© 2018 by Acta Neurobiologiae Experimentalis
Detrimental eects of chia (Salvia hispanicaL.)
seeds on learning and memory in aluminum
chloride‑induced experimental Alzheimer’s disease
Yasemin Bilgic1, Enver Ahmet Demir1*, Nilufer Bilgic2, Hatice Dogan1, Okan Tutuk1 and Cemil Tumer1
1 Department of Physiology, Faculty of Medicine, Hatay Mustafa Kemal University, Hatay, Turkey,
2 Department of Molecular Biochemistry and Genetics, Faculty of Medicine, Hatay Mustafa Kemal University, Hatay, Turkey,
* Email: firstname.lastname@example.org
Polyphenols and omega‑3 fatty acids are thought to have benecial eects in Alzheimer’s disease, the most common cause of
dementia. Seeds of chia (Salvia hispanicaL.) are highly rich in these nutrients, and thus, the present study investigated the eects of
chia seeds on behavior and cognition in an aluminum‑induced Alzheimer’s disease model in rats. Experimental animals received chia
supplementation either during the generation of the model (i.e., pretreatment) or after the model was established (i.e., treatment).
Abattery of behavioral and cognitive tests were performed, including open‑eld, elevated plus maze, Porsolt’s forced swim, and
Morris’ water maze, to evaluate anxiety‑ and depression‑like behaviors, and learning and memory. Results showed that chia
supplementation was ineective against Alzheimer’s‑related anxiety, whereas depression‑like behaviors were attenuated with both
pretreatment and treatment. There was no improvement in learning and memory with chia treatment. Rather, cognitive performance
in chia‑pretreated animals was remarkably worse as compared to their non‑treated disease‑induced counterparts. Hippocampal
concentrations of amyloid‑β42, amyloid precursor protein, and total tau protein were similarly increased in all disease‑induced
animals (despite chia supplementation), as compared to the controls. Based on these ndings, chia supplementation during the
progression of Alzheimer’s disease may exacerbate the disease. Although the results presented here emerge from an experimental/
preclinical study, we suggest cautious and careful use of chia, especially in early‑stage Alzheimer’s patients, until future research in
dierent experimental settings is conducted.
Key words: Alzheimer’s disease, chia, anxiety, depression, learning, memory
To date, all of the neurodegenerative diseases, in‑
cluding Alzheimer’s disease, remain without a cu‑
rative treatment. Nevertheless, ceaseless efforts of
neuroscientists are paving the way with remarkable
palliative treatment options towards discovering a cu‑
rative treatment. Alzheimer’s disease is the most com‑
mon form of dementia (Alzheimer’s Association 2018).
Since the first identification of Alzheimer’s disease,
our knowledge about its pathophysiology has substan‑
tially grown and carried us to the point where we can
identify and test potential medical and nutritional
interventions to break the neurodegeneration cycle.
Among the possible interventions, polyphenols such
as rosmarinic acid, quercetin, rutin, and caffeic acid
are extensively studied and shown to possess promis‑
ing antidegenerative actions (Bhullar and Rupasinghe
2013, Kelsey et al. 2010, Nabavi et al. 2015). In addi‑
tion to polyphenols, several reports underline disease
relieving features of an omega‑3‑rich diet in Alzhei‑
mer’s (Barberger‑Gateau et al. 2007, Huang et al. 2005,
Kalmijn et al. 1997). Although clinical data regarding
benefits of omega‑3 fatty acids in Alzheimer’s disease
are controversial to some extent (see Canhada et al.
2017), the systematical review by Hooijmans et al.
Received 8 August 2018, accepted 29 October 2018
Acta Neurobiol Exp 2018, 78: 322–331
Chia seeds impair learning/memory in experimental Alzheimer’s disease 323Acta Neurobiol Exp 2018, 78: 322–331
(2012) suggests that omega‑3 supplementation can al‑
leviate Alzheimer’s pathology in experimental models.
Hooijmans et al. (2012) also noted that contradictory
clinical reports likely result from inadequate duration
of the supplementation.
The seeds of chia (Salvia hispanica L.), a native Lat‑
in American herbaceous plant, are highly rich in the
abovementioned polyphenols (Pellegrini et al. 2018) as
well as omega‑3 fatty acids (especially alpha‑linolenic
acid) (Sargi et al. 2013). These ingredients have caused
chia seeds to be called a “superfood” and a “functional
food” (Muñoz et al. 2013, van den Driessche et al. 2018).
Supplementation with chia seeds is generally considered
to be safe, with relatively rare and nonspecific adverse
effects (Ulbricht et al. 2009). In recent years, there has
been an increase in research regarding health benefits of
chia, including antioxidant, antiobesogenic, antidiabet‑
ic, cardioprotective, and antitumoral effects. Surprising‑
ly, however, there are no studies examining the effects of
chia seeds on neurodegenerative diseases and in partic‑
ular, Alzheimer’s disease.
