Long-Term Cannabidiol Treatment Prevents the Development of Social Recognition Memory Deficits in Alzheimer's Disease Transgenic Mice

Abstract

Impairments in cognitive ability and widespread pathophysiological changes caused by neurotoxicity, neuroinflammation, oxidative damage, and altered cholesterol homeostasis are associated with Alzheimer's disease (AD). Cannabidiol (CBD) has been shown to reverse cognitive deficits of AD transgenic mice and to exert neuroprotective, anti-oxidative, and anti-inflammatory properties in vitro and in vivo. Here we evaluate the preventative properties of long-term CBD treatment in male AβPPSwe/PS1ΔE9 (AβPP × PS1) mice, a transgenic model of AD. Control and AD transgenic mice were treated orally from 2.5 months of age with CBD (20 mg/kg) daily for 8 months. Mice were then assessed in the social preference test, elevated plus maze, and fear conditioning paradigms, before cortical and hippocampal tissues were analyzed for amyloid load, oxidative damage, cholesterol, phytosterols, and inflammation. We found that AβPP × PS1 mice developed a social recognition deficit, which was prevented by CBD treatment. CBD had no impact on anxiety or associative learning. The prevention of the social recognition deficit was not associated with any changes in amyloid load or oxidative damage. However, the study revealed a subtle impact of CBD on neuroinflammation, cholesterol, and dietary phytosterol retention, which deserves further investigation. This study is the first to demonstrate CBD's ability to prevent the development of a social recognition deficit in AD transgenic mice. Our findings provide the first evidence that CBD may have potential as a preventative treatment for AD with a particular relevance for symptoms of social withdrawal and facial recognition.

Full text

AUTHOR COPY
Journal of Alzheimer’s Disease 42 (2014) 1383–1396
DOI 10.3233/JAD-140921
IOS Press
1383
Long-Term Cannabidiol Treatment Prevents
the Development of Social Recognition
Memory Deficits in Alzheimer’s Disease
Transgenic Mice
David Chenga,b, Adena S. Spiroc,d, Andrew M. Jennerc,d, Brett Garnerc,dand Tim Karla,b,e,∗
aNeuroscience Research Australia, Randwick, NSW, Australia
bSchool of Medical Sciences, University of New South Wales, NSW, Australia
cIllawarra Health and Medical Research Institute, University of Wollongong, NSW, Australia
dSchool of Biological Sciences, University of Wollongong, NSW, Australia
eSchizophrenia Research Institute, Darlinghurst, NSW, Australia
Handling Associate Editor: Gary Arendash
Accepted 14 May 2014
Abstract. Impairments in cognitive ability and widespread pathophysiological changes caused by neurotoxicity, neuroinflam-
mation, oxidative damage, and altered cholesterol homeostasis are associated with Alzheimer’s disease (AD). Cannabidiol
(CBD) has been shown to reverse cognitive deficits of AD transgenic mice and to exert neuroprotective, anti-oxidative, and
anti-inflammatory properties in vitro and in vivo. Here we evaluate the preventative properties of long-term CBD treatment in
male A␤PPSwe/PS1E9 (A␤PP ×PS1) mice, a transgenic model of AD. Control and AD transgenic mice were treated orally
from 2.5 months of age with CBD (20 mg/kg) daily for 8 months. Mice were then assessed in the social preference test, elevated
plus maze, and fear conditioning paradigms, before cortical and hippocampal tissues were analyzed for amyloid load, oxidative
damage, cholesterol, phytosterols, and inflammation. We found that A␤PP ×PS1 mice developed a social recognition deficit,
which was prevented by CBD treatment. CBD had no impact on anxiety or associative learning. The prevention of the social
recognition deficit was not associated with any changes in amyloid load or oxidative damage. However, the study revealed a
subtle impact of CBD on neuroinflammation, cholesterol, and dietary phytosterol retention, which deserves further investigation.
This study is the first to demonstrate CBD’s ability to prevent the development of a social recognition deficit in AD transgenic
mice. Our findings provide the first evidence that CBD may have potential as a preventative treatment for AD with a particular
relevance for symptoms of social withdrawal and facial recognition.
Keywords: Alzheimer’s disease, amyloid load, behavior, cannabidiol, cholesterol, neuroinflammation, oxidative stress, phytos-
terol, social recognition memory, transgenic A␤PPSwe/PS1E9 mice
INTRODUCTION
Alzheimer’s disease (AD) is a neurodegenerative
disease, which is associated with progressive mem-
∗Correspondence to: Tim Karl, PhD, Neuroscience Research
Australia, Barker Street, Randwick, NSW 2031, Australia. Tel.: +61
2 9399 1838; Fax: +61 2 9399 1005; E-mail: t.karl@neura.edu.au.
ory loss. Other behavioral and cognitive symptoms
include social withdrawal, poor facial recognition
ability, increased motor agitation, and likelihood of
wandering [1, 2]. AD is characterized by two main
postmortem pathological hallmarks; amyloid-␤(A␤)
protein aggregation forming plaque deposits and tau
protein hyperphosphorylation resulting in neurofib-
rillary tangles. Microglia, the resident immune cells
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved

AUTHOR COPY
1384 D. Cheng et al. / Protective Effects of Cannabidiol
of the central nervous system, are activated for the
phagocytosis of A␤[3–5], but impaired clearance or
reuptake of A␤results in the release of inflamma-
tory cytokines such as interleukin-1␤(IL-1␤), tumor
necrosis factor-␣(TNF-␣), and chemokines, that cause
neuroinflammation. Brain tissue damage is further
exacerbated by the release of glutamate and reactive
oxygen and nitrogen species, resulting in neurotoxic-
ity and oxidative damage, respectively [6]. Increased
oxidative stress may be an early indication of AD risk
[7, 8]. Disturbances in brain cholesterol metabolism
are associated with the major pathological features
of AD (including A␤and tau pathology). In partic-
ular, decreased cholesterol synthesis correlates with
the severity of neurodegeneration and dementia [9,
10], while late-stage AD patients also show decreased
cholesterol circulation [11, 12]. Interestingly, dietary
phytosterols (or plant sterols) found naturally in many
foods (such as vegetable oils, nuts, grains, and grain-
derived products) [13] can either interfere with critical
functional processes in AD or decrease amyloidogenic
processing [14]. Some phytosterols may even be rele-
vant additional biomarkers for AD [15].
Current treatments available to AD patients do
not slow the progression of the disease and only
offer limited benefits for the cognitive abilities of
patients (reviewed in [16]). Thus, it is important to
explore novel alternative treatment strategies. The phy-
tocannbinoid cannabidiol (CBD) may be a potential
new candidate for AD therapy (for review, see [17]).
CBD is derived from the cannabis sativa plant and is
devoid of psychoactive properties. It has neuroprotec-
tive, anti-inflammatory, and anti-oxidative properties
[18–22], thereby countering a number of AD-relevant
pathological symptoms. In vitro studies have found
that CBD prevents A␤-induced tau protein hyperphos-
phorylation [23], neurotoxicity [23, 24], attenuates cell
death, and promotes neurogenesis in mouse hippocam-
pal cells [25, 26]. These biological functions of CBD
promise therapeutic value for the neurodegenerative
and neurotoxic components of AD. Indeed, in vivo
studies reported that CBD reduced A␤-induced neu-
roinflammation in rats and mice [25, 27] and rescued
learning deficits in the Morris water maze in a phar-
macological mouse model of AD [28]. The memory
restoring properties of CBD were linked to a reduc-
tion in microglial activation and pro-inflammatory
cytokines (i.e., decreased IL-6) [28].
Current research suggests existing interventions
may be administered too late in the disease process
when the damage caused by AD pathology is already
too severe [17, 29, 30]. Thus, in the current study, we
evaluated for the very first time the effectiveness of
long-term oral CBD treatment to prevent the devel-
opment of cognitive deficits and AD-relevant brain
pathophysiology in an established transgenic mouse
model of familial AD [31]. The double transgenic
A␤PPSwe/PS1E9 (A␤PP ×PS1) mouse model co-
expresses mutant amyloid-␤protein precursor (A␤PP)
and presenilin 1 (PS1) genes [31–34]. Amyloid plaques
are found as early as at 4 months of age in these AD
transgenic mice [35]. Our past research established
that male A␤PP ×PS1 mice demonstrate social recog-
nition deficits, increased anxiety, and task-specific
hyperlocomotion whereas sensorimotor gating and
spatial memory were intact at 10–12 months of age
[36]. Importantly, we also demonstrated recently that
3-weeks of CBD treatment effectively reversed the
social and object recognition memory deficits of
A␤PP ×PS1 males [37]. In the present study male
A␤PP ×PS1 mice were treated with CBD (20 mg/kg)
or vehicle using a daily voluntary oral administration
scheme for 8 months beginning at 2.5 months of age
when AD-like pathophysiology is still sparse (i.e., no
A␤burden reported for 4 months old A␤PP ×PS1
mice: [35]). Following this, mice were assessed in
social recognition memory, associative memory (i.e.,
fear conditioning), and anxiety, before brain samples
were analyzed for amyloid load, oxidative damage
(i.e., markers of cerebral lipid oxidation), cholesterol
levels as well as dietary phytosterols, and neuroin-
flammation markers. We selected the cytokines TNF-␣
and IL-1␤as both have been most strongly implicated
in the promotion of AD pathology in humans and in
AD transgenic mouse models [38, 39]. Furthermore,
inflammation driven by these cytokines is attenuated
by CBD [27, 28].
METHODS
Animals
Double transgenic mice expressing chimeric
mouse/human A␤PP (Mo/HuA␤PP695swe/Swedish
mutations K595N/M596L) and mutant human
PS1 (PS1/E9) mice were obtained from Jackson
Laboratory [Bar Harbor, USA; strain name: B6C3-
Tg(A␤PPSwe/PS1E9)85Dbo/Mmjax; stock no.
004462] and maintained as hemizygotes on the con-
genic C57BL/6JxC3H/HeJ background as described
previously [31–33, 40]. Male double transgenic mice
(A␤PP ×PS1) and their non-transgenic littermates
(WT) were bred and group-housed in independently
ventilated cages (Airlaw, Smithfield, Australia) at

