RES E A R C H Open Access
The effect of abscisic acid chronic
treatment on neuroinflammatory markers
and memory in a rat model of high-fat diet
induced neuroinflammation
Sandra Sánchez-Sarasúa
1†
, Salma Moustafa
1†
, Álvaro García-Avilés
1
, María Fernanda López-Climent
2
,
Aurelio Gómez-Cadenas
2
, Francisco E. Olucha-Bordonau
1
and Ana M. Sánchez-Pérez
1*
Abstract
Background: Western diet and lifestyle are associated with overweight, obesity, and type 2 diabetes, which, in turn,
are correlated with neuroinflammation processes. Exercise and a healthy diet are important in the prevention of these
disorders. However, molecules inhibiting neuroinflammation might also be efficacious in the prevention and/or
treatment of neurological disorders of inflammatory etiology. The abscisic acid (ABA) is a phytohormone involved in
hydric-stress responses. This compound is not only found in plants but also in other organisms, including mammals. In
rodents, ABA can play a beneficial role in the regulation of peripheral immune response and insulin action. Thus, we
hypothesized that chronic ABA administration might exert a protective effect in a model of neuroinflammation
induced by high-fat diet (HFD).
Methods: Male Wistar rats were fed with standard diet or HFD with or without ABA in the drinking water for 12 weeks.
Glucose tolerance test and behavioral paradigms were performed to evaluate the peripheral and central effects of
treatments. One-Way ANOVA was performed analyzed statistical differences between groups.
Results: The HFD induce d insu lin re sistance periph erally and increased the levels of proinflammatory markers
in in the brain. We o bserved that A BA restor ed glucose tolerance in HFD-fed rats, as expected. In addition,
chronic ABA treatment rescued cognitive p erformance in these animals, whil e not affecting control diet fed
animals. Moreover, it counteracted the changes induced by HFD in the hypothalamus; microglia activations
and TNFα mRNA levels.
Conclusion: These results suggest that ABA might be come a new therapeutic molecule improving the
neuroinflammatory status and insulin resistance.
Keywords : Obesity, Working memory, Microglia, Phytohormones
Background
Obesity, a leading cause of type 2 diabetes [1], correlates
with cognitive impairment. Insulin sensitizers have been
proposed as a promising tool for the redu ction of
obesity-induced insulin resistance and inflammation
processes. The thiazolidinediones (TDZ) are a family of
synthetic insulin sensitizer molecules; however, some of
them have undesirable side effects [2–4]. Thus, alterna-
tive compounds with analogous properties but fewer
side effects are needed. The phytohormone abscisic acid
(ABA) was found in mammalian cells more than 25 years
ago [5]. Since then, several studies have proposed it as a
universal signaling molecule [6, 7]. Structurally, ABA is
very similar to the TDZs. Indeed, ABA can improve
glucose tolerance [8], reduce the level of TNFα, and
decrease adipocyte cell size in an in vivo model of obes-
ity induced by HFD [9]. Moreover, in human and murine
* Correspondence: sanchean@uji.es
†
Equal contributors
1
Department of Medicine, University of Jaume I, Vicente Sos Banyat s/n,
12071 Castellón de la Plana, Castellón, Spain
Full list of author information is available at the end of the article
© The Author(s). 2016 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Sánchez-Sarasúa et al. Nutrition & Metabolism (2016) 13:73
DOI 10.1186/s12986-016-0137-3
pancreatic cell lines (RIN-m and INS-2 cells), ABA can
increase glucose-stimulated insulin secretion [10]. This
effect can be repressed using pertussis toxin and PKA
inhibitors [11]. Dietary ABA also stimulates granulocyte
function and mac rophage infiltration in the adipose tis-
sue [12]. In mammalian cells, the lanthionine synthetase
C-like protein 2 (LANCL2) shows high homology with
the ABA receptor in plants, the Arabidopsis GCR2.