Based on the ingredients of chia seeds, the present
study was designed to test the hypothesis that chia seed
supplementation results in antidegenerative effects in
an experimental model of Alzheimer’s disease. To test
this hypothesis we investigated cognitive, behavioral
and Alzheimer’s‑associated parameters in an aluminum
chloride‑induced disease model.
Laboratory conditions and preparations
The study was conducted in the Application and
Research Center for Experimental Researches at Hatay
Mustafa Kemal University under standardized labora‑
tory conditions (22±2°C temperature, 55±10% relative
humidity, 12:12‑h light/dark cycle). The animals were
provided with ad libitum water and standard or chia‑rich
pellets. Chia seeds were purchased from local suppli‑
ers, and finely ground and nicely blended in crushed
standard rat chow [36.2% (w/w)]. Aluminum chlo‑
ride (AlCl3) (Merck, Germany) and D‑galactose (D‑gal)
(Sigma‑Aldrich, Germany) were dissolved in physiolog‑
ical saline in separate beakers (0.07 M and 0.8 M, respec‑
tively). Chia‑rich pellets were freshly prepared every
other day, whereas solutions were prepared weekly and
stored in the refrigerator (4°C). The study was approved
by the local ethics committee at Hatay Mustafa Kemal
University (#2017/4‑1 rev.#2018/6‑3).
Adult male Wistar albino rats were randomly as‑
signed into one of four groups: (1) Control (Con, n=8),
(2) Alzheimer (Alz, n=10), (3) Pretreatment (Pre, n=10),
and (4) Treatment (Tre, n=10). The animals in Alz, Pre,
and Tre groups intraperitoneally received 10 mg/kg/day
AlCl3 and 150 mg/kg/day D‑gal for 21 days whereas the
Con group received saline injections. The combination
of AlCl3 and D‑gal was preferred based on previous re‑
ports (Xiao et al. 2011, Chiroma et al. 2018), which note
that this combination generates pathologies resembling
to Alzheimer’s disease such as memory deficits, neu‑
ronal loss, increased acetylcholinesterase activity, and
tauopathy in Wistar albino rats. Con and Alz groups were
fed standard rat chow throughout the study. The animals
in the Pre group were fed chia‑rich pellets during the
induction of the experimental model, whereas the Tre
group were fed chia‑rich pellets for 21 days following
the induction. All animals were subjected to a battery of
behavioral and cognitive tests to evaluate anxiety‑ and
depression‑like behaviors, and learning and memory.
The study plan is shown at Fig. 1.
Fig.1. Study plan. Con: Control, Alz: Alzheimer, Pre: Pretreatment, Tre: Treatment.
324 Y. Bilgic et al. Acta Neurobiol Exp 2018, 78: 322–331
Behavioral and cognitive tests
Open‑field and elevated plus maze tests were per‑
formed to assess anxiety‑like behaviors. Depres‑
sion‑like behaviors were evaluated by using Porsolt’s
forced swim test. Learning and memory performance
of the animals was examined by means of Morris’ wa‑
ter maze test. The methods for each test are briefly ex‑
Open‑Field (OF) test
The test apparatus consists of an open‑top
cube‑shaped box (70x70 cm) which is virtually divided
to central (40x40 cm) and peripheral zones (30 cm to the
wall). The animals were gently released into the appa‑
ratus and left to move freely for 5 min. Total distance
moved (cm), velocity (cm/s) and time in center zone (s)
were estimated with an animal tracking software (Etho‑
Vision XT, Noldus, The Netherlands) whereas numbers
of defecations and rearing behaviors were manually re‑
corded. The apparatus was cleaned with 70% ethanol be‑
tween trials to eliminate olfactory cues.
Elevated plus maze (EPM) test
The elevated plus maze test was conducted in
a plus‑shaped apparatus with two open and two closed
arms (50x10 cm arms and 10x10 cm center). The closed
arms were surrounded by opaque walls with the height
of 50 cm. The animals were gently placed into the appa‑
ratus and left to explore freely for 5 min. Time in open/
closed arms (s) was measured with an animal tracking
software (EthoVision XT, Noldus, The Netherlands),
and numbers of rearing and head‑dipping behaviors
were manually noted. The apparatus was cleaned with
70% ethanol between trials to eliminate olfactory cues.
Porsolt’s forced swim test
The forced swim test was performed after anxiety
tests in a cylindrical pool (25 cm dia.) that was filled
with warm water (25±1°C) to the height of 35 cm. In
the acclimation (pre‑test) session, the animals were re‑
leased into the pool and left to habituate for 15 min.