AUTHOR COPY
D. Cheng et al. / Protective Effects of Cannabidiol 1385
Animal BioResources (Moss Vale, Australia). Test
mice were transported to Neuroscience Research
Australia (NeuRA) at around 10 weeks of age,
where they were group-housed in Polysulfone cages
(1144B: Techniplast, Rydalmere, Australia) with corn
cob bedding (PuraCob Premium: Able Scientific,
Perth, Australia) and some tissues for nesting. Mice
were kept under a 12 : 12 h light:dark schedule
[light phase: white light (illumination: 210lx); lights
on 0700–1900 h]. Environmental temperature was
automatically regulated at 21 ±1◦C and relative
humidity was 40–60%. Food (Gordon’s Rat and
Mouse Maintenance Pellets: Gordon’s Specialty
Stockfeeds, Yanderra, Australia) and water were pro-
vided ad libitum, except where specified. Adult, male
A/J mice from Animal Resources Centre (Canning
Vale, Australia) were placed in the animal enclosures
as standard opponents for the social preference test.
Research and animal care procedures were approved
by the University of New South Wales Animal
Care and Ethics Committee in accordance with the
Australian Code of Practice for the Care and Use of
Animals for Scientific Purposes.
Drug treatment
Powdered cannabidiol (CAS: 13956-29-1, THC
Pharm GmbH, Frankfurt/Main, Germany) was used
at a dose of 20 mg/kg body weight, based on pre-
vious studies evaluating the behavioral properties of
different doses of CBD [41] and the effectiveness of
20 mg/kg CBD to reverse spatial memory deficits of a
pharmacological mouse model of AD [42]. Although
chronic administration of CBD appears to be well tol-
erated by transgenic mice [37, 43], the stress of chronic
injections may impact behavioral, cognitive, and/or
physiological results. Thus, the gel pellet preparation
and the oral administration regime were adapted from
Zhang and colleagues [44, 45]: CBD or vehicle were
dissolved in a highly palatable sweetened and choco-
late flavored gel pellet, and administered at a volume
of 8 ml/kg body weight. Due to the insolubility of CBD
in water, CBD was first dissolved in 100% ethanol and
an equal amount of Tween 80 (Sigma-Aldrich Co., St
Louis, USA), then vortexed vigorously. CBD was dis-
solved in gel pellets with a final composition of 2.0%
ethanol, 2.0% Tween 80, 15.2% Splenda (Splenda Low
Calorie Sweetener: Johnson & Johnson Pacific Pty,
Broadway, Australia), 8.7% gelatine (Davis Gelatine:
GELITA Australia Pty, Josephville, Australia), 20.1%
chocolate flavoring (Queen Flavouring Essence Imi-
tation Chocolate: Queen Fine Foods Pty, Alderley,
Australia), and 52.0% water for irrigation. Vehicle gel
pellets were identical but contained no CBD. Mice
were initially habituated to vehicle gel pellets in their
home cages for seven days. Following this, the mice
were isolated within their home cages for the treat-
ment by placing a plastic divider in the home cage.
Then animals were given either a vehicle or a CBD
gel pellet (treatments were quasi-randomized), which
they consumed within 2–5 min. The plastic divider was
removed once mice had consumed the gel pellets. Mice
were treated daily, late in the afternoon, to avoid poten-
tial acute effects of CBD confounding test outcomes
(Table 1).
Behavioral phenotyping
Starting at 10 months of age, mice were tested in a
number of behavioral tests (Table 1), with an inter-test
interval of at least 48 h as described earlier (n= 8–14
mice per genotype/treatment) [36, 46, 47]. All tests
were conducted during the first 5 h of the light phase
to minimize effects of circadian rhythm.
Social preference test (SPT)
The SPT was used to assess sociability and social
recognition memory [48] and performed as described
Table 1
Age of A␤PPSwe/PS1E9 (A␤PP ×PS1) mice and their WT counterparts (in days ±SEM) at the start of treatment, throughout behavioral
testing and at the end of treatment
Treatment Vehicle CBD
Genotype WT A␤PP ×PS1 WT A␤PP ×PS1
Age at start of treatment 91.5 ±11.5 97.1 ±18.3 89.0 ±8.2 95.9 ±12.5
Number of days treated prior to start of testing 228.1 ±38.5 234.8 ±31.3 226.1 ±33.0 237.8 ±35.5
Social Preference Test 319.6 ±34.5 331.9 ±41.5 315.1 ±30.8 333.8 ±38.6
Elevated plus maze 324.1 ±34.8 336.0 ±41.7 319.3 ±31.3 338.1 ±38.9
Y-Maze 326.9 ±34.5 338.4 ±41.5 321.8 ±31.2 340.6 ±38.6
Fear conditioning 329.9 ±34.5 341.4 ±41.5 324.8 ±31.2 343.6 ±38.6
Tissue collection 333.1 ±34.8 345.0 ±41.7 328.3 ±31.3 347.1 ±38.9
Total Days of treatment 241.6 ±38.9 247.9 ±31.6 239.3 ±33.4 251.2 ±35.8