Silencing the expression of endo genous LANCL2 in
granulocyte cells can abrogate ABA induction of Ca
2+
responses, whereas overexpression of LANCL2 enhances
the ABA-mediated effects [13]. Because of its role in the
mediation of ABA effects, LANCL-2 has been proposed
as a therapeutic target for the treatment of infl ammatory
diseases and diabetes [14].
Furthermore, ABA shows some molecular structural
similarities to retinoic acid (RA). RA has beneficial cog-
nitive effects, rectifying memory deficits in rodent
models of Alzheimer disease. However, the clinical and
animal model data show an association betw een RA
administration and the symptoms of depressio n [15].
Chronic treatment with ABA has beneficial antidepres-
sant effects demonstrated by increased sucrose intake,
increased swimming in the forced swim test, and re-
duced expression of CRH and RARα mRNA in the rat
hypothalamus in control rats, with no reported side ef-
fect s [16]. Moreover, preliminary data show an improved
spatial memory in rats treated with ABA [17]. On the
basis of these data, we hypothesized that as ABA can
modulate peripheral insulin resistance and immune re-
sponse, it might also exert similar action centrally. Thus,
it this work aims to show whether dietary ABA could re-
store cognitive deficits resulting from a high fat diet
(HFD) induced neuroinflammation. HFD elevates the
levels of neuroinflammation markers in the brain [18]
and it might constitute a link between obesity and
degenerative disorders via insulin resistance [19]. Fur-
thermore, HFD has been shown to induce memory loss
through elevation of inflammatory markers in hippo-
campus [20].
We administered ABA (20 mg/L) in the drinking water
to rats on either HFD or standard diet (SD). After eleven
weeks of treatment, we compared the behavior of the
four groups using two memory paradigms: the novel ob-
ject recognition (NOR) and the T-maze. In addition, we
measured ABA levels in the blood and cerebellum of all
four groups using HPLC. We analyzed the microglia
proliferation using immunohistochemistry. We demon-
strated that the ABA administered in drinking water im-
proved glucose tolerance and cognitive performance and
decreased the levels of inflammatory markers in the
hypothalamic areas. Our results confirm a therapeutic
potential of this phytohormon e in the peripheral meta-
bolic alterations. The data also strongly suggest the
potential beneficial effects of ABA in disorde rs of neu-
roinflammatory etiology, which has not been demon-
strated before.
Methods
The aim of the experiment is to evaluate the central
effects of the phytohormone, ABA in a model of neuro-
inflammation elicited by HFD, during 12-week treat-
ment. Behavioral tests started on the 11
th
week, and
sacrifices were carried out in the 12
th
week of ABA and
diet treatm ent (Fig. 1).
Animals and diet
Eight-week-old male Wistar rats were obtained from the
Janvier Labs (Saint-Berthevin, France) and kept at the
animal facility of the University Jaume I. The procedures
followed the directive 86/609/EEC of the European
Community on the protection of animals used for ex-
perimental and other scientific purposes. The experi-
ments were approved by the Ethics Committee of the
University Jaume I. The animals were maintained on a
12 h:12 h light–dark cycle and housed in pairs to reduce
stress due to social isolation. Rats were divided randomly
into four experimental groups:SD,controlanimalsfedthe
standard rodent diet (Ssniff, Soest, Germany); SD-ABA,
animals fed standard diet supplemented with ABA
(Fernandez-Rapado, Spain) in the drinking water (20 mg/
L); HFD, animals fed a high-fat diet (5736 kcal/kg, Ssniff)
(Table 1); and HFD-ABA, animals fed high fat diet and
ABA in drinking water (20 mg/L). We based our estima-
tions in previous papers where ABA had been administered
in the food pellet. Considering the daily food (20 g/day) and
water average intake (50 mL/day), we estimated the amount
of ABA per mL of drinking water, in order to doses in the
same range as previous papers (100 mg/Kg) [21]. In Guri
study, the period of ABA treatment was shorter (36 days)
Fig. 1 Experiment design. Behavioral tests started in the 11
th
week, and the animals were sacrificed in the 11
th
and 12
th
week
Sánchez-Sarasúa et al. Nutrition & Metabolism (2016) 13:73 Page 2 of 11
than our study (12 weeks), therefore we considered that the
ABA administration could be lowered (1 mg/day) to
achieve optimal chronic effects. The four groups were fed
ad libitum for 12 weeks. The HFD diet induces inflamma-
tory effects in the peripheral tissues [21] and the brain [22].