The test session, in which the ethological analysis was
carried out from video recordings by using a software
(Behavioral Observation Research Interactive Software,
Italy) (Friard and Gamba 2016), was conducted 24‑h
later. In the test session, the animals were allowed to
swim for 5 min and behaviors except for those required
to keep the nose above the water surface (i.e., swim‑
ming, climbing, diving) were considered mobility. The
water was changed and pool was cleaned thoroughly
Morris’ water maze test
A round pool with the diameter of 150 cm was filled
with warm water (maintained at 25±1°C with water
heaters) to the height of 50 cm and the water was dark‑
ened with food coloring. A platform was placed 2 cm
below the water level on one of the virtually designat‑
ed quadrants. The animals were taught the location of
the platform for 4 consecutive days in learning ses‑
sions. Each learning session was consisted of 4 trials
with 90 s cut‑off and 30 s inter‑trial interval. On the
day after the last learning session, the platform was re‑
moved and the animals were left to swim freely for 90 s
(probe trial). All trials were video‑recorded and latency
to the platform (s), total distance moved (cm), velocity
(cm/s), time in the target quadrant (s), distance moved
in the target quadrant (cm), and average distance to
the platform zone (cm) were estimated by using a soft‑
ware (EthoVision XT, Noldus, The Netherlands).
The animals were exsanguinated under ketamine/
xylazine (80/12 mg/kg) anesthesia following the be‑
havioral/cognitive tests. The brains were excised and
hippocampi were dissected on ice. The hippocampal
tissues were homogenized in a proprietary lysis re‑
agent (T‑PER, Thermo Fisher Scientific, USA) and tissue
levels of amyloid‑β42, amyloid precursor protein (APP),
and total tau protein (t‑tau) were measured with com‑
mercial ELISA kits (Elabscience, China) according to the
instructions of manufacturer. Amyloid‑β42, which ac‑
cumulates in Alzheimer’s disease, is the cleavage prod‑
uct of APP and tau protein is a microtubule‑associated
protein which is strongly connected with the progres‑
sion of the disease (Kamentani et al. 2018). Bradford’s
(1976) method was used to quantify total protein con‑
tents. The results were reported as ng/mg protein.
Parametric data were analyzed with ordinary or re‑
peated measures one‑way ANOVAs, post‑hoc Tukey’s
test, or paired Student’s t‑tests, as appropriate. Krus‑
kal‑Wallis test, post‑hoc Dunn’s test, or Mann‑Whitney
U tests were used for non‑parametric data, as appropri‑
Chia seeds impair learning/memory in experimental Alzheimer’s disease 325Acta Neurobiol Exp 2018, 78: 322–331
ate. Results are shown as mean ± standard error of the
mean (SEM) for parametric data, and 25–75% percen‑
tiles for non‑parametric data. Results were considered
statistically significant at a p<0.05 threshold.
Body weights and chow consumption
Initial body weights of the animals were similar
among groups (p>0.05). There was no difference be‑
tween final body weights (data not shown), although
Tre animals had significantly higher body weight
change than Alz and Pre groups (F(3,34)=7.21, p=0.001;
post‑hoc p<0.001 and p=0.035, respectively) (Fig. 2A).
No significant difference in bod weight was found be‑
tween groups consuming standard vs, chia‑rich chows
(p>0.05), as depicted in Fig. 2B.
The total distance moved and velocity were ex‑
amined for locomotor activity. Both parameters did
not differ between groups in either open‑field test
(Fig. 3A, B) or in the probe trial of Morris’ water maze
test (data not shown).
We employed the elevated plus maze test to de‑
termine anxiety‑like behaviors. More time spent in
the open‑arms is thought to reflect lower anxiety‑like
Fig.2. (A) Initial body weights & Δ–Weight (B) Chow consumption. (A) Initial
body weights are shown with solid bars whereas cumulative body weight
changes (Δ–Weight) are shown with patterned bars. Asterisks (*) indicate
signicance versus Treatment (p<0.05). (B) There was no statistical sig‑
nicance for the consumption of standard and chia‑rich pellets (p>0.05)
(Control: Con, Alzheimer: Alz, Pretreatment: Pre, Treatment: Tre).
Fig.3. Locomotor activity in the open‑eld test. (A) Total distance moved
and (B) Velocity. There was no statistical signicance for either (A) total
distance moved or (B) velocity (p>0.05) (Control: Con, Alzheimer: Alz,
Pretreatment: Pre, Treatment: Tre).
326 Y. Bilgic et al. Acta Neurobiol Exp 2018, 78: 322–331
behavior. Compared to the controls, Alz, Pre and Tre
animals spent significantly less time in the open‑arms
(F(3,34)=6.9, p=0.001; post‑hoc p=0.037, p<0.001 and
p=0.006, respectively) and there was no difference be‑
tween these groups (Fig. 4A). Except for rearing, which
represents exploratory behavior, we did not find statis‑
tically significant differences between groups for other
ethological measures, including defecation, stretch‑
ing and head‑dipping (data not shown). The number
of rearing behaviors was significantly lower in the Pre
group relative to controls (Fig. 4B).
