protein
phosphorylation" (GO:0006468) with 8 upregulated proteins and 7 down‑regulated proteins in Prdx6−/− mice. c STRING showed a predicted
functional protein–protein interaction network of proteins in the KEGG pathway termed "MAP kinase signaling pathway" with memory‑associated
proteins and PRDX6. The signiﬁcant nodes were labeled in red for the MAPK signaling pathway (FDR 1.25e−12) and green for the Ras signaling
pathway (FDR 3.33e−11)
Page 12 of 17
Phasuketal. Mol Brain (2021) 14:42
primarily involved in fear memory formation—the
hippocampus , amygdala , and prefrontal cor-
tex . Previous studies report that PRDX6 is highly
expressed in the astrocytes  under various con-
ditions but not known in TFC. We thus performed
double staining with anti-PRDX6 and anti-GFAP
(astrocyte marker) antibodies to examine whether
PRX6 is also expressed in astrocytes after TFC. Our
results demonstrated that PRDX6 is expressed in the
hippocampal astrocytes within the CA1, CA2, CA3,
and DG (Fig.7a, b). We also recorded the expression
of PRDX6 in the amygdala (Additional file1: Fig. S5a,
b) and prefrontal cortex (Additional file1: Fig. S6a, b).
e present study reports that the loss of the Prdx6
gene in the brain led to enhanced trace fear memory to
context. e intracerebroventricular injection (i.c.v) of
LV-mPRDX6 could reverse the enhanced contextual
fear response, a hippocampal-dependent memory, of
Prdx6−/− mice. We conﬁrmed that the observed eﬀect
was attributable to PRDX6. Proteomic and western blot
analysis revealed that mitogen-activated protein kinase
Fig. 6 Activation of MAPK signaling pathways in the hippocampus of Prdx6−/− mice during retrieval of contextual fear memory. a Hippocampal
protein samples were collected 20 min after the contextual test for the retrieval process to validate proteins in the MAPK signaling pathway
(mmu04010), including TrkB (NTRK2), AKT, pERK1/2, and cPLA2. (b–f; upper panels) Immunoblots of TrkB, AKT, pERK1/2, tERK1/2, cPLA2, PSD95,
and β‑actin expression in the hippocampus during memory retrieval. (b–f; lower panels) Quantiﬁcation data for the expression levels of TrkB, AKT,
phosphorylated ERK1/2, cPLA2, and PSD95 in the hippocampi of mice (n = 3–5/group). All data represent the mean ± the SEM. *p < 0.05. TrkB
tyrosine receptor kinase B, AKT or PKB Protein kinase B, PSD95 postsynaptic protein density 95
Page 13 of 17
Phasuketal. Mol Brain (2021) 14:42
Fig. 7 PRDX6 protein is highly expressed in the astrocytes throughout the hippocampus. a Sagittal section of the brain showing colocalization
of PRDX6 (red) with GFAP (green), scale bar 1000 μm. b Confocal images of PRDX6‑GFAP colocalization in the astrocytes of CA1, CA2, CA3, and DG
subregions of the hippocampus, scale bar 20 μm
Page 14 of 17
Phasuketal. Mol Brain (2021) 14:42
(MAPK) signaling pathways were highly activated in
the hippocampi of Prdx6−/− mice during the expression
of contextual fear memory. ese results suggest that
PRDX6 plays a critical role in the regulation of fear mem-
In humans, the feeling of intense fear has been deﬁned
using the Diagnostic and Statistical Manual of Mental
Disorders, ﬁfth edition (DSM-V) as one primary symp-
tom of PTSD [45, 46]. ree brain regions, the hippocam-
pus, amygdala, and prefrontal cortex, are important for
fear memory formation . Here we demonstrated that
PRDX6 is expressed in the astrocytes of the amygdala
(Additional ﬁle1: Fig. S5), prefrontal cortex (Additional
ﬁle1: Fig. S6), and hippocampus (Fig.7a, b) after TFC.
e activation of hippocampal astrocytes plays a crucial
role in synaptic plasticity and contextual fear memory
[47, 48]. It is known that PRDX6 can modulate astrocyte
activation [49, 50]. Whether PRDX6 may regulate the
activation of astrocytes during the synaptic process and
memory formation requires further study to verify.