The body weight and food and water intake were moni-
tored twice a week per cage.
Novel object recognition (NOR)
Training chamber consisted of a black wooden box
(81 × 61 × 52.3 cm). Rats were habituated to the testing
environment for 30 min on two consecutive days. On
the first day, the rat s spent a 10-min habituation period
in the experimental box, without objects. On the second
day, each rat was placed in the box for the familiarization
phase and allowed to explore two identical objects (heavy
metallic jars). One group (easy task group) was left for
10 min, and the second group (difficult task group) was
left to explore for 3 min. All the rats were then returned
to their home cages for a 1-h intertrial interval. The box
and objects were cleaned with a 30 % ethanol solution. In
the test phase, the rats were returned to the box and
allowed to explore a familiar and novel object for further
5 min (easy task group) or 3 min (difficult task group)
[23]. Familiar and novel objects were alternated between
left and right to prevent location predisposition. Both trial
and test phases were recorded using a video tracking
system (Smart 2.5.19, Panlab, Barcelona, Spain) for subse-
quent behavioral analysis. Exploration was defined as time
spent sniffing within 1–4 cm of the object or touching it,
always with the head oriented towards the object. Climb-
ing over the object or running around it was not consid-
ered exploration.
T-maze
The T-maze was a three-arm maze; one arm (119.3 ×
18.2 cm) was longer than the other two, which were
identical (21.1 × 34.5 cm); the entire maze was placed
above the floor. The longer arm was chosen a s the start
arm. The rats were habituated to the behavior room for
30 min. On the test day, the animals were allowed to
explore the maze for 5 min, with access to two of the
three arms (the home or start arm and the familiar arm).
The rat was then returned to its home cage for a 2-h in-
tertrial interval (difficult task group) or 90-min intertrial
(easy task group), during which the maze was cleaned
with 30 % ethanol. The rat was then placed back in the
maze; this time the animal had access to all arms for
5 min [24]. The number of entries to the novel arm and
the time rat spent in each arm was recorded using a
video tracking system (Smart 2.5.19).
Glucose tolerance test
Rats were fasted overnight, and a drop of blood was
taken from the tail before (t = 0) and 30 min (t = 30) and
120 min (t = 120) after glucose administration (2 g/kg).
Plasma glucose was measured using Glucomen LX Plus
glucometer.
HPLC ABA measurements
ABA was analyzed in all four groups using LC/ESI-
MSMS, essentially as described in [25], with slight
modifications. Briefly, 1 g of frozen tissue was extracted
in ultrapure water using a tissue homogenizer (Ultra-
Turrax, Ika-Werke, Staufen, Germany) after spiking with
100 ng of d
6
-ABA. After extraction and centrifugation,
pH of the supernatant was adjusted to 3.0, and it was par-
titioned twice against diethyl-ether (Panreac, Barcelona,
Spain). The organic layers were combined and evaporated
in a centrifuge vacuum evaporator (Jouan, Saint-Herblain,
France). The dry residue was then resuspended in water–
methanol (9:1) solution, filtered and injected into a
UPLC™ Acquity system (Waters, Milford, MA, USA). The
analyte was then separated on a reversed phase UPLC
C18 column (Nucleodur C18, 1.8 μm, 50 × 2.0 mm,
Macherey-Nagel, Barcelona, Spain). The solvents were
methanol and water supplemented with 0.01 % acetic acid,
applied at a flow rate of 300 μLmin
−1
. ABA was quanti-
fied with a Quattro LC triple quadrupole mass spectrom-
eter (Micromass, Waters, Manchester, UK) connected
online to the output of the column through an orthogonal
Z-spray electrospray ion source. Quantitation of this hor-
mone was achieved by external calibration with known
amounts of standards.