AUTHOR COPY
1386 D. Cheng et al. / Protective Effects of Cannabidiol
previously [36, 37]. Test animals were isolated for an
hour prior to the start of testing. During the habit-
uation trial, mice were allowed to freely explore a
three-chamber apparatus, consisting of a center cham-
ber (length: 9 cm; width: 18 cm; depth: 20 cm) and
two outer chambers (16 cm ×18 cm ×20 cm), freely
for 5 min. For the sociability test, an unfamiliar stan-
dard opponent (male A/J mouse) was placed in one of
two animal enclosures (i.e., opponent chamber) in a
quasi-randomized fashion (mouse enclosures allowed
nose contact between mice but prevented fighting). The
test mouse was returned to the apparatus and allowed to
explore all three chambers and the animal enclosures
for 10 min. Finally, test animals were observed in a
10 min social recognition test. For this, a second, unfa-
miliar standard opponent was placed in the previously
empty chamber so that the test mouse had the choice
to explore either the familiar mouse (from the previ-
ous trial) or the novel, unfamiliar mouse. The inter-trial
interval (ITI) was 5 min. The chambers and enclosures
were cleaned with 30% ethanol in-between trials and
fresh corn cob bedding was added to the chambers
prior to each test trial. AnyMazeTM (Stoelting, Wood
Dale, USA) tracking software was used to determine
the time spent in the different chambers, number of
entries and distance travelled by the test mice in each
trial. Two mice (1 WT-VEH and 1 WT-CBD) were
excluded from the sociability test due to recording
issues.
Elevated plus maze (EPM)
The EPM assesses the natural conflict between
the tendency of mice to explore a novel environ-
ment and avoidance of a brightly lit, elevated, and
open area [49, 50] and was employed to determine
potential effects of chronic CBD treatment on anxiety
behavior. The ‘+’ apparatus consisted of two alter-
nate open arms (35 cm ×6cm; without side walls) and
two alternate enclosed arms (35 cm ×6 cm; height of
enclosing walls 28 cm) connected by a central platform
(6 cm ×6 cm), elevated 70 cm above the floor. Mice
were placed at the center of the ‘+’ of the grey PVC
plus maze (for further details of apparatus, see [51])
facing an enclosed arm and were allowed to explore
the maze for 5 min. The time spent on open arms,
the percentage of entries onto open arms over total
arm entries (open arm entries) and the distance trav-
elled on the open and enclosed arms were recorded
using AnyMazeTM tracking software. One mouse was
excluded (A␤PP ×PS1-CBD group) for falling off the
apparatus.
Fear conditioning (FC)
FC assesses hippocampus- and amygdala-
dependent associative learning whereby a previously
neutral stimulus elicits a fear response after it has been
paired with an aversive stimulus. On conditioning
day, mice were placed into the test chamber (Model
H10-11R-TC, Coulbourn Instruments, USA) for
2 min. An 80 dB conditioned stimulus (CS) was
presented twice for 30 s with a co-terminating 0.4 mA
2-s foot shock (unconditioned stimulus; US) with an
inter-pairing interval of 2 min. The test concluded
2 min later. The next day (context test), mice were
returned to the apparatus for 7 min. On day 3 (cue
test), animals were placed in an altered context for
9 min. After 2 min (pre-CS/baseline), the CS was
presented continuously for 5 min. The test concluded
after another 2 min, absent the CS (for more details,
see [52, 53]). Time spent freezing was measured on all
three experimental days using Any-MazeTM software.
Biochemical analyses
Mice were anaesthetized and blood was collected
through cardiac puncture. Blood samples were cen-
trifuged (5000 rpm, 5 min, 4◦C) in a microcentrifuge
(Model No. 5415R, Eppendorf, Hamburg, Germany),
and the plasma fraction was collected and stored at
−80◦C. Euthanized mice were perfused with phos-
phate buffered saline (PBS) transcardially as described
previously [54]. Brains were sagittally divided and the
right hemisphere was snap frozen in liquid nitrogen
before being stored at −80◦C. Cortex and hippocampal
samples were dissected and weighed on dry ice prior
to biochemical analyses [Sample numbers for ELISA
and GC-MS were: n= 8 for WT-vehicle, n= 10 for
A␤PP ×PS1-vehicle, n= 10 for WT-CBD, and n=10
for A␤PP ×PS1-CBD].
Aβenzyme-linked immunosorbent assay (ELISA)
for Aβpathology
Frozen cortex (20–30 mg) and hippocampal samples
(∼5 mg) were homogenized and prepared as TBS solu-
ble, and guanidine HCl (gHCl) soluble (TBS insoluble)
fractions and stored at −80◦C as described previously
[54]. Both TBS-soluble and gHCl-soluble fractions
were used in enzyme-linked immunosorbent assay
(ELISA) to investigate the effect of CBD on A␤levels
in transgenic mice. Protein was quantified using the
bicinchoninic (BCA assay) method.
A␤40 and A␤42 protein in TBS-soluble and gHCl-
soluble fractions of brain homogenates were quantified
using Beta Amyloid x-40 and x-42 ELISA kits (Cat

AUTHOR COPY
D. Cheng et al. / Protective Effects of Cannabidiol 1387
No. SIG-38954 and SIG-38956 respectively, Covance,
Emeryville, USA) as described previously [54, 55].
Gas chromatography-mass spectroscopy (GC-MS)
for cholesterol, oxidative damage, and CBD
plasma levels
An Agilent 7000B triple quadrupole mass selective
detector interfaced with an Agilent 7890A GC sys-
tem gas chromatograph, equipped with an automatic
sampler and computer workstation (Agilent Technolo-
gies, Santa Clara, USA) was used to analyze markers
of oxidative damage in the cortical samples and CBD
presence in plasma samples. GC-MS triple quadrupole
provided very high analytical sensitivity for all ana-
lytes measured. Limits of detection (LOD: 0.05 ng/ml)
were significantly less (at least 10-fold) than the levels
of each analyte measured in plasma and brain. 150 ␮l
of plasma were used for the analysis of CBD. The
concentration obtained from the GC-MS was there-
fore multiplied by a factor of 6.67 to give the total
amount of CBD per ml of plasma as shown in the
Results section. The injection port and GC-MS inter-
face were kept at 270◦C and separations were carried
out on a fused silica capillary column (20 m ×0.18 mm
i.d. ×0.18 m film thickness, Restek Rxi-5 ms). Helium
was the carrier gas with a flow rate of 0.8 ml/min (aver-
age velocity = 59 cm/s).
F2-isoprostanes, oxidised sterols (oxysterols), and
cholesterol
Frozen cortex samples (10–20 mg) were homoge-
nized and hydrolyzed overnight for GC-MS analysis as
described previously [56]. Samples were loaded into
solid phase extraction columns (UCT CUQAX223
3 ml; United Chemical Technologies, Bristol, USA).
Sterols and oxysterols, arachidonic acid, DHA, and
F2-isoprostanes were eluted from the SPE column
separately. The sterol/oxysterol fractions were deriva-
tized in 20 ␮l acetonitrile and 20 ␮l Selectra-SIL
BSTFA [N,O-bis(trimethylsilyl)trifluoroacetamide]
containing 1% TMCS (trimethylchlorosilane; United
Chemical Technologies, Bristol, USA) prior to GC-
MS analysis. Quantification of cholesterol oxidation
products (COP) was as previously described [57].
Cholesterol was quantified using lathosterol-d6heavy
isotope standard in a separate (0.6 ␮l split ratio
25 : 1). Relative molar response factors of all analytes
were calculated from calibration curves constructed
from different concentrations in triplicate. The F2-
isoprostane and fatty acid fractions were prepared
and analyzed by GC-MS as described previously
[56]. Quantification of F2-isoprostanes and fatty
acids were calculated by comparison of specific SRM
transitions with their corresponding heavy isotope
internal standards.
Quantification of CBD in plasma
Concentration of cannabidiol in plasma was quan-
tified using the GC-MS as previously described with
slight modifications [58, 59]. Plasma samples (150 ␮l)
were treated using sodium acetate buffer pH 4.0, with
MTBE and Hexane (1 : 1 v/v), rotated for 30 min and
centrifuged at 1500 rpm for 2 min at 4◦C, dried down,
derivatized in 20␮l BSTFA and 20 ␮l 1% TMCS,
and incubated at 70◦C for 30 min. Derivatized sam-
ples were dried down, reconstituted in 40 ␮l of toluene
and analyzed using the GC-MS (1 ␮l splitless). MRM
was performed using EI mode similar to sterol anal-
ysis. Column temperature was held for 1 min and
increased 40◦C/min to 210◦C, then 20◦C/min to 300◦C
and held for 4 min. Quantification of CBD was calcu-
lated by comparison with specific MRM transitions
corresponding with its heavy isotope internal standard
(CBD-d3, Lipomed, Arlesheim, Switzerland).
Inflammatory markers (quantitative polymerase
chain reaction)
RNA extraction: Frozen cortex samples (10–20mg)
were homogenized in Tri-reagent (TRIzol Reagent,
cat no. 15596-018, Life Technologies, Mulgrave,
VIC, Australia) as described previously [57]. RNA
levels were quantified using a NanoDrop 2000 spec-
trophotometer (Thermo Fisher Scientific, Scoresby,
Australia), and diluted in RNAse-free water to obtain
a concentration of 0.5 ␮g/␮l. cDNA was synthesized
using a Tetro cDNA Synthesis kit (Bioline, Alexandria,
Australia), according to manufacturer instructions.
The SensiFAST SYBR No-ROX kit (Bioline, Alexan-
dria, Australia) was used to determine levels of
inflammatory markers. Template concentration was
100 ng (1 : 10 dilution; 1␮l cDNA). Forward and
reverse primers for interleukin-1␤(IL-1␤; forward, 5-
CAACCAACAAGTGATATTCTCCATG-3; reverse,
5-GATCCACACTCTCCAGCTGCA-3), and tumor
necrosis factor-␣(TNF-␣; forward, 5-CATCTTCTC-
AAAATTCGAGTGACAA-3; reverse: 5-TGGGAG-
TAGACAAGGTACAACCC-3) were used as bio-
markers for the quantification of inflammation in trans-
genic mice, with ␤-actin as a housekeeping gene.
Polymerase chain reaction (PCR) assays were reacted
(3-step cycling; IL-1␤: 45 cycles; TNF-␣: 50 cycles)
and analyzed using Roche LightCycler 480 (Roche
Diagnostics, Castle Hill, Australia). Three mice were
excluded (1×WT-CBD, 1×A␤PP ×PS1-vehicle and