As shown at Fig. 5A, in the Porsolt’s forced swim
test, which was performed to assess depression‑like
behaviors, Alz animals had significantly lower mobility
compared to other groups (F(3,34)=5.6, p=0.003; post‑hoc
p=0.004 vs Con, p=0.032 vs Pre, and p=0.020 vs. Tre). No
difference in mobility was found between Con, Pre and
Tre groups (p>0.05). The proportion of climbing time
(to total mobility) did not between groups (data not
shown). Since diving had a rare occurrence, it could
not be analyzed. There were no group differences in
the number of head twitches (p>0.05) (Fig. 5B).
Learning and memory
The spatial learning and memory performance of the
animals was evaluated using Morris’ water maze test. As
depicted at Fig. 6, all groups took gradually less time to
locate the escape platform in learning trials (RM‑ANO‑
VA, p<0.001), but in the last learning session, controls
showed significantly faster latency to the platform
than other animals (KW‑test’s H=17.5, p<0.001; post‑hoc
Fig. 4. (A) In the elevated plus maze test, the animals in which the ex‑
perimental Alzheimer’s disease was induced displayed anxiety‑like be‑
haviors, but neither pretreatment nor treatment with chia seeds had
any effect on these behaviors. (B) Exploratory behavior, inferred from
the number of rearing, was decreased in chia‑pretreated animals. As‑
terisks (*) indicate significance versus Control, and section sign (§) ver‑
sus Treatment (p<0.05). (Control: Con, Alzheimer: Alz, Pretreatment:
Pre, Treatment: Tre).
Fig.5. (A) Mobility and (B) Number of head twitches in the Porsolt’s
forced swim test. (A) Non‑treated disease‑induced animals exhibited
significantly less mobility than both controls and chia‑supplemented
groups. Asterisk (*) indicates significance versus Control, and section
sign (§) versus Alzheimer (p<0.05). (B) There was no statistical signifi‑
cance for the number of head twitches between groups (p>0.05) (Con‑
trol: Con, Alzheimer: Alz, Pretreatment: Pre, Treatment: Tre).
Chia seeds impair learning/memory in experimental Alzheimer’s disease 327Acta Neurobiol Exp 2018, 78: 322–331
p=0.037 vs. Alz, p<0.001 vs. Pre, and p=0.003 vs. Tre), In
addition, the Pre group took longer to find the platform
compared to Alz group (p=0.036). In the probe trial, the
time spent in the target quadrant, which is interpreted
as the main measure of memory retention, was signifi‑
cantly greater in controls than other animals. Further‑
Fig.6. Learning trials of the Morris’ water maze test. All animals gradually showed better performance at finding the platform; however, con‑
trols spent significantly lesser time than others in the last learning trial. Different letters in each group indicate statistical significance between
learningsessions (in‑group comparison) (p<0.05). Asterisks (*) indicate significance between groups shown with arrow begin and end in each day
(inter‑group comparison) (p<0.05) (Control: Con, Alzheimer: Alz, Pretreatment: Pre, Treatment: Tre).
Fig.7. (A) The time spent in the target quadrant, (B) Ratio of in‑target distance to total distance moved and (C) Average distance to the platform
zone. (A) Disease‑induced animals, either chia supplemented or not, spent lesser time than controls in the target quadrant. Also, the animals
pretreated with chia elapsed less time in the target quadrant compared to non‑treated disease‑induced animals. Both for (B) the proportional
in‑target distance and (C) distance to the platform zone, the results were alike. Asterisks (*) indicate significance versus Control, and section sign
(§) versus Alzheimer (p<0.05) (Control: Con, Alzheimer: Alz, Pretreatment: Pre, Treatment: Tre).
328 Y. Bilgic et al. Acta Neurobiol Exp 2018, 78: 322–331
more, the Pre group spent less time than Alz and Tre
groups in the target quadrant, as illustrated at Fig. 7A.
Results for additional parameters were consistent with
these results, specifically the ratio of distance moved
in the target quadrant to total distance moved (Fig. 7B;
F(3,34)=13.6, p<0.001; post‑hoc Con vs. Alz: p=0.044, Con vs.
Pre: p<0.001, Con vs. Tre: p<0.001, Alz vs. Pre: p=0.006)
and the average distance to the platform zone (Fig. 7C;
F(3,34)=11, p<0.001; post‑hoc Con vs. Alz: p=0.049, Con vs.
Pre: p<0.001, Con vs. Tre: p=0.009, Alz vs. Pre: p=0.016).
As shown at Fig. 8A, the hippocampal concentrations
of amyloid‑β42 were higher in Alz, Pre and Tre animals
as compared to controls (KW‑test’s H=14.6, p=0.002;
post‑hoc p<0.001, p=0.017 and p=0.002, respectively).