In the present study, we used systemic Prdx6 knock-
out strain, which lacks PRDX6 in the whole brain ,
for trace fear conditioning (TFC). Since intracerebroven-
tricular injection of mouse PRDX6 lentivirus reduced
contextual fear memory, we thus focused on identifying
the function of PRDX6 in the required brain region—the
hippocampus [51, 52]. We found that PRDX6 is colocal-
ized with an astrocytic marker, EGFP, within the hip-
pocampus. Although high expression level of PRDX6
in astrocytes was conﬁrmed, its expression in other cell
types was not examined in the present study. Given that
PRDX6 expression in diﬀerent cell types would aﬀect ani-
mal behavior, designing a construct containing a neuron
or oligodendrocyte speciﬁc promotor may help identify
related molecular and cellular mechanisms regarding
PRDX6′s function in memory formation.
Our results also demonstrate that the Prdx6−/− mice
displayed hyperlocomotion activity. is phenotype
conﬁrms that enhanced freezing behavior exhibited in
Prdx6−/− mice was attributable to the lack of PRDX6,
not reduced locomotor activity. Anxiety-like, motivation,
and exploration behaviors may also aﬀect response to
fear conditioning [53, 54]. ese behaviors are normal in
Prdx6−/− mice, indicating loss of PRDX6 does not cause
these phenotypes. is series of behavior tests suggest
that the ablation of Prdx6 is speciﬁcally responsible for
the enhanced fear memory.
e excessive fear expression to TFC exhibited by
Prdx6−/− mice was also observed in activating transcrip-
tion factor 3 (Atf3) deﬁcient mice . ATF3 is a leucine
zipper-containing (bZIP) transcription factor-induced
upon stress . Using a computer-based search program
(Alggen Promo software, version 8.3), we found that the
promoter region of the Prdx6 gene contains binding sites
for activating transcription factor 3 (ATF3). Moreover,
proteomic analysis (Additional ﬁle 2: Table S1) reveals
that expression of gelsolin (GSN), an actin-severing pro-
tein essential for synaptic plasticity , is reduced in the
hippocampus of Prdx6−/− mice. is phenomenon is also
recorded in the Atf3−/− mice after TFC . Besides, in
rats subjected to predator-scent-stress (PSS), a PTSD-like
model, gelsolin (Gsn) expression levels were also down-
regulated . We thus speculate that ATF3, PRDX6, and
GSN may participate in the same or related pathways for
the regulation of fear memory. Further experiments are
necessary to verify their relationship.
Inhibition of the memory retrieval process is proved
to attenuate excessive fear response . Stress and
stress hormone, glucocorticoid (GC) may positively or
negatively aﬀect fear response involving stress coping
mechanisms [58, 59]. PRDX6 can be regulated by dexa-
methasone, a glucocorticoid analog suggesting a possible
role of PRDX6 in stress coping mechanisms, including
fear response . Previous studies have shown the phys-
iological and pathological role of reactive oxygen species
(ROS) in fear response , and the hippocampal pyram-
idal neurons of the CA1 and CA3 subregions related to
fear memory retrieval are more vulnerable to oxidative
stress[51, 62, 63]. During contextual memory retrieval,
ROS level in the hippocampal CA1 and CA3 regions of
the Prdx6−/− mice remained similar as wild-type mice
(Additional ﬁle 1: Fig. S4), indicating that PRDX6 did
not regulate the expression of contextual fear memory
through modulating ROS level in the hippocampus.