AUTHOR COPY
1388 D. Cheng et al. / Protective Effects of Cannabidiol
1×A␤PP ×PS1-CBD) as outliers (2 standard devia-
tions away from mean).
Statistical analyses
One-way ANOVA was used to analyze effect
of ‘treatment’ on A␤levels in A␤PP ×PS1 mice.
Two-way analysis of variance (ANOVA) was used
to analyze behavioral parameters and biochemical
data obtained for oxidative damage, CBD levels and
quantification of inflammation for main effects of
‘genotype’ and ‘treatment’ in all tests. Repeated mea-
sures (RM) ANOVA was used to evaluate the effects of
‘chamber’ (SPT) and ‘1 min block’ (FC) as published
previously [36, 47]. Performance in the SPT was also
assessed using one sample t-tests to clarify whether
the percentage of time spent in the opponent/novel
chamber was greater than chance (50%). Differences
were regarded as significant if p< 0.05. Data are shown
as means ±standard error of means (SEM). F-values
and degrees of freedom are presented for ANOVAs
and significant ‘genotype’ and ‘treatment’ effects are
shown in figures and tables as ‘∗’(
∗p< 0.05, ∗∗p< 0.01)
and ‘#’ (#p<0.05) respectively whereas RM-ANOVA
results for social novelty preference are presented by
‘+’(
+p< 0.05, ++ p< 0.01, +++ p< 0.001). Analyses
were conducted using SPSS 20.0 for Windows.
RESULTS
Behavior
Sociability and social recognition
RM ANOVA revealed an effect of ‘chamber’
[F(1,39) = 197.9, p< 0.001] (Fig. 1A). All mice spent
more time investigating the social opponent over the
empty chamber, indicating intact sociability for all
mice regardless of genotype and treatment. T-tests
for percentage of time spent with the novel mouse
confirmed that all mice demonstrated significant lev-
els of sociability above chance [WT-VEH: t(6) = 7.4,
p< 0.001; A␤PP ×PS1-VEH: t(13) = 6.2, p< 0.001;
WT-CBD: t(8) = 12.3, p< 0.001; A␤PP ×PS1-CBD:
t(12) = 8.6, p< 0.001] (data not shown).
In the SPT, RM ANOVA revealed a significant effect
of ‘chamber’ for time spent investigating the novel over
the familiar mouse [F(1,41) = 23.6, p< 0.001]. Impor-
tantly, a significant interaction between ‘genotype’
and ‘treatment’ was found [F(1,41) = 4.8, p< 0.05],
where only vehicle-treated AD transgenic mice did not
develop a preference for the novel mouse (Fig. 1B).
Two-way ANOVA also revealed a trend toward an
effect of CBD treatment [F(1,41) = 3.1, p= 0.09].
Indeed, ANOVA split by ‘genotype’ revealed that CBD
increased the time AD transgenic mice spent with the
novel mouse [F(1,25) = 5.0, p< 0.05], with no such
effect observed in WT mice [F(1,16) = 0.2, p= 0.7]
(Fig. 1B) showing that CBD had a beneficial effect
on social recognition memory. T-tests confirmed that
all animals, except vehicle-treated A␤PP ×PS1 mice,
spent a significantly greater percentage of time with
the novel mouse than the familiar mouse [WT-VEH:
t(7) = 2.5, p< 0.05; A␤PP ×PS1-VEH: t(13) = 0.3,
p= 0.8; WT-CBD: t(9) = 3.3, p< 0.01; A␤PP ×PS1-
CBD: t(12) = 3.7, p< 0.01] (data not shown).
Anxiety
A␤PP ×PS1 transgenic mice demonstrated WT-like
locomotion and anxiety (p> 0.05 for total distance
travelled, time spent on open arms and open arm
entries). Chronic treatment with CBD had no effect
on EPM behaviors (all p> 0.05; Table 2).
Associative learning
All mice responded to the electric foot shocks during
conditioning (i.e., vocalization detected in all mice).
Two-way ANOVA found transgenic mice demon-
strated increased amounts of freezing at baseline (i.e.,
first 2 min pre-conditioning) regardless of treatment
[F(1,41) = 4.5, p< 0.05]. However, freezing duration
during the first 2 min of the context and cue trials
was similar for all mice across test conditions (all
p> 0.05; Table 3) and all mice exhibited intact context
memory regardless of treatment [p> 0.05; Fig. 2A].
Furthermore, memory of the cue was intact as all ani-
mals showed increased freezing post cue presentation
[RM ANOVA: F(1,41) = 52.9, p< 0.001], regardless of
‘genotype’ or ‘treatment’ (Fig. 2B and Table 3).
Brain pathophysiology
Amyloid load
One-way ANOVA revealed that CBD had no
effect on soluble and insoluble A␤40 or A␤42 in
the cortex of A␤PP ×PS1 mice, although insolu-
ble A␤42 was slightly higher after CBD treatment
[‘treatment’: Soluble A␤40: F(1,18) = 0.3, p= 0.6;
Insoluble A␤40: F(1,18) = 2.4, p= 0.1; Soluble A␤42:
F(1,18) = 0.1, p= 0.7; Insoluble A␤42 : F(1,18) = 3.5,
p= 0.08] (Table 4). Similarly, A␤levels remained
unchanged after CBD treatment in the hippocam-
pus [‘treatment’: Soluble A␤40: F(1,17) = 0.4, p= 0.6;
Insoluble A␤40: F(1,18) = 1.1, p= 0.3; Soluble A␤42:

AUTHOR COPY
D. Cheng et al. / Protective Effects of Cannabidiol 1389
Fig. 1. Sociability and social recognition were measured using the social preference test. Graphs show total time spent [s] in test chambers by
the test mice containing A) either an unfamiliar mouse (i.e., opponent) or an empty mouse enclosure (i.e., empty); or B) either a familiar or an
unfamiliar (i.e., novel) mouse. Data for non-transgenic wild type-like control (WT) and double transgenic A␤PPSwe/PS1E9 (A␤PP ×PS1)
male mice after vehicle or cannabidiol (CBD) treatment are shown as means + SEM. Significant ‘treatment’ effects are indicated with ‘#’
(#p<0.01). RM ANOVA for novelty preference are presented by ‘+’(
+p< 0.05, ++ p< 0.01, +++ p< 0.001).
Table 2
Anxiety-related behaviors (i.e., time spent on and entries onto open arms) and locomotion (total distance travelled) in the elevated plus
maze (EPM). Parameters for wild type-like control mice (WT) and double transgenic A␤PPSwe/PS1E9 (A␤PP ×PS1) mice after vehicle
or cannabidiol (CBD) treatment are shown as mean ±SEM
Vehicle CBD
WT A␤PP ×PS1 WT A␤PP ×PS1
Time spent on open arms [s] 7.2 ±2.7 6.0 ±2.3 6.7 ±1.6 7.3 ±2.7
Entries onto open arms [%] 13.3±4.0 7.8 ±3.1 11.6 ±2.1 11.2 ±2.7
Total distance travelled [m] 7.1 ±0.9 8.0 ±1.4 7.2 ±0.9 7.9 ±1.0
Fig. 2. Fear-associated learning was assessed in the fear conditioning test. Time spent freezing during A) the context test and B) the cue test for
each is shown per ‘1 min bin’. Data for non-transgenic control wild type-like (WT) and double transgenic A␤PPSwe/PS1E9 (A␤PP ×PS1)
mice after vehicle or cannabidiol (CBD) treatment are shown as means ±SEM.
F(1,15) = 0.3, p= 0.6; Insoluble A␤42 : F(1,18) = 0.1,
p= 0.7] (Table 4).
Oxidative damage
Total F2-isoprostanes (free and esterified corrected
for arachidonic acid; AA) were not significantly
altered in A␤PP ×PS1 mice when compared to
their WT littermates, regardless of ‘treatment’
(all p> 0.05) (Table 5). We also measured the
levels of oxysterols in the cortex. For enzymati-
cally oxidized sterols, A␤PP ×PS1 mice demon-
strated significantly decreased overall levels of
24-hydroxycholesterol compared to WT littermates
[‘genotype’: F(1,34) = 4.9, p< 0.05], whereas ‘treat-
ment’ had no effect on sterols [F(1,34) = 0.07, p= 0.8]
and no ‘genotype’ by ‘treatment’ interactions were
found. No differences were found across all four
groups for 27-hydroxycholesterol, and the reactive
species oxidized sterols, 7␤-hydroxycholesterol and
7-ketocholesterol (all p> 0.05) (Table 5).