As expected, the concentration of APP was similarly
increased in the experimental Alzheimer’s disease‑in‑
duced groups (KW‑test’s H=8.4, p=0.038; post‑hoc Con vs.
Alz: p=0.012, Con vs. Pre: p=0.031 and Con vs. Tre: p=0.011)
(Fig. 8B). Also, higher amounts of t‑tau were found in Alz,
Pre and Tre groups as compared to Con group (KW‑test’s
H=15.2, p=0.002; post‑hoc p=0.001, p<0.001 and p=0.007,
respectively) (Fig. 8C). There was no difference between
Alz, Pre and Tre animals in regard to aforementioned
hippocampal measurements (p>0.05).
The present study investigated the behavioral and
cognitive effects of chia‑rich feeding in rats, using an
aluminum chloride‑induced experimental Alzheimer’s
disease model. Our main results can be summarized as
follows: the disease (i) generated anxiety‑like behaviors,
but neither pretreatment (i.e., supplementation while
the model is being established) nor treatment (i.e., sup‑
plementation after the model is established) with chia
had any anxiolytic effects, (ii) provoked depression‑like
behaviors, and both pretreatment and treatment alle‑
viated depressive behaviors, (iii) impaired the learning
and memory performance of animals, and pretreat‑
ment (but not treatment) exacerbated the impairment
in learning and memory, and (iv) increased the hippo‑
campal concentrations of Alzheimer’s‑associated pa‑
rameters [amyloid‑β42, APP and t‑tau] which remained
elevated with pretreatment or treatment.
Modern humans are inevitably exposed to alumi‑
num through polluted air, contaminated diet, medi‑
cations, and even the skin (Exley 2013). Although this
environmental exposure is believed to be in minute
amounts (Campbell 2002), aluminum has been shown to
accumulate in aging neurons due to lifetime exposure
(Walton 2006). Previous studies also show elevated lev‑
els of aluminum in the brains and cerebrospinal fluid
of Alzheimer’s disease patients (Virk and Eslick 2015).
Several authors have emphasized the role of chron‑
ic aluminum exposure in the pathophysiology of Alz‑
Fig. 8. Hippocampal concentrations of (A) Amyloid‑β42, (B) Amyloid
precursor protein and (C) Total tau protein. Controls had significantly
lower concentrations of amyloid‑β42, APP, and t‑tau. There was no
difference between disease‑induced animals. Asterisks (*) indicate
significance versus Treatment (p<0.05) (Control: Con, Alzheimer: Alz,
Pretreatment: Pre, Treatment: Tre).
Chia seeds impair learning/memory in experimental Alzheimer’s disease 329Acta Neurobiol Exp 2018, 78: 322–331
heimer’s disease (Gupta et al. 2005, Wang et al. 2016);
however, no direct evidence exists that supports or
rejects a causal relationship between these variables.
Nevertheless, the existing experimental data indicate
that aluminum‑based experimental models show neu‑
ronal alterations that resemble those observed in Alz‑
heimer’s disease (Castorina et al. 2010, Shaw and Toml‑
jenovic 2013, Walton 2007, 2014).
A significant proportion of Alzheimer’s patients suf‑
fer from comorbid affective disorders, even though the
disease is fundamentally a memory deteriorating dis‑
ease (Even and Weintraub 2010, Novais and Starkstein
2015). Affective disorders and Alzheimer’s disease seem
to be reciprocally linked based on abnormal myelin‑
ation (Nihonmatsu‑Kikuchi et al. 2013). Furthermore,
they share some other similar neuroimmunologic, neu‑
roendocrine and oxidative disturbances (Rodrigues et
al. 2014). Despite these similarities, depressive symp‑
toms are not correlated with the severity of Alzhei‑
mer’s disease, which suggests distinct pathophysio‑
logical mechanisms (Lee and Lyketsos 2003). Hence, an
intervention that relieves depression is not expected to
have antidegenerative efficacy in Alzheimer’s disease.
Despite an extensive literature search, we are aware
of only a few reports examining the behavioral and
cognitive effects of chia seeds. In the behavioral study
by Nemeth et al. (2014), which was conducted in guin‑
ea pigs, no influence of chia seeds on locomotion was
reported. These results are in accordance with our own
data in rats. Although depression‑like behaviors were
not evaluated in the study by Nemeth et al. (2014), sa‑
liva cortisol concentrations were found to be lower in
chia‑supplemented animals. Cortisol is physiologically
a stress confronting hormone, but in depression, dys‑
regulated hypothalamic‑pituitary‑adrenal (HPA) axis
response leads to increased cortisol levels, which can
be reduced by antidepressants (Maric and Adzic 2013).