Immunocytochemistry (ICC)
Rats were anesthetized with an overdose of pentobarbital
(120 mg/kg Eutanax, Fatro, Barcelona, Spain) and trans-
cardially perfused with saline (0.9 %) followed by 4 %
paraformaldehyde (PFA) fixative in 0.1 M phosphate
buffer, pH 7.4. After the perfusion, the brains were re-
moved from the skulls and postfixed overnight at 4 °C in
PFA. Then, the brains were immersed in 30 % sucrose
for 48 h for cryoprotection. Sliding Microtome Leica
SM2010R (Leica Microsystems, Heidelberg, Germany)
Table 1 Composition of high fat diet
Crude Nutrients % Additives per kg
Crude protein 24.4 Vitamin A (IU) 15000
Crude fat 34.6 Vitamin D
3
(IU) 1500
Crude fibre 6.0 Vitamin E (mg) 150
Crude ash 5.5 Vitamin K
3
(mg) 20
Starch 0.1 Vitamin C (mg) 30
Sugar 9.4 Copper (mg) 12
Fat 60 kJ%
Energy 21.6 MJ (or kcal) ME/kg
Sánchez-Sarasúa et al. Nutrition & Metabolism (2016) 13:73 Page 3 of 11
was used to obtain 40-μm-thick coronal frozen sections.
The brains were cut in rostrocaudal direction; six series
of slices were collected from each brain and store d at
−20 °C for analysis.
For Iba1 staining, we used a goat anti-Iba1 (Abcam,
Cambridge, UK). Briefly, free-floating sections were
rinsed twice in 0.05 M Tris-buffered saline (TBS,
pH 8.0) and once in TBS with 0.2 % Triton X-100, at
room temperature . Sections were incubated in 4 % nor-
mal donkey serum for 1 h to reduce nonspecific labeling.
Afterward, the sections were incubated in the primary
antibody solution diluted 1:500 in 0.01 M phosphate
buffered saline (PBS) containing 2 % normal donkey
serum, TBS with 0.2 % Triton X-100, and 2 % bovine
serum albumin for 24 h at room temperature. After
washing off the excess of the primary antibody, the
sections wer e incubated in biotinylated donkey anti-goat
secondary antibody (Jackson) (1:200 in TBS, 0.2 %
Triton X-100). Two hours later, the sections were
transferred to the avidin–biotin–horseradish peroxid-
ase complex solution (Standard Elite ABC kit, Vector
Laboratories Burlingame, CA USA) for 90 min, followed
by two rinses with TBS with 0.2 % Triton X-100, and two
more with 0.05 M Tris/HCl pH 8.0. Then, the sections
were processed in 0.05 M Tris/HCl pH 8.0 containing
3.125 mg of DAB and 2 μLofH
2
O
2
for 15–20 min. The
reaction was stopped by adding an excess of 0.05 M Tris/
HCl pH 8.0, followed by several rinses in PBS. Finally, the
sections were mounted onto gelatin-coated slides, air-
dried, dehydrated in alcohol, cleared in xylene, and cover-
slipped with DPX mounting medium.
RNA extraction and RT-PCR
Total RNA was extracted from the hypothalamus (n =4–6
per group) using RNeasy Lipid Tissue Mini Kit (Qiagen,
Valencia, CA, USA) according to the manufacturer’s
protocol. The RNA samples were resuspended in 50 μLof
nuclease-free water. RNA concentration and quantifica-
tion of total RNA was performed using Thermo Scientific
NanoDrop 2000c, with the OD260/OD280. Genomic
DNA was removed using DNase I, RNase-free (Life Tech-
nologies, USA), for 30 min at 37 °C. The reaction was
stopped by addition of 1 μl of EDTA for 10 min at 65 °C.