AUTHOR COPY
1390 D. Cheng et al. / Protective Effects of Cannabidiol
Table 3
Fear-associated memory in the fear conditioning paradigm. Freezing (i.e., time spent freezing [s]) at baseline, and during context test and cue test
for non-transgenic wild type-like control (WT) and double transgenic A␤PPSwe/PS1E9 (A␤PP ×PS1) male mice after vehicle or cannabidiol
(CBD) treatment are presented as mean ±SEM
Vehicle CBD
WT A␤PP ×PS1 WT A␤PP ×PS1
Baseline (first 2 min)
Conditioning freezing [s] 2.3 ±1.1 5.8 ±1.7 4.7 ±1.5 8.5 ±1.9
Context freezing [s] 32.0 ±7.8 47.1 ±7.4 37.0 ±6.0 36.7 ±7.0
Context
Total time spent freezing [s] 132.2 ±29.6 152.3 ±28.1 131.6 ±15.4 150.7 ±25.8
Cue
Time spent freezing 2min prior to cue onset [s] 18.0 ±5.9 29.1 ±5.7 25.5 ±5.2 28.6±4.4
Time spent freezing 2min post cue onset [s] 40.4 ±7.3+++ 37.4 ±5.8+38.7 ±6.7+40.4 ±5.2+
Significant effects of cue presentation on freezing response are indicated by ‘+’(
+p< 0.05 and +++ p< 0.001)
Fig. 3. Neuroinflammation markers in cortical tissue. Quantitative PCR was used to measure the concentration of A) interleukin-1␤and B)
TNF-␣derived from the cortex of control (WT) and double transgenic A␤PPSwe/PS1E9 (A␤PP ×PS1) male mice after vehicle or cannabidiol
(CBD) treatment. Concentrations [ng/ ␮l] are presented as mean±SEM.
Table 4
Amyloid-␤Soluble and insoluble amyloid load in double trans-
genic A␤PPSwe/PS1E9 (A␤PP ×PS1) male mice after vehicle or
cannabidiol (CBD) treatment are shown as means ±SEM
A␤PP ×PS1 Vehicle CBD
Cortex
Soluble A␤40 [pg/mg] 1033.2 ±211.2 904.6 ±118.8
Soluble A␤42 [pg/mg] 654.3 ±102.6 613.4 ±65.8
Insoluble A␤40 [pg/mg] 8184.6 ±701.0 9758.3 ±751.6
Insoluble A␤42 [pg/mg] 25601.4 ±2138.6 30897.8 ±1847.6
Hippocampus
Soluble A␤40 [pg/mg] 464.4 ±99.3 390.8 ±69.6
Soluble A␤42 [pg/mg] 155.7 ±196.8 196.8 ±54.1
Insoluble A␤40 [pg/mg] 9854.5 ±2217.7 12776.1 ±1738.7
Insoluble A␤42 [pg/mg] 22295.1 ±7937.3 26370.0 ±8467.1
Cholesterol
Cholesterol was increased in cortical tissues
of A␤PP ×PS1 mice compared to WT animals
[F(1,34) = 12.1, p< 0.01] and CBD increased choles-
terol levels [F(1,34) = 11.0, p= 0.01]. Further one-way
ANOVA revealed that cholesterol was signifi-
cantly higher in vehicle-treated AD transgenic mice
[F(1,16) = 7.7, p< 0.05] compared to control mice,
while CBD increased the cholesterol levels in WT mice
[F(1,16) = 25.1, p< 0.001] but not A␤PP ×PS1 mice
[F(1,18) = 1.3, p= 0.3] (Table 5).
Two-way ANOVA revealed a significant ‘genotype’
by ‘treatment’ interaction for the cortical levels of
the dietary phytosterol, brassicasterol [F(1,34) = 6.1,
p< 0.05], which was caused by CBD increasing bras-
sicasterol levels in A␤PP ×PS1 mice only [WT:
F(1,16) = 0.5, p= 0.5 A␤PP ×PS1: F(1,18) = 6.9,
p< 0.05; Table 5]. Furthermore, the analysis detected
a ‘genotype’ effect in CBD-treated mice [vehicle:
F(1,16) = 0.2, p= 0.7, CBD: F(1,18) = 9.9, p< 0.01;
Table 5]. The dietary phytosterol, campesterol, was
also increased in A␤PP ×PS1 mice [F(1,34) = 4.4,
p< 0.05]. More specifically, cortical campesterol was
elevated in CBD-treated A␤PP ×PS1 mice [vehicle:
F(1,16) = 0.2, p= 0.6, CBD: F(1,18) = 9.0, p< 0.01].

AUTHOR COPY
D. Cheng et al. / Protective Effects of Cannabidiol 1391
Table 5
Oxysterols, F2-isoprostanes, cholesterol, and phytosterol levels. Oxidative damage and total cholesterol in the cortex of non-transgenic wild
type-like control (WT) and double transgenic A␤PPSwe/PS1E9 (A␤PP ×PS1) male mice after vehicle or cannabidiol (CBD) treatment.
Concentrations (in pg) are presented as mean ±SEM
Vehicle CBD
WT A␤PP ×PS1 WT A␤PP ×PS1
Oxidized sterols
Reactive species oxidized
7␤-hydroxycholesterol [ng/mg] 0.29 ±0.02 0.31 ±0.05 0.30 ±0.02 0.32 ±0.02
7-ketocholesterol [ng/mg] 0.53 ±0.04 0.50 ±0.04 0.47 ±0.03 0.57 ±0.05
Enzymatically oxidized
24-hydroxycholesterol [ng/mg] 38.5 ±2.8 34.2 ±3.0 39.4 ±2.7 31.9 ±1.8*
27-hydroxycholesterol [pg/mg] 48.3 ±5.3 38.9 ±5.7 46.1 ±2.5 43.4 ±4.6
F2-isoprostanes (normalized for arachidonic acid)
Total [pg/ng] 10.0 ±1.0 8.1 ±0.5 9.1 ±0.4 8.3 ±0.5
Cholesterol
Total cholesterol [ng/mg] 20.5 ±1.0 29.6 ±2.8 ∗29.3 ±1.4### 33.1 ±1.4
Dietary phytosterol
Brassicasterol [pg/mg] 51.2 ±3.8 48.3 ±6.1 46.7 ±4.9 70.4 ±5.8∗∗#
Campesterol [ng/mg] 15.1 ±1.8 16.5 ±2.3 15.0 ±1.7 21.2 ±1.1∗
Significant effects of ‘genotype’ are indicated by ‘∗’(
∗p< 0.05 and ∗∗ p< 0.01) and effects of ‘treatment’ by ‘#’ (#p< 0.05 and ### p< 0.001)
There was also a trend for CBD to increase the corti-
cal levels of this phytosterol in transgenic mice [WT:
F(1,16) = 0.0, p= 1.0, A␤PP ×PS1: F(1,18) = 3.4,
p= 0.08] (Table 5).
Inflammatory markers
Two-way ANOVA revealed no significant differ-
ences in the levels of mRNA for two inflammatory
cytokine markers across test conditions [‘genotype’:
IL-1␤: F(1,39) = 1.0, p= 0.3 – ‘TNF-␣’: F(1,39) = 1.1,
p= 0.3] (Fig. 3A-B). There was no significant effect of
‘treatment’ on these cytokines [IL-1␤: F(1,39) = 1.3,
p= 0.3 – TNF-␣: F(1,39) = 2.5, p= 0.1] either,
although cytokine levels of CBD-treated A␤PP ×PS1
mice appeared closer to corresponding WT levels
than levels of vehicle-treated AD transgenic mice
(Fig. 3A-B).
CBD plasma levels
Two-way ANOVA revealed that all mice treated
with CBD demonstrated significantly increased levels
of plasma CBD (ng/ml) [WT-CBD: 2.1 ±0.6;
A␤PP ×PS1-CBD: 1.9 ±0.4 – ‘treatment’:
F(1,30) = 21.3, p< 0.001]. No significant ‘geno-
type’ differences or interactions were found (all
p> 0.05). CBD could not be detected in mice that
were treated with vehicle (values < 0.05 ng/ml).
DISCUSSION
Our study demonstrates for the first time the effects
of long-term oral CBD treatment on the social recog-
nition memory and pathophysiology of a double trans-
genic A␤PP ×PS1 mouse model for AD. We provide
first evidence of a possible impact of CBD on dietary
phytosterols, which can exert beneficial effects on cog-
nition. We also suggest that the therapeutic effect of
CBD may be linked to neuroinflammatory processes
or changes in cholesterol but further research using
additional CBD doses will be necessary to clarify this.
The SPT determined that vehicle-treated
A␤PP ×PS1 mice exhibit a social recognition
memory deficit, confirming our earlier findings [36,
37]. Importantly, long-term CBD treatment pre-
vented this social recognition deficit from occurring
in A␤PP ×PS1 mice. We previously found that
intraperitoneal administration of CBD for three weeks
reversed this cognitive deficit in the same AD mouse
model [37]. Other recent studies also report social
recognition deficits in AD transgenic mouse models,
providing evidence for the increasing relevance of
social recognition memory testing for AD research
[60, 61]. Anxiety can confound the performance
of mice in cognitive tests [62] and acute CBD has
been found to modify anxiety-related behaviors [41,
63–67]. However, the A␤PP ×PS1 transgene did not
influence anxiety parameters nor did CBD treatment.
The beneficial effect of CBD on social recognition
memory was not associated with a direct effect on A␤
levels. Insoluble and soluble levels of A␤40 and A␤42
were no different between vehicle and CBD-treated
A␤PP ×PS1 mice in cortex and hippocampus. Simi-
larly, another study described improvements in spatial
memory in A␤PPSwe/PS1E9 mice on a C57BL/6J
background, which was not accompanied by changes