Although cortisol is not a depression marker by itself,
but rather an associated hormone, the observed anti‑
depressant‑like effects of chia in the present study may
be derived from its action on the HPA axis. Also, the
finding of ineffectiveness of chia supplementation on
social stress in the study of Nemeth’s team may be in‑
terpreted as absence of any influence on anxiety‑like
behaviors, which is also similar to our results.
Regarding the cognitive effects of chia seeds, a re‑
cent nutritional intervention trial was conducted in
undergraduate students by Onneken (2018). The au‑
thor founds an improvement in both memory and in‑
telligence tests following chia seed supplementation
(5 g/day) for 21 days. Based on these results, the au‑
thor of the study concluded that “chia consumption is
highly recommendable for dealing with Alzheimer’s
disease”, despite the fact that the included partici‑
pants were all young, healthy individuals (average age:
21.3 years). Our results instead suggest that chia seeds
while an experimental Alzheimer’s disease is progress‑
ing tremendously impairs learning and memory. We
also found no benefit of chia seed supplementation af‑
ter the disease has already emerged. Recently, Rui et al.
(2018) reported similar findings in a senescence‑accel‑
erated mouse‑prone 8 mouse line, which displays the
phenotype of accelerated aging. In their study, they
emphasized the absence of cognitive improvement
with chia supplementation and increased activity in
both amyloidogenic and non‑amyloidogenic pathways,
which subsequently bolstered the amyloid pathology.
The ineffectiveness of chia in cognitive improvement
was also confirmed in male guinea pigs by Nemeth et
al. (2015). Undoubtedly, it is clear that experimental/
preclinical researches cannot be directly extrapolat‑
ed to clinical practice; however, these studies are in‑
valuable in terms of translational medicine, especially
where scant knowledge exists.
To further explain our cognitive findings, we should
revisit the chemical properties of aluminum. Alumi‑
num is a prooxidant, but redox‑inactive metal. Exley
(2004) and Ruipérez et al. (2012) have attributed the
prooxidant potency of aluminum, in part, to its iron
reducing action. Aluminum stimulates iron overload in
tissues by altering the cellular uptake of iron (Cannata
Andia 1996). Polyphenols, which are generally known
antioxidants, exert prooxidant activity in this reduced
iron‑enriched medium (Decker 1997, Margină et al.
2015, Osborn and Akoh 2003). Given that chia seeds
are highly rich in omega‑3 fatty acids and polyphe‑
nols, this study was designed to observe chia’s effects
on both progressing and already emerged disease. We
found that the learning and memory impairing effect
of chia is apparent in animals in which the disease
was progressing, but not in those in which the disease
has already emerged. We hypothesize that this result
may be due to the prooxidation favoring environment.
Presumably, the animals in which the disease already
emerged had time to excrete aluminum and were able
to escape from its devastating consequences. Further‑
more, these animals may have had a positive energy
balance via the help of increased beta‑oxidation of
alpha‑linoleic acid (Fu and Sinclair 2000). In this con‑
text, how can our results of Alzheimer’s‑associated bio‑
chemical parameters be interpreted? Indeed, amyloid/
tau hypothesis is insufficient to completely explain the
pathophysiology, although amyloidogenic accumula‑
tion remains to be a prominent feature of the disease
(Kametani and Hasegawa 2018). Hence, one should also
consider other hypothesis, including oxidative stress.
Therefore, no change in these parameters in our study
supports the assumption of prooxidant action of the
330 Y. Bilgic et al. Acta Neurobiol Exp 2018, 78: 322–331
chia supplementation within the perspective of the ox‑
Although this is the first study to examine the be‑
havioral and cognitive effects of chia seeds in an exper‑
imental Alzheimer’s disease model, it has some limita‑
tions that should be taken into account when interpret‑
ing results. Although we demonstrated an impairment
in learning and memory with chia supplementation,
we speculate without any relevant analysis that this
impairment may be the result of oxidative stress. We
also restricted our analyses to basic Alzheimer’s‑asso‑
ciated biochemical parameters. Further examination,
probably including of the cholinergic pathway, protein
modifications, neuroimmune reactions and neuromod‑
ulations, are needed to clarify the exact mechanism(s).
Next, we employed an aluminum‑induced model, which
is reported to resemble the pathophysiological features
of Alzheimer’s disease; however, our results need to be
confirmed in other non‑transgenic and transgenic mod‑
els before deprecating chia supplementation in patients
with Alzheimer’s disease. Nonetheless, based on the
knowledge of the neuronal accumulation of aluminum
in Alzheimer’s disease, we suggest cautious and careful
use of chia supplementation – particularly in early‑stage
patients whose disease is progressing – until further re‑
search in different experimental settings is conducted.
This paper has been generated from the parts of
a master’s thesis prepared by Y. Bilgic and supervised
by E. A. Demir, and submitted to the Institute of Health
Sciences, Hatay Mustafa Kemal University. We would like
to express our sincere thanks to Okan Gonder and Faruk
Arslan for their help at laboratory analyses.