The first strand cDNA was synthesized using the Prime-
Script™ RT Reagent Kit (Perfect Real Time) (Takara Bio
Inc., Shiga Japan). For each reaction, 1 μg of RNA was
used for reverse transcription, in a mixture of 4 μLof5×
PrimeScript Buffer; 1 μl of PrimeScript RT, 1 μL of Oligo
dT Primer (50 μM), and 1 μL of random primer
(100 μM). Enzyme mix was adjusted to a final volume of
20 μL at room temperature. The mixture was incubated at
37 °C for 15 min and heated at 85 °C for 15 min to
terminate the reaction. The cDNA was subsequently
stored at −20 °C. RT-PCR was performed in a volume of
10 μLwith5μL of Maxima SYBR Green/ROX qPCR
Master Mix (2X) (Applied Biosystems Life Technologies,
Carlsbad, CA, USA), 1 μL of primer and 1 μLofcDNA.
All PCR reactions were performed under the following
conditions: initial cycle at 98 °C for 10 min followed by
40 cycles at 98 °C for 10 s, 60 °C for 10 s, and 72 °C for
20s. Gene expression in the hypothalamus and hippocam-
pus was quantified using a StepOnePlus Real-Time PCR
system (Applied Biosystems Life Technologies). The RT-P
CR primers for TNFα were Forward 5′GACCCTCAC
ACTCAGATCATCTTCT3′ and reverse 5′TGCTACG
ACGTGGGCTACG3′. Each sample was tested in tripli-
cate. Data were analyzed using the comparative critical
threshold method, with the amount of target gene nor-
malized to the housekeeping gene ß-actin. Relative gene
expression was calculated using 2 -
ΔΔCt
relative to control.
Results
Animals were weighed, and water and food consumption
was monitored twice a week. Behavioral tests started on
the 11
th
week, and sacrifices were carried out in the 12
th
week of ABA and diet treatme nt (Fig. 1). As expected,
overnutrition affected the body weight, but ABA admin-
istration did not affect the weig ht gain in either group.
The results are presented as the means ± SE (n = 16 per
group). Control animals fed SD increased their body
weight from 444 ± 11.7 g (week 1) to 585 ± 12.0 g (week
10). Similarly, animals on SD supplemented with ABA
increased their body weight from 446 ± 10.7 g (week 1)
to 593 ± 10.2 g (week 10). This represents an increment
of 132 ± 2.4 % and 133 ± 1.5 %, for SD and SD with
ABA, respectively. However, animals fed HFD and HFD-
ABA increased their body weight from 448 ± 11.7 and
453 ± 12.2 g (week 1) to 659 ± 16.0 and 669 ± 16.2 g
(week 10), an increase of 148 ± 3.2 % and 148 ± 2.8 %, re-
spectively (Fig. 2a). Data were analyzed using a two-way
ANOVA; time (F
(10,600)
=994, p < 0.0001) and diet
(F
(3,60)
= 6.08, p = 0.001) had a significant effect. Food
and water consumption was mea sured per cage. Food
intake data is represented by the mean (g) of food con-
sumed ± SE per cage each week (n = 8 per group).
Weekly consumption was steady during the 12 weeks
treatment, but the diet clearly affected food intake. The
data were analyzed using two-way ANOVA (F
(3,28)
=75.36
p < 0.0001). To determine the ABA intake, we monitored
weekly water consumption (Fig. 2c). Based on this infor-
mation, we calculated ABA intake for both HFD and SD
fed animals, and confirmed that both groups had a similar
average weekly intake of the hormone, 12.30 ± 1.4 and
13.27 ± 0.8 mg/week/cage, respectively (Fig. 2d).