AUTHOR COPY
1392 D. Cheng et al. / Protective Effects of Cannabidiol
in A␤levels [68]. The same study also found that lev-
els of insoluble A␤40 and A␤42 in the parietal cortex
did not correlate with cognitive deficits [68]. Nonethe-
less, various in vitro studies show CBD can attenuate
A␤-induced processes [23–26], reverse A␤-induced
memory impairments in rodents [28] and reduce A␤
formation [69].
The in vivo formation of isoprostanes is a marker for
cerebral lipid oxidation and directly correlated with
an increase in oxidative stress [7, 70–72]. Patients
with AD are also known to have increased concentra-
tions of F2-isoprostanes in CSF even prior to disease
diagnosis [7, 8, 71]. Levels of oxidation were not sig-
nificantly altered in A␤PP ×PS1 mice in comparison
to their age-matched WT littermates, nor did we detect
changes in the level of lipid oxidation in the cortex
of CBD-treated animals, despite its known antioxidant
properties [73, 74]. These findings may be due to age
as nucleic acid oxidation is significantly higher in 3
and 5 months old A␤PP ×PS1 mice compared to age-
matched control mice. Importantly, this phenomenon
is not evident in 10 and 15 month old mice, which is
the age when brain tissue was collected for the current
study [75].
Cholesterol was increased in A␤PP ×PS1 mice
compared to WT mice, while CBD treatment increased
cholesterol levels in WT mice. Our finding of increased
cholesterol in A␤PP ×PS1 mice could indicate either
an impaired reuptake process, or a compensatory
mechanism for protection against neurodegeneration
in AD mice. Maintenance of sufficient cholesterol is
important to help combat synapse loss and neurode-
generation [76] and such a response is consistent with
the reduced levels of 24-OH cholesterol detected in
the A␤PP ×PS1 mice compared to WT. Formation of
24-OH cholesterol is the major pathway of cholesterol
removal from the brain [77]. Insufficient amounts of
cholesterol may interrupt essential processes such as
myelin formation, synaptic transmission and cognitive
ability in mice [78, 79], while a reduction in oxys-
terols has been shown to correlate with the severity
of dementia and brain atrophy [9, 10, 80]. Interest-
ingly, 8-month old A␤PPSwe/PS1E9 mice on a pure
C57BL/6J background did not demonstrate signifi-
cantly different levels of cholesterol, while 15-month
old transgenic mice had significantly lower cholesterol
levels compared to control mice [81]. It is noteworthy
that decreased levels of cholesterol in cerebral spinal
fluid and plasma have been found in patients with AD
[11, 12].
Phytosterols are present naturally in a variety of dif-
ferent foods, including grains (e.g., sorghum and bran)
found in mouse food pellets. CBD increased the lev-
els of brassicasterol and campesterol in A␤PP ×PS1
mice. The accumulation and long-term consump-
tion of dietary phytosterols do not interfere with
memory [82, 83]. On the other hand, dietary sup-
plementation of a fish oil-rich diet with phytosterols
reduced insoluble A␤42 in A␤PPSwe/PS1E9 mice on
a C57BL/6J background [68], while the phytosterol
stigmasterol attenuated scopolamine-induced spatial
memory deficits of mice [84]. These findings suggest
a potentially beneficial effect of increased phytosterol
levels for cognitive symptoms in the AD brain. It is
possible that CBD interacted with AD pathophysiol-
ogy by increasing the retention of specific phytosterols.
Further research needs to be conducted in order to
understand the effect of increased phytosterol levels
in AD brains and how CBD might be involved in these
processes.
Daily long-term administration of 20 mg/kg CBD
did not result in a statistically significant effect.
However, A␤PP ×PS1 mice have previously been
shown to exhibit elevated levels of neuroinflam-
mation (increased nitric oxide species and TNF-␣)
in the hippocampus [85] and inflammatory changes
have been linked to impaired spatial memory of
A␤PP ×PS1 mice [86]. Furthermore, Martin-Moreno
and colleagues have demonstrated that A␤-induced
neuroinflammation was decreased after CBD treatment
[28]. Thus, we suggest that CBD might be able to
combat increased inflammation in A␤PP ×PS1 mice
thereby impacting on the cognitive performance of
these mice. Future research should consider additional
CBD doses to determine the effects of long-term CBD
treatment on neuroinflammation in AD mouse models.
Furthermore, an escalating CBD dosage regime could
be used as the dosage of AD-approved treatments is
often increased over time as the condition of patients
deteriorates [87].
In conclusion, our study is the first to demonstrate
that long-term CBD treatment can prevent the devel-
opment of a social recognition deficit in A␤PP ×PS1
mice. The findings suggest the mechanism involved
in this prevention may be linked to CBD-induced
retention of dietary phytosterols or neuroinflammatory
processes in the brain of AD mice. We provide the first
evidence that CBD has potential to be used as a long-
term preventative treatment option for AD and may
be especially relevant for symptoms of social with-
drawal and facial recognition. The behavioral inertness
of CBD and the fact that CBD is well tolerated in
humans [88, 89] suggests that preclinical research find-
ings could easily be followed up in clinical trials.