Alzheimer’s Association (2018) 2018 Alzheimer’s disease facts and gures.
Alzheimer’s Dement 14: 367–429.
Barberger‑Gateau P, Raaitin C, LetenneurL, Berr C, Tzourio C, DartiguesJF,
Alpérovitch A (2007) Dietary patterns and risk of dementia: the Three‑City
cohort study. Neurology 69: 1921–1930.
Bhullar KS, Rupasinghe HPV (2013) Polyphenols: multipotent therapeu‑
tic agents in neurodegenerative diseases. Oxid Med Cell Longev 2013:
Bradford MM (1976) A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein‑dye
binding. Anal Biochem 72: 248–254.
Campbell A (2002) The potential role of aluminium in Alzheimer’s disease.
Nephrol Dial Transplant 17: 17–20.
Canhada S, Castro K, Perry IS, Luft VC (2017) Omega‑3 fatty acids’ supple‑
mentation in Alzheimer’s disease: A systematic review. Nutr Neurosci
Cannata Andia JB (1996) Aluminium toxicity: its relationship with bone and
iron metabolism. Nephrol Dial Transplant 11: 69–73.
Castorina A, Tiralongo A, Giunta S, Carnazza ML, Scapagnini G, D’Agata V
(2010) Early eects of aluminum chloride on beta‑secretase mRNA ex‑
pression in a neuronal model of beta‑amyloid toxicity. Cell Biol Toxicol
Chiroma SM, Mohd Moklas MA, Mat Taib CN, Baharuldin MTH, Amon Z
(2018) d‑galactose and aluminium chloride induced rat model with cog‑
nitive impairments. Biomed Pharmacother 103: 1602–1608.
Decker EA (1997) Phenolics: prooxidants or antioxidants? Nutr Rev 55:
Even C, Weintraub D (2010) Case for and against specicity of depression
in Alzheimer’s disease. Psychiatry Clin Neurosci 64: 358–366.
Exley C (2004) The pro‑oxidant activity of aluminum. Free Radic Biol Med
Exley C (2013) Human exposure to aluminium. Environ Sci Process Impacts
Friard O, GambaM (2016) BORIS: afree, versatile open‑source event‑log‑
ging software for video/audio coding and live observations. Methods
Ecol Evol 7: 1325–1330.
Fu Z, Sinclair AJ (2000) Increased α‑linolenic acid intake increases tissue
α‑linolenic acid content and apparent oxidation with little eect on tis‑
sue docosahexaenoic acid in the guinea pig. Lipids 35: 395–400.
Gupta VB, Anitha S, Hegde ML, ZeccaL, Garruto RM, Ravid R, ShankarSK,
Stein R, Shanmugavelu P, Jagannatha Rao KS (2005) Aluminium in
Alzheimer’s disease: are we still at a crossroad? Cell Mol Life Sci 62:
Hooijmans CR, Pasker‑de Jong PCM, de Vries RBM, Ritskes‑Hoitinga M
(2012) The eects of long‑term omega‑3 fatty acid supplementation on
cognition and Alzheimer’s pathology in animal models of Alzheimer’s
disease: asystematic review and meta‑analysis. J Alzheimers Dis 28:
Huang TL, Zandi PP, Tucker KL, Fitzpatrick AL, Kuller LH, Fried LP, Burke GL,
Carlson MC (2005) Benets of fatty sh on dementia risk are stronger
for those without APOE epsilon4. Neurology 65: 1409–1414.
Kalmijn S, Launer LJ, Ott A, Witteman JC, Hofman A, Breteler MM (1997)
Dietary fat intake and the risk of incident dementia in the Rotterdam
Study. Ann Neurol 42: 776–782.
Kametani F, HasegawaM (2018) Reconsideration of amyloid hypothesis
and tau hypothesis in Alzheimer’s disease. Front Neurosci 12: 25.
Kelsey NA, Wilkins HM, Linseman DA (2010) Nutraceutical antioxidants as
novel neuroprotective agents. Molecules 15: 7792–7814.
Lee HB, Lyketsos CG (2003) Depression in Alzheimer’s disease: heteroge‑
neity and related issues. Biol Psychiatry 54: 353–362.
Margină D, IlieM, Grădinaru D, Androutsopoulos VP, Kouretas D, TsatsakisAM
(2015) Natural products‑friends or foes? Toxicol Lett 236: 154–167.
Maric NP, AdzicM (2013) Pharmacological modulation of HPA axis in de‑
pression – new avenues for potential therapeutic benets. Psychiatr
Danub 25: 299–305.
Muñoz LA, Cobos A, Diaz O, Aguilera JM (2013) Chia seed (Salvia hispanica):
an ancient grain and anew functional food. Food Rev Int 29: 394–408.