e proteomic and western blot analysis revealed
that upregulation of several proteins (Additional ﬁle2:
Table S1) involved in the MAPK signaling pathway in
the hippocampus of Prdx6−/− mice during the retrieval
stage of contextual fear memory. It is known that ongo-
ing protein synthesis is required for maintaining GluA1-
AMPA receptors at the synapses for cue memory
retrieval . Tropomyosin receptor kinase B (TrkB)
and its downstream molecules (AKT and ERK1/2) par-
ticipate in the regulation of production and traﬃcking of
GluA1-AMPA receptors . A previous study revealed
immediate upregulation of total TrkB after a probe test
. Another study showed upregulation of total AKT
in the basolateral amygdala (BLA) 15min after reexpo-
sure to conditioned context . We also found imme-
diate upregulation of these proteins in the hippocampi
of Prdx6−/− mice. Further studies are necessary to ver-
ify whether upregulation of these proteins results from
local protein synthesis during contextual fear memory
retrieval. One previous research has shown that PRDX6
participates in the modulation of ERK1/2 activity in
the lung . Another study revealed that inhibition of
Page 15 of 17
Phasuketal. Mol Brain (2021) 14:42
ERK1/2 before a memory test blocks contextual fear
memory retrieval . ese studies suggest that hyper-
phosphorylation of ERK1/2 is associated with enhanced
contextual fear memory in the Prdx6−/− mice. Among
diﬀerential expression proteins listed on Additional
ﬁle2: TableS1, total TrkB is highly expressed in the hip-
pocampus of Prdx6−/− mice. is neurotrophin receptor
is encoded by the neurotrophic receptor tyrosine kinase
2 (Ntrk2) gene and plays an important role in neuronal
plasticity and fear memory . Piazza and colleagues
reported that the mice administered with stress hormone
GC exhibited enhanced fear response via the activation
TrkB/MAPK pathway . us, hyperactivation of
ERK1/2 in the absence of PRDX6 may be correlated with
increased TrkB level.
Interestingly, we also observed upregulation of cyto-
solic phospholipase A2 (cPLA2), a downstream target
of ERK1/2 , in the Prdx6−/− mice, which may be the
compensation eﬀect for the functional loss of aiPLA2-
PRDX6 . is increased cPLA2 level may promote
contextual fear memory retrieval in Prdx6−/− mice,
since blocking cPLA2 activity before memory test sup-
presses memory retrieval . TrkB signaling can also
activate phosphoinositide 3-kinase (PI3K) and protein
kinase B (AKT) [73, 74]. Blocking of PI3K reduced activa-
tion of ERK1/2 and AKT, in turn, impaired fear memory
retrieval . ese pieces of data suggest that activation
of TrkB signaling and its downstream molecules-ERK1/2,
cPLA2, and AKT in the hippocampus may help enhance
retrieval of fear memory in Prdx6−/− mice. Other brain
regions may also be responsible for the Prdx6−/− mice’s
enhanced contextual fear response, particularly the
amygdala, but the protein changes were not examined in
the present study. A further experiment is worth pursu-
ing the amygdala’s signiﬁcance in regarding this pheno-
type of the Prdx6−/− mice.
In conclusion, this study is the ﬁrst to report PRDX6′s
function in negative regulation of contextual fear mem-
ory along with hyperactivation of the MAPK pathway in
the hippocampus during the retrieval stage of contextual
memory. e results obtained from this study reveal the
physiological role of PRDX6 in memory formation and
help better understand the mechanism underlying home-
ostatic fear regulation. It also suggests that PRDX6 may
be a potential drug target for treating fear-dysregulated
disorders like PTSD.
The online version contains supplementary material available at https ://doi.
org/10.1186/s1304 1‑021‑00754 ‑1.
Additional le1: Figure S1–S6. The characteristics of Prdx6−/− mice (Fig.
S1). Expression of EGFP in the hippocampus of Prdx6−/− mice (Fig. S2).
Immunostaining of mPRDX6 expressed in the amygdala and prefrontal
cortex (Fig. S3). The level of reactive oxygen species (ROS) measured by
superoxide‑sensitive DHE staining in hippocampal CA1 and CA3 regions
of Prdx6+/+ and Prdx6−/− mice (Fig. S4). Co‑localization of PRDX6 with GFAP
in the amygdala (Fig. S5) and prefrontal cortex (Fig. S6).
Additional le2: TableS1–S5. This ﬁle contains the list of diﬀerential
expression proteins (DEPs) in the hippocampus of Prdx6+/+ and Prdx6−/−
mice after a contextual test (TableS1). First 25th enrich terms of GO
biological process of diﬀerential expression proteins (DEPs) (TableS2). Up‑
and down‑regulated enrich proteins diﬀerentially expressed relative to
GO biological process termed MAPK signaling pathway during contextual
memory retrieval in Prdx6−/− mice (TableS3). List of KEGG pathways for
GO biological process termed "protein phosphorylation" (GO:0006468) of
up‑and down‑regulated proteins (TableS4). And the details of antibodies
and vectors used in this study (TableS5).