AUTHOR COPY
D. Cheng et al. / Protective Effects of Cannabidiol 1393
Future studies using cytokine arrays or an ‘omics’
approach may reveal which biochemical/genetic path-
ways contribute to the beneficial effects of CBD.
It will also be important to clarify what receptors
mediate the therapeutic-like effects of CBD: the per-
oxisome proliferator-activated receptor-␥[25, 69],
N-methyl-D-aspartate receptors [90, 91], and recep-
tors of the endocannabinoid system [92] are promising
targets.
ACKNOWLEDGMENTS
TK is supported by the Schizophrenia Research
Institute utilizing infrastructure funding from NSW
Ministry of Health and a career development fellow-
ship (1045643) from the National Health and Medical
Research Council (NHMRC). BG is supported by an
NHMRC Senior Research Fellowship (630445). DC
received an Australian Postgraduate Award scholar-
ship from the University of New South Wales and a
supplementary scholarship provided by Neuroscience
Research Australia. We thank Jac Kee Low and Warren
Logge (Neuroscience Research Australia, Randwick,
Australia) for assisting in treatment of our test mice,
Kalani Ruberu and Hongyun (Henry) Li (University
of Wollongong, Wollongong, Australia) for provid-
ing extensive ELISA and qPCR training, Lei Zhang
(Garvan Institute of Medical Research, Darlinghurst,
Australia) for providing assistance in the gel pel-
let setup, Jerry Tanda for critical comments on the
manuscript, and the staff of the Australian BioRe-
sources and Adam Bryan at Neuroscience Research
Australia for taking care of our test mice.
Authors’ disclosures available online (http://www.j-
alz.com/disclosures/view.php?id=2337).
REFERENCES
[1] Chung JA, Cummings JL (2000) Neurobehavioral and
neuropsychiatric symptoms in Alzheimer’s disease: Charac-
teristics and treatment. Neurol Clin 18, 829-846.
[2] Reisberg B, Ferris SH, de Leon MJ, Crook T (1982) The
Global Deterioration Scale for assessment of primary degen-
erative dementia. Am J Psychiatry 139, 1136-1139.
[3] Haga S, Akai K, Ishii T (1989) Demonstration of microglial
cells in and around senile (neuritic) plaques in the Alzheimer
brain. An immunohistochemical study using a novel mono-
clonal antibody. Acta Neuropathol 77, 569-575.
[4] Itagaki S, McGeer PL, Akiyama H, Zhu S, Selkoe D (1989)
Relationship of microglia and astrocytes to amyloid deposits
of Alzheimer disease. J Neuroimmunol 24, 173-182.
[5] Rogers J, Lue LF (2001) Microglial chemotaxis, activation,
and phagocytosis of amyloid beta-peptide as linked phenom-
ena in Alzheimer’s disease. Neurochem Int 39, 333-340.
[6] Streit WJ (2004) Microglia and Alzheimer’s disease patho-
genesis. J Neurosci Res 77, 1-8.
[7] Dalle-Donne I, Rossi R, Colombo R, Giustarini D, Milzani
A (2006) Biomarkers of oxidative damage in human disease.
Clin Chem 52, 601-623.
[8] Pratico D, Clark CM, Liun F, Rokach J, Lee VY, Trojanowski
JQ (2002) Increase of brain oxidative stress in mild cognitive
impairment: A possible predictor of Alzheimer disease. Arch
Neurol 59, 972-976.
[9] Papassotiropoulos A, Lutjohann D, Bagli M, Locatelli S,
Jessen F, Rao ML, Maier W, Bjorkhem I, von Bergmann K,
Heun R (2000) Plasma 24S-hydroxycholesterol: A peripheral
indicator of neuronal degeneration and potential state marker
for Alzheimer’s disease. Neuroreport 11, 1959-1962.
[10] Solomon A, Leoni V, Kivipelto M, Besga A, Oksengard
AR, Julin P, Svensson L, Wahlund LO, Andreasen N, Win-
blad B, Soininen H, Bjorkhem I (2009) Plasma levels of
24S-hydroxycholesterol reflect brain volumes in patients
without objective cognitive impairment but not in those with
Alzheimer’s disease. Neurosci Lett 462, 89-93.
[11] Kolsch H, Heun R, Jessen F, Popp J, Hentschel F, Maier W,
Lutjohann D (2010) Alterations of cholesterol precursor levels
in Alzheimer’s disease. Biochim Biophys Acta 1801, 945-950.
[12] Mielke MM, Zandi PP, Sjogren M, Gustafson D, Ostling S,
Steen B, Skoog I (2005) High total cholesterol levels in late
life associated with a reduced risk of dementia. Neurology 64,
1689-1695.
[13] Marangoni F, Poli A (2010) Phytosterols and cardiovascular
health. Pharmacol Res 61, 193-199.
[14] Burg VK, Grimm HS, Rothhaar TL, Grosgen S, Hundsdorfer
B, Haupenthal VJ, Zimmer VC, Mett J, Weingartner O, Laufs
U, Broersen LM, Tanila H, Vanmierlo T, Lutjohann D, Hart-
mann T,Grimm MO (2013) Plant sterols the better cholesterol
in Alzheimer’s disease? A mechanistical study. J Neurosci 33,
16072-16087.
[15] Vanmierlo T, Popp J, Kolsch H, Friedrichs S, Jessen F, Stoffel-
WagnerB, Bertsch T, Hartmann T,Maier W, von BergmannK,
Steinbusch H, Mulder M, Lutjohann D (2011) The plant sterol
brassicasterol as additional CSF biomarker in Alzheimer’s
disease. Acta Psychiatr Scand 124, 184-192.
[16] Massoud F, Leger GC (2011) Pharmacological treatment of
Alzheimer disease. Can J Psychiatry 56, 579-588.
[17] Karl T, Cheng D, Garner B, Arnold JC (2012) The therapeu-
tic potential of the endocannabinoid system for Alzheimer’s
disease. Expert Opin Ther Targets 16, 407-420.
[18] Booz GW (2011) Cannabidiol as an emergent therapeutic
strategy for lessening the impact of inflammation on oxidative
stress. Free Radic Biol Med 51, 1054-1061.
[19] Iuvone T, Esposito G, De Filippis D, Scuderi C, Steardo L
(2009) Cannabidiol: A promising drug for neurodegenerative
disorders? CNS Neurosci Ther 15, 65-75.
[20] Krishnan S, Cairns R, Howard R (2009) Cannabinoids for
the treatment of dementia. Cochrane Database Syst Rev,
CD007204.
[21] Scuderi C, Esposito G, Blasio A, Valenza M, Arietti P, Steardo
L Jr, Carnuccio R, De Filippis D, Petrosino S, Iuvone T, Di
Marzo V, Steardo L (2011) Palmitoylethanolamide counter-
acts reactive astrogliosis induced by beta-amyloid peptide. J
Cell Mol Med 15, 2664-2674.
[22] Zuardi AW (2008) Cannabidiol: From an inactivecannabinoid
to a drug with wide spectrum of action. Rev Bras Psiquiatr
30, 271-280.
[23] Esposito G, De Filippis D, Carnuccio R, Izzo AA, Iuvone T
(2006) The marijuana component cannabidiol inhibits beta-
amyloid-induced tau protein hyperphosphorylation through