Nabavi SF, Tenore GC, Daglia M, Tundis R, Loizzo MR, Nabavi SM (2015)
The cellular protective eects of rosmarinic acid: from bench to bed‑
side. Curr Neurovasc Res 12: 98–105.
NemethM, Millesi E, Wagner K‑H, Wallner B (2014) Eects of diets high in
unsaturated fatty acids on socially induced stress responses in Guinea
pigs. PloS One 9: e116292.
NemethM, Millesi E, Wagner K‑H, Wallner B (2015) Sex‑specic eects of
diets high in unsaturated fatty acids on spatial learning and memory in
Guinea Pigs. PloS One 10: e0140485.
Nihonmatsu‑Kikuchi N, Hayashi Y, Yu XJ, Tatebayashi Y (2013) Depression
and Alzheimer’s disease: novel postmortem brain studies reveal apos‑
sible common mechanism. J Alzheimers Dis 37: 611–621.
Novais F, Starkstein S (2015) Phenomenology of depression in Alzheimer’s
disease. J Alzheimers Dis 47: 845–855.
Chia seeds impair learning/memory in experimental Alzheimer’s disease 331Acta Neurobiol Exp 2018, 78: 322–331
Onneken P (2018) Salvia hispanica L (Chia Seeds) as brain superfood: how
seeds increase intelligence. J Nutr Food Sci 8: 684.
Osborn HT, Akoh CC (2003) Eects of natural antioxidants on iron‑cata‑
lyzed lipid oxidation of structured lipid‑based emulsions. J Am Oil Chem
Soc 80: 847–852.
Pellegrini M, Lucas‑Gonzalez R, Sayas‑Barberá E, Fernández‑López J,
Pérez‑Álvarez JA, Viuda‑Martos M (2018) Bioaccessibility of phenolic
compounds and antioxidant capacity of chia (Salvia hispanicaL.) seeds.
Plant Foods Hum Nutr 73: 47–53.
Rodrigues R, Petersen RB, Perry G (2014) Parallels between major depres‑
sive disorder and Alzheimer’s disease: role of oxidative stress and ge‑
netic vulnerability. Cell Mol Neurobiol 34: 925–949.
Rui Y, LvM, Chang J, Xu J, QinL, Wan Z (2018) Chia seed does not improve
cognitive impairment in SAMP8 mice fed with high fat diet. Nutrients
Ruipérez F, Mujika JI, Ugalde JM, Exley C, Lopez X (2012) Pro‑oxidant activity
of aluminum: promoting the Fenton reaction by reducing Fe(III) to Fe(II).
J Inorg Biochem 117: 118–123.
Sargi SC, Silva BC, Santos HMC, Montanher PF, Boeing JS, Santos JúniorOO,
Souza NE, Visentainer JV (2013) Antioxidant capacity and chemical com‑
position in seeds rich in omega‑3: chia, ax, and perilla. Food Sci Technol
Shaw CA, Tomljenovic L (2013) Aluminum in the central nervous system
(CNS): toxicity in humans and animals, vaccine adjuvants, and autoim‑
munity. Immunol Res 56: 304–316.
Ulbricht C, ChaoW, Nummy K, Rusie E, Tanguay‑Colucci S, Iannuzzi C,
Plammoottil J, VargheseM, WeissnerW (2009) Chia (Salvia hispanica):
Asystematic review by the natural standard research collaboration. Rev
Recent Clin Trials 4: 168–174.
van den Driessche JJ, Plat J, Mensink RP (2018) Eects of superfoods on risk
factors of metabolic syndrome: asystematic review of human interven‑
tion trials. Food Funct 9: 1944–1966.
Virk SA, Eslick GD (2015) Aluminum levels in brain, serum, and cerebrospi‑
nal uid are higher in Alzheimer’s disease cases than in controls: ase‑
ries of meta‑analyses. J Alzheimers Dis 47: 629–638.
Walton JR (2006) Aluminum in hippocampal neurons from humans with
Alzheimer’s disease. Neurotoxicology 27: 385–394.
Walton JR (2007) An aluminum‑based rat model for Alzheimer’s disease
exhibits oxidative damage, inhibition of PP2A activity, hyperphosphor‑
ylated tau, and granulovacuolar degeneration. J Inorg Biochem 101:
Walton JR (2014) Chronic aluminum intake causes Alzheimer’s disease: ap‑
plying Sir Austin Bradford Hill’s causality criteria. J Alzheimers Dis 40:
Wang Z, Wei X, Yang J, Suo J, Chen J, Liu X, Zhao X (2016) Chronic exposure
to aluminum and risk of Alzheimer’s disease: A meta‑analysis. Neurosci
Lett 610: 200–206.
Xiao F, Li XG, Zhang XY, Hou JD, Lin LF, Gao Q, Luo HM (2011) Combined
administration of D‑galactose and aluminium induces Alzheimer‑like le‑
sions in brain. Neurosci Bull 27: 143–155.