aiPLA2: Acidic calcium‑independent phospholipase A2; ATF3: Activating
transcription factor 3; CA1: Cornu ammonis 1; CA3: Cornu ammonis 3; cPLA2:
Cytosolic phospholipase A2; DEPs: Diﬀerential expression proteins; DG: Den‑
tate gyrus; DHE: Dihydroethidium; DSM‑V: Diagnostic and Statistical Manual of
Mental Disorders, ﬁfth edition; ERK1/2: Extracellular signal‑regulated protein
kinases 1 and 2; FDR: False discovery rate; GC: Glucocorticoid; GO: Gene ontol‑
ogy; GPx: Glutathione peroxidase; LC/MS–MS: Liquid chromatography–tan‑
dem mass spectrometry; LPCAT : Lysophosphatidylcholine acyltransferase; LV‑
mPRDX6: Lentivirus‑carrying mouse PRDX6; MAPK: Mitogen‑activated protein
kinase; NTRK2: Neurotrophic receptor tyrosine kinase 2; PI3K: Phosphoinositide
3‑kinase; PRDX6: Peroxiredoxin 6; PTSD: Post‑traumatic stress disorder; ROS:
Reactive oxygen species.
The authors highly appreciated everyone for your contributions to this study.
We are also thankful for the service from the Core Research Laboratory, Tzu Chi
SP: conceptualized and performed experiments, analyzed data, and wrote the
manuscript. TP, PS: performed experiments and wrote parts of the manuscript.
CHY, SR: analyzed data and advised experimental design. HKC: performed
experiments. SPH: provided essential mouse strain and advised experimental
designs. CCC, NP, SC, SNT: advised experimental designs. IYL: supervised the
research, analyzed the data, and prepared manuscripts. All authors read and
approved the ﬁnal manuscript.
This work was supported by the Ministry of Science and Technology (MOST),
Taiwan (MOST‑107‑2410‑H320‑DOI‑MY3), and Tzu Chi University/Tzu Chi
Data and materials availability
The data that support the ﬁndings of this study are available from the cor‑
responding author upon reasonable request.
Ethics approval and consent to participate
All animal experiments were done following the Taiwan Ministr y of Science
and Technology guidelines for animals’ ethical treatment. Experiments were
approved by the Institutional Animal Care and Use Committee of Tzu Chi
University, Taiwan (Approval #104099).
Consent for publication
The authors report no biomedical ﬁnancial interests or potential conﬂicts of
1 Institute of Medical Sciences, Tzu Chi University, Hualien, Taiwan. 2 Depart‑
ment of Physiology, Faculty of Medicine Siriraj Hospital, Mahidol University,
Page 16 of 17
Phasuketal. Mol Brain (2021) 14:42
Bangkok, Thailand. 3 Department of Laboratory Medicine and Biotechnology,
Tzu Chi University, Hualien, Taiwan. 4 National Center for Genetic Engineering
and Biotechnology, National Science and Technology Development Agency,
Pathum Thani, Thailand. 5 Institute of Biomedical Sciences, Academia Sinica,
Taipei, Taiwan. 6 Department of Molecular Biology and Human Genetics, Tzu
Chi University, Hualien, Taiwan. 7 Department of Anatomy, Faculty of Medi‑
cine Siriraj Hospital, Mahidol University, Bangkok, Thailand. 8 Department
of Anatomy, Faculty of Medical Science, Naresuan University, Phitsanulok,
Thailand. 9 Centre of Excellence in Medical Biotechnology, Faculty of Medical
Science, Naresuan University, Phitsanulok, Thailand.
Received: 12 December 2020 Accepted: 16 February 2021
1. Jhang J, Lee H, Kang MS, Lee HS, Park H, Han JH. Anterior cingulate cortex
and its input to the basolateral amygdala control innate fear response.
Nat Commun. 2018;9(1):2744.
2. Raio CM, Phelps EA. The inﬂuence of acute stress on the regulation of
conditioned fear. Neurobiol Stress. 2015;1:134–46.
3. Stevenson CW. Role of amygdala‑prefrontal cortex circuitry in regulat‑
ing the expression of contextual fear memory. Neurobiol Learn Mem.
4. Etkin A, Wager TD. Functional neuroimaging of anxiety: a meta‑analysis
of emotional processing in PTSD, social anxiety disorder, and speciﬁc
phobia. Am J Psychiatry. 2007;164(10):1476–88.