It has been reported that dietary ABA given in the
food pellets (100 mg/kg food) can improve glucose toler-
ance [21]. To confirm that, in our model, ABA (20 mg/L
of drinking water) improved the glucose homeostasis,
Sánchez-Sarasúa et al. Nutrition & Metabolism (2016) 13:73 Page 4 of 11
we performed a glucose tolerance test by intraperitoneal
injection of glucose (2 mg/kg) in fasted animals. After
11 weeks of treatment, the animals fasted overnight
(12–13 h). The basal glucose level s were similar in all
groups; 30 min after glucose injection, blood glucose
levels increased. No significant differences were ob-
served between the groups. However, two hours after
injection, glucose levels in HFD group remained higher
(235.4 ± 36 mg/dL) than in HFD-AB A group (158.9 ±
11.23 mg/dL). In the latter group, the glucose levels
were similar to those in control groups, SD (143.9 ±
10.24 mg/dL) and SD-ABA (152 .6 ± 15.52 mg/dL).
The data were analyzed using one-way A NOVA, with
Newman–K euls post-hoc tes t (p <0.05) (Fig. 3).
HFD can induce memory impairments in rodents [26]
and humans [27]. To examine the effect of chronic ABA
administration on cognitive performance, the anim als
were subjected to two behavioral paradigms that evalu-
ate the memory in rodents, NOR and T-maze test. NOR
exploits the innate exploratory preference of novel ob-
jects exhibited by rodents. This paradigm examines the
capability of the animal to remember a familiar object
when presented with a new one. We observed no signifi-
cant differences in the time spent exploring the identical
objects during familiarization phase (Fig. 4a and c).
During the test, all four groups spent much more time
exploring the novel object than the familiar one, and
neither the diet nor ABA treatment changed these pa-
rameters (Fig. 4b). However, we observed differences
when both the familiarization time and test time were
reduced to 3 min. In the HFD group, the times for
familiar and novel object did not differ significantly, sug-
gesting impairment in remembering the familiar object.
HFD-ABA animals behaved in the same way as the con-
trol groups (SD and SD-ABA), indicating that ABA
could abroga te the HFD-induced impairment (Fig. 4d).
In the T-maze test, we recorded the number of entries
to both maze arms, one of which was familiar and the
other was novel because it had been closed during the
habituation. We observed that rats fed SD, SD-ABA, and
HFD-ABA had a significantly higher number of en tries
to the novel arm than to the familiar one . This was not
the case for the animals fed HFD only, suggesting that this
diet might impair the memory of the familiar arm (Fig. 5a).
The data were analyzed using unpaired Student’s t-test,
comparing familiar and novel arm; *p <0.05<**p <0.01.
Interestingly, when the test was performed with a longer
inter-trial time between familiarization and test phases,
which may be a more difficult working memory task,
HFD-ABA did not rescue the alternation behavior shown
by the HFD group. We found that the difference between
the number of entries to the two arms was no longer sig-
nificant for HFD-ABA animals, in similarity with HFD
group. Both SD and SD-ABA group maintained a signifi-
cant difference in the exploratory behavior, entering the
novel arm more often (Fig. 5b).
ABA levels in the blood and brain (cerebellum) were
measured using HPLC. We detected variable amounts of
circulating ABA in treated animals. However, we were
unable to detect ABA under the same conditions in
untreated animals, confirming that we, indeed, were
observing exogenous ABA (Table 2). It has been
Fig. 2 Body weight in grams (a), food intake (b), water intake (c) and ABA weekly comsumption per cage (d) of rats fed high-fat diet (HFD, white
circles); HFD with ABA (HFD-ABA, black circles); standard diet (SD, white triangles), and SD-ABA (black triangles) for 10–11 weeks; n = 16 per group.
Values are mean ± SEM. #, p < 0.05; ##, p < 0.01 for HFD vs. SD. *, p < 0.05; **, p < 0.01, ***, p < 0.001 for HFD-ABA vs. SD
Sánchez-Sarasúa et al. Nutrition & Metabolism (2016) 13:73 Page 5 of 11