5. Jovanovic T, Norrholm SD, Blanding NQ, Phifer JE, Weiss T, Davis M, et al.
Fear potentiation is associated with hypothalamic‑pituitary‑adrenal axis
function in PTSD. Psychoneuroendocrinology. 2010;35(6):846–57.
6. Mahan AL, Ressler KJ. Fear conditioning, synaptic plasticity and the
amygdala: implications for posttraumatic stress disorder. Trends Neurosci.
7. Fisher AB. Peroxiredoxin 6: a bifunctional enzyme with glutathione
peroxidase and phospholipase A(2) activities. Antioxid Redox Signal.
8. Arevalo JA, Vazquez‑Medina JP. the role of peroxiredoxin 6 in cell signal‑
ing. Antioxidants (Basel). 2018;7(12):172.
9. Asuni AA, Guridi M, Sanchez S, Sadowski MJ. Antioxidant peroxiredoxin
6 protein rescues toxicity due to oxidative stress and cellular hypoxia
in vitro, and attenuates prion‑related pathology in vivo. Neurochem Int.
10. Shanshan Y, Beibei J, Li T, Minna G, Shipeng L, Li P, et al. Phospholipase A2
of peroxiredoxin 6 plays a critical role in cerebral ischemia/reperfusion
inﬂammatory injury. Front Cell Neurosci. 2017;11:99.
11. Power JH, Asad S, Chataway TK, Chegini F, Manavis J, Temlett JA, et al.
Peroxiredoxin 6 in human brain: molecular forms, cellular distribution
and association with Alzheimer’s disease pathology. Acta Neuropathol.
12. Figueiredo HF, Dolgas CM, Herman JP. Stress activation of cortex and
hippocampus is modulated by sex and stage of estrus. Endocrinology.
13. Asuni AA, Gray B, Bailey J, Skipp P, Perry VH, O’Connor V. Analysis
of the hippocampal proteome in ME7 prion disease reveals a pre‑
dominant astrocytic signature and highlights the brain‑restricted
production of clusterin in chronic neurodegeneration. J Biol Chem.
14. Tan L, Zhao Y, Jiang B, Yang B, Zhang H. Knockdown of PRDX6 in microglia
reduces neuron viability after OGD/R injury. Xi Bao Yu Fen Zi Mian Yi Xue
Za Zhi. 2016;32(8):1014–20.
15. Hassan W, Gomes VdC, Pinton S, Batista Teixeira da Rocha J, Landeira‑
Fernandez J. Association between oxidative stress and contextual fear
conditioning in Carioca high‑ and low‑conditioned freezing rats. Brain
16. Schaeﬀer EL, Zorron PuL, Gagliotti DA, Gattaz WF. Conditioning training
and retrieval increase phospholipase A(2) activity in the cerebral cortex of
rats. J Neural Transm (Vienna). 2009;116(1):41–50.
17. Pai CS, Sharma PK, Huang HT, Loganathan S, Lin H, Hsu YL, et al. The
Activating Transcription Factor 3 (Atf3) homozygous knockout mice
exhibit enhanced conditioned fear and down regulation of hippocampal
GELSOLIN. Front Mol Neurosci. 2018;11:37.
18. Briscione MA, Jovanovic T, Norrholm SD. Conditioned fear associated
phenotypes as robust, translational indices of trauma‑, stressor‑, and
anxiety‑related behaviors. Front Psychiatry. 2014;5:88.
19. Haaker J, Maren S, Andreatta M, Merz CJ, Richter J, Richter SH, et al.
Making translation work: Harmonizing cross‑species methodology in
the behavioural neuroscience of Pavlovian fear conditioning. Neurosci
Biobehav Rev. 2019;107:329–45.
20. Dai JX, Han HL, Tian M, Cao J, Xiu JB, Song NN, et al. Enhanced contextual
fear memory in central serotonin‑deﬁcient mice. Proc Natl Acad Sci USA.
21. Yoshii A, Constantine‑Paton M. Postsynaptic BDNF‑TrkB signal‑
ing in synapse maturation, plasticity, and disease. Dev Neurobiol.
22. He Y, Xu W, Xiao Y, Pan L, Chen G, Tang Y, et al. Overexpression of perox‑
iredoxin 6 (PRDX6) promotes the aggressive phenotypes of esophageal
squamous cell carcinoma. J Cancer. 2018;9(21):3939–49.
23. Wang X, Phelan SA, Forsman‑Semb K, Taylor EF, Petros C, Brown A, et al.
Mice with targeted mutation of peroxiredoxin 6 develop normally but are
susceptible to oxidative stress. J Biol Chem. 2003;278(27):25179–90.
24. Vyas Y, Lee K, Jung Y, Montgomery JM. Inﬂuence of maternal zinc sup‑
plementation on the development of autism‑associated behavioural
and synaptic deﬁcits in oﬀspring Shank3‑knockout mice. Mol Brain.
25. Lebow M, Neufeld‑Cohen A, Kuperman Y, Tsoory M, Gil S, Chen A. Sus‑
ceptibility to PTSD‑like behavior is mediated by corticotropin‑releasing
factor receptor type 2 levels in the bed nucleus of the stria terminalis. J
26. Katano T, Takao K, Abe M, Yamazaki M, Watanabe M, Miyakawa T, et al.
Distribution of Caskin1 protein and phenotypic characterization of its
knockout mice using a comprehensive behavioral test battery. Mol Brain.
27. Huang CH, Chiang YW, Liang KC, Thompson RF, Liu IY. Extracellular
signal‑regulated kinase 1/2 (ERK1/2) activated in the hippocampal CA1
neurons is critical for retrieval of auditory trace fear memory. Brain Res.
28. Sukkaew A, Suksatu A, Roytrakul S, Smith DR, Ubol S. Proteomic analysis
of CHIKV‑infected human ﬁbroblast‑like synoviocytes: identiﬁcation of
host factors potentially associated with CHIKV replication and cellular
pathogenesis. Microbiol Immunol. 2020;64(6):445–57.
29. Tyanova S, Temu T, Sinitcyn P, Carlson A, Hein MY, Geiger T, et al. The Per‑
seus computational platform for comprehensive analysis of (prote)omics
data. Nat Methods. 2016;13(9):731–40.
30. Losuwannarak N, Maiuthed A, Kitkumthorn N, Leelahavanichkul A, Roytra‑
kul S, Chanvorachote P. Gigantol targets cancer stem cells and destabi‑
lizes tumors via the suppression of the PI3K/AKT and JAK/STAT pathways
in ectopic lung cancer xenografts. Cancers (Basel). 2019;11(12):2032.
31. Forester CM, Zhao Q, Phillips NJ, Urisman A, Chalkley RJ, Oses‑Prieto JA,
et al. Revealing nascent proteomics in signaling pathways and cell dif‑
ferentiation. Proc Natl Acad Sci USA. 2018;115(10):2353–8.
32. Dybas JM, O’Leary CE, Ding H, Spruce LA, Seeholzer SH, Oliver PM. Inte‑
grative proteomics reveals an increase in non‑degradative ubiquitylation
in activated CD4(+) T cells. Nat Immunol. 2019;20(6):747–55.
33. Bardou P, Mariette J, Escudie F, Djemiel C, Klopp C. jvenn: an interactive
Venn diagram viewer. BMC Bioinform. 2014;15:293.
34. Mi H, Muruganujan A, Ebert D, Huang X, Thomas PD. PANTHER version 14:
more genomes, a new PANTHER GO‑slim and improvements in enrich‑
ment analysis tools. Nucleic Acids Res. 2019;47(D1):D419–26.
35. Howe EA, Sinha R, Schlauch D, Quackenbush J. RNA‑Seq analysis in MeV.
36. Seo JS, Park JY, Choi J, Kim TK, Shin JH, Lee JK, et al. NADPH oxidase medi‑
ates depressive behavior induced by chronic stress in mice. J Neurosci.
37. Yau SY, Bettio L, Vetrici M, Truesdell A, Chiu C, Chiu J, et al. Chronic
minocycline treatment improves hippocampal neuronal structure, NMDA
receptor function, and memory processing in Fmr1 knockout mice.
Neurobiol Dis. 2018;113:11–22.
38. Han RW, Liu ZP, Lin HR, Tian AW, Xiao YF, Wei J, et al. Role of lateral amyg‑
dala calstabin2 in regulation of fear memory. Mol Brain. 2020;13(1):35.
Page 17 of 17
Phasuketal. Mol Brain (2021) 14:42
39. Kikuchi M, Takase K, Hayakawa M, Hayakawa H, Tominaga SI, Ohmori
T. Altered behavior in mice overexpressing soluble ST2. Mol Brain.
40. Zhang JB, Chen L, Lv ZM, Niu XY, Shao CC, Zhang C, et al. Oxytocin is
implicated in social memory deﬁcits induced by early sensory depriva‑
tion in mice. Mol Brain. 2016;9(1):98.
41. Miller MW, Wolf EJ, Logue MW, Baldwin CT. The retinoid‑related orphan
receptor alpha (RORA) gene and fear‑related psychopathology. J Aﬀect
42. Besnard A, Caboche J, Laroche S. Recall and reconsolidation of contextual
fear memory: diﬀerential control by ERK and Zif268 expression dosage.
PLoS ONE. 2013;8(8):e72006.
43. Mazzucchelli C, Brambilla R. Ras‑related and MAPK signalling in neuronal
plasticity and memory formation. Cell Mol Life Sci. 2000;57(4):604–11.
44. Hamezah HS, Durani LW, Yanagisawa D, Ibrahim NF, Aizat WM, Bellier JP,
et al. Proteome proﬁling in the hippocampus, medial prefrontal cortex,
and striatum of aging rat. Exp Gerontol. 2018;111:53–64.
45. Regier DA, Kuhl EA, Kupfer DJ. The DSM‑5: classiﬁcation and criteria
changes. World Psychiatry. 2013;12(2):92–8.
46. Cahill SP, Pontoski K. Post‑traumatic stress disorder and acute stress disor‑
der I: their nature and assessment considerations. Psychiatry (Edgmont).
47. Ota Y, Zanetti AT, Hallock RM. The role of astrocytes in the regula‑
tion of synaptic plasticity and memory formation. Neural Plast.
48. Ben Menachem‑Zidon O, Avital A, Ben‑Menahem Y, Goshen I, Kreisel T,
Shmueli EM, et al. Astrocytes support hippocampal‑dependent memory
and long‑term potentiation via interleukin‑1 signaling. Brain Behav
49. Pankiewicz JE, Diaz JR, Marta‑Ariza M, Lizinczyk AM, Franco LA, Sadowski
MJ. Peroxiredoxin 6 mediates protective function of astrocytes in Abeta
proteostasis. Mol Neurodegener. 2020;15(1):50.
50. Yun HM, Choi DY, Oh KW, Hong JT. PRDX6 exacerbates dopaminergic
neurodegeneration in a MPTP mouse model of Parkinson’s disease. Mol
51. Jimenez JC, Berry JE, Lim SC, Ong SK, Kheirbek MA, Hen R. Contextual fear
memory retrieval by correlated ensembles of ventral CA1 neurons. Nat
52. Raineki C, Holman PJ, Debiec J, Bugg M, Beasley A, Sullivan RM. Functional
emergence of the hippocampus in context fear learning in infant rats.
53. Maren S, Fanselow MS. Appetitive motivational states diﬀer in their ability
to augment aversive fear conditioning in rats (Rattus norvegicus). J Exp
Psychol Anim Behav Process. 1998;24(3):369–73.
54. McHugh TJ, Tonegawa S. Spatial exploration is required for the formation
of contextual fear memory. Behav Neurosci. 2007;121(2):335–9.
55. Green TA, Alibhai IN, Unterberg S, Neve RL, Ghose S, Tamminga CA, et al.
Induction of activating transcription factors (ATFs) ATF2, ATF3, and ATF4
in the nucleus accumbens and their regulation of emotional behavior. J
56. Hlushchenko I, Koskinen M, Hotulainen P. Dendritic spine actin dynamics
in neuronal maturation and synaptic plasticity. Cytoskeleton (Hoboken).
57. Daskalakis NP, Cohen H, Cai G, Buxbaum JD, Yehuda R. Expression proﬁl‑
ing associates blood and brain glucocorticoid receptor signaling with
trauma‑related individual diﬀerences in both sexes. Proc Natl Acad Sci
58. Revest JM, Kaouane N, Mondin M, Le Roux A, Rouge‑Pont F, Vallee M,
et al. The enhancement of stress‑related memory by glucocorticoids
depends on synapsin‑Ia/Ib. Mol Psychiatry. 2010;15(12):1125, 40–51.
59. Schutsky K, Ouyang M, Castelino CB, Zhang L, Thomas SA. Stress and glu‑
cocorticoids impair memory retrieval via beta2‑adrenergic, Gi/o‑coupled
suppression of cAMP signaling. J Neurosci. 2011;31(40):14172–81.
60. Chowdhury I, Fisher AB, Christoﬁdou‑Solomidou M, Gao L, Tao JQ,
Sorokina EM, et al. Keratinocyte growth factor and glucocorticoid
induction of human peroxiredoxin 6 gene expression occur by
independent mechanisms that are synergistic. Antioxid Redox Signal.
61. Massaad CA, Klann E. Reactive oxygen species in the regulation of synap‑
tic plasticity and memory. Antioxid Redox Signal. 2011;14(10):2013–54.
62. Rajasethupathy P, Sankaran S, Marshel JH, Kim CK, Ferenczi E, Lee SY,
et al. Projections from neocortex mediate top‑down control of memory
retrieval. Nature. 2015;526(7575):653–9.
63. Medvedeva YV, Ji SG, Yin HZ, Weiss JH. Diﬀerential vulnerability of CA1
versus CA3 pyramidal neurons after ischemia: possible relationship to
sources of Zn accumulation and its entry into and prolonged eﬀects on
mitochondria. J Neurosci. 2017;37(3):726–37.
64. Lopez J, Gamache K, Schneider R, Nader K. Memory retrieval requires
ongoing protein synthesis and NMDA receptor activity‑mediated AMPA
receptor traﬃcking. J Neurosci. 2015;35(6):2465–75.
65. Fortin DA, Srivastava T, Dwarakanath D, Pierre P, Nygaard S, Derkach VA,
et al. Brain‑derived neurotrophic factor activation of CaM‑kinase kinase
via transient receptor potential canonical channels induces the transla‑
tion and synaptic incorporation of GluA1‑containing calcium‑permeable
AMPA receptors. J Neurosci. 2012;32(24):8127–37.
66. Boondam Y, Songvut P, Tantisira MH, Tapechum S, Tilokskulchai K, Paka‑
prot N. Inverted U‑shaped response of a standardized extract of Centella
asiatica (ECa 233) on memory enhancement. Sci Rep. 2019;9(1):8404.
67. Slouzkey I, Maroun M. PI3‑kinase cascade has a diﬀerential role in acquisi‑
tion and extinction of conditioned fear memory in juvenile and adult rats.
Learn Mem. 2016;23(12):723–31.
68. Yun HM, Park KR, Lee HP, Lee DH, Jo M, Shin DH, et al. PRDX6 promotes
lung tumor progression via its GPx and iPLA2 activities. Free Radic Biol
69. Yamada K, Nabeshima T. Brain‑derived neurotrophic factor/TrkB signaling
in memory processes. J Pharmacol Sci. 2003;91(4):267–70.
70. Andersen JM, Myhre O, Fonnum F. Discussion of the role of the extracel‑
lular signal‑regulated kinase‑phospholipase A2 pathway in production
of reactive oxygen species in Alzheimer’s disease. Neurochem Res.
71. Blanchard H, Taha AY, Cheon Y, Kim HW, Turk J, Rapoport SI. iPLA2beta
knockout mouse, a genetic model for progressive human motor
disorders, develops age‑related neuropathology. Neurochem Res.
72. Fujita S, Ikegaya Y, Nishiyama N, Matsuki N. Ca2+‑independent phos‑
pholipase A2 inhibitor impairs spatial memory of mice. Jpn J Pharmacol.
73. Chen X, Garelick MG, Wang H, Lil V, Athos J, Storm DR. PI3 kinase signaling
is required for retrieval and extinction of contextual memory. Nat Neuro‑
74. Nakai T, Nagai T, Tanaka M, Itoh N, Asai N, Enomoto A, et al. Girdin
phosphorylation is crucial for synaptic plasticity and memory: a potential
role in the interaction of BDNF/TrkB/Akt signaling with NMDA receptor. J
Springer Nature remains neutral with regard to jurisdictional claims in pub‑
lished maps and institutional aﬃliations.