AUTHOR COPY
1394 D. Cheng et al. / Protective Effects of Cannabidiol
Wnt/beta-catenin pathway rescue in PC12 cells. J Mol Med
84, 253-258.
[24] Iuvone T, Esposito G, Esposito R, Santamaria R, Di Rosa
M, Izzo AA (2004) Neuroprotective effect of cannabidiol, a
non-psychoactive component from Cannabis sativa, on beta-
amyloid-induced toxicity in PC12 cells. J Neurochem89 ,134-
141.
[25] Esposito G, Scuderi C, Valenza M, Togna GI, Latina V, De Fil-
ippis D, Cipriano M, Carratu MR, Iuvone T, Steardo L (2011)
Cannabidiol Reduces Abeta-Induced Neuroinflammation and
Promotes Hippocampal Neurogenesis through PPARgamma
Involvement. PLoS One 6, e28668.
[26] Wolf SA, Bick-Sander A, Fabel K, Leal-Galicia P, Tauber S,
Ramirez-Rodriguez G, Muller A, Melnik A, Waltinger TP,
Ullrich O, Kempermann G (2010) Cannabinoid receptor CB1
mediates baseline and activity-induced survival of new neu-
rons in adult hippocampal neurogenesis. Cell Commun Signal
8, 12.
[27] Esposito G, Scuderi C, Savani C, Steardo L Jr, De Filippis D,
Cottone P, Iuvone T, Cuomo V, Steardo L (2007) Cannabidiol
in vivo blunts beta-amyloid induced neuroinflammation by
suppressing IL-1beta and iNOS expression. Br J Pharmacol
151, 1272-1279.
[28] Martin-Moreno AM, Reigada D, Ramirez BG, Mechoulam
R, Innamorato N, Cuadrado A, de Ceballos ML (2011)
Cannabidiol and other cannabinoids reduce microglial acti-
vation in vitro and in vivo: Relevance to Alzheimer’s disease.
Mol Pharmacol 79, 964-973.
[29] Hampel H (2012) Current insights into the pathophysi-
ology of Alzheimer’s disease: Selecting targets for early
therapeutic intervention. Int Psychogeriatr 24(Suppl 1),
S10-S17.
[30] Riedel WJ (2014) Preventing cognitive decline in preclinical
Alzheimer’s disease. Curr Opin Pharmacol 14, 18-22.
[31] Borchelt DR, Ratovitski T, van Lare J, Lee MK, Gonzales
V, Jenkins NA, Copeland NG, Price DL, Sisodia SS (1997)
Accelerated amyloid deposition in the brains of transgenic
mice coexpressing mutant presenilin 1 and amyloid precursor
proteins. Neuron 19, 939-945.
[32] Jankowsky JL, Fadale DJ, Anderson J, Xu GM, Gonzales
V, Jenkins NA, Copeland NG, Lee MK, Younkin LH, Wag-
ner SL, Younkin SG, Borchelt DR (2004) Mutant presenilins
specifically elevate the levels of the 42 residue beta-amyloid
peptide in vivo: Evidence for augmentation of a 42-specific
gamma secretase. Hum Mol Genet 13, 159-170.
[33] Jankowsky JL, Slunt HH, Gonzales V, Jenkins NA, Copeland
NG, Borchelt DR (2004) APP processing and amyloid depo-
sition in mice haplo-insufficient for presenilin 1. Neurobiol
Aging 25, 885-892.
[34] Machova E, Rudajev V, Smyckova H, Koivisto H, Tanila H,
Dolezal V (2010) Functional cholinergic damage develops
with amyloid accumulation in young adult APPswe/PS1dE9
transgenic mice. Neurobiol Dis 38, 27-35.
[35] Wang J, Tanila H, Puolivali J, Kadish I, van Groen T (2003)
Gender differences in the amount and deposition of amyloid-
beta in APPswe and PS1 double transgenic mice. Neurobiol
Dis 14, 318-327.
[36] Cheng D, Logge W, Low JK, Garner B, Karl T (2013) Novel
behavioural characteristics of the APP(Swe)/PS1DeltaE9
transgenic mouse model of Alzheimer’s disease. Behav Brain
Res 245, 120-127.
[37] Cheng D, Low JK, Logge W, Garner B, Karl T (2014) Chronic
cannabidiol treatment improves social and object recognition
in double transgenic APP/PS1E9 mice. Psychopharmacology
(Berl), doi: 10.1007/s00213-014-3478-5.
[38] Bhaskar K, Maphis N, Xu G, Varvel NH, Kokiko-Cochran
ON, Weick JP, Staugaitis SM, Cardona A, Ransohoff RM,
Herrup K, Lamb BT (2014) Microglial derived tumor necrosis
factor-alpha drives Alzheimer’s disease-related neuronal cell
cycle events. Neurobiol Dis 62, 273-285.
[39] Torres KC, Lima GS, Fiamoncini CM, Rezende VB,
Pereira PA, Bicalho MA, Moraes EN, Romano-Silva MA
(2014) Increased frequency of cluster of differentiation 14
(CD14+) monocytes expressing interleukin 1 beta (IL-1beta)
in Alzheimer’s disease patients and intermediate levels in
late-onset depression patients. Int J Geriatr Psychiatry 29,
137-143.
[40] Jankowsky JL, Slunt HH, Ratovitski T,Jenkins NA, Copeland
NG, Borchelt DR (2001) Co-expression of multiple trans-
genes in mouse CNS: A comparison of strategies. Biomol
Eng 17, 157-165.
[41] Long LE, Chesworth R, Huang XF, McGregor IS, Arnold
JC, Karl T (2010) A behavioural comparison of acute
and chronic Delta9-tetrahydrocannabinol and cannabidiol in
C57BL/6JArc mice. Int J Neuropsychopharmacol 13, 861-
876.
[42] Martin-Moreno AM, Brera B, Spuch C, Carro E, Garcia-
Garcia L, Delgado M, Pozo MA, Innamorato NG, Cuadrado
A, de Ceballos ML (2012) Prolonged oral cannabinoid admin-
istration prevents neuroinflammation, lowers beta-amyloid
levels and improves cognitive performance in Tg APP 2576
mice. J Neuroinflammation 9,8.
[43] Long LE, Chesworth R, Huang XF, Wong A, Spiro A, McGre-
gor IS, Arnold JC, Karl T (2012) Distinct neurobehavioural
effects of cannabidiol in transmembrane domain neuregulin
1 mutant mice. PLoS One 7, e34129.
[44] Zhang L (2011) Voluntary oral administration of drugs
in mice. Protocol Exchange, doi:10.1038/protex.2011.
236.
[45] Zhang L, Lee NJ, Nguyen AD, Enriquez RF, Riepler SJ,
Stehrer B, Yulyaningsih E, Lin S, Shi YC, Baldock PA,
Herzog H, Sainsbury A (2010) Additive actions of the
cannabinoid and neuropeptide Y systems on adiposity and
lipid oxidation. Diabetes Obes Metab 12, 591-603.
[46] Karl T, Bhatia S, Cheng D, Kim WS, Garner B (2012) Cogni-
tive phenotyping of amyloid precursor protein transgenic J20
mice. Behav Brain Res 228, 392-397.
[47] Logge W, Cheng D, Chesworth R, Bhatia S, Garner B, Kim
WS, Karl T (2012) Role of Abca7 in mouse behaviours rele-
vant to neurodegenerative diseases. PLoS One 7, e45959.
[48] Moy SS, Nadler JJ, Perez A, Barbaro RP,Johns JM, Magnuson
TR, Piven J, Crawley JN (2004) Sociability and preference
for social novelty in five inbred strains: An approach to
assess autistic-like behavior in mice. Genes Brain Behav 3,
287-302.
[49] Montgomery KC (1955) The relation between fear induced by
novel stimulation and exploratory behavior. J Comp Physiol
Psychol 48, 254-260.
[50] Montgomery KC, Monkman JA (1955) The relation between
fear and exploratory behavior. J Comp Physiol Psychol 48,
132-136.
[51] Karl T, Duffy L, Herzog H (2008) Behavioural profile of a
new mouse model for NPY deficiency. Eur J Neurosci 28,
173-180.
[52] Chesworth R, Downey L, Logge W,Killcross S, Karl T (2012)
Cognition in female transmembrane domain neuregulin 1
mutant mice. Behav Brain Res 226, 218-223.
[53] Duffy L, Cappas E, Lai D, Boucher AA, Karl T (2010) Cog-
nition in transmembrane domain neuregulin 1 mutant mice.
Neuroscience 170, 800-807.

AUTHOR COPY
D. Cheng et al. / Protective Effects of Cannabidiol 1395
[54] Kim WS, Li H, Ruberu K, Chan S, Elliott DA, Low JK, Cheng
D, Karl T, Garner B (2013) Deletion of Abca7 increases cere-
bral amyloid-beta accumulation in the J20 mouse model of
Alzheimer’s disease. J Neurosci 33, 4387-4394.
[55] Elliott DA, Tsoi K, Holinkova S, Chan SL, Kim WS, Halliday
GM, Rye KA, Garner B (2011) Isoform-specific proteolysis
of apolipoprotein-E in the brain. Neurobiol Aging 32, 257-
271.
[56] Bhatia S, Jenner AM, Li H, Ruberu K, Spiro AS, Shepherd
CE, Kril JJ, Kain N, Don A, Garner B (2013) Increased
apolipoprotein D dimer formation in Alzheimer’s disease hip-
pocampus is associated with lipid conjugated diene levels. J
Alzheimers Dis 35, 475-486.
[57] Cheng D, Jenner AM, Shui G, Cheong WF, Mitchell TW,
Nealon JR, Kim WS, McCann H, Wenk MR, Halliday GM,
Garner B (2011) Lipid pathway alterations in Parkinson’s
disease primary visual cortex. PLoS One 6, e17299.
[58] Nadulski T, Pragst F, Weinberg G, Roser P, Schnelle
M, Fronk EM, Stadelmann AM (2005) Randomized,
double-blind, placebo-controlled study about the effects of
cannabidiol (CBD) on the pharmacokinetics of Delta9-
tetrahydrocannabinol (THC) after oral application of THC
verses standardized cannabis extract. Ther Drug Monit 27,
799-810.
[59] Klein C, Karanges E, Spiro A, Wong A, Spencer J, Huynh T,
Gunasekaran N, Karl T, Long LE, Huang XF, Liu K, Arnold
JC, McGregor IS (2011) Cannabidiol potentiates Delta(9)-
tetrahydrocannabinol (THC) behavioural effects and alters
THC pharmacokinetics during acute and chronic treatment in
adolescent rats. Psychopharmacology (Berl) 218, 443-457.
[60] Faizi M, Bader PL, Saw N, Nguyen TV, Beraki S, Wyss-Coray
T, Longo FM, Shamloo M (2012) Thy1-hAPP(Lond/Swe+)
mouse model of Alzheimer’s disease displays broad behav-
ioral deficits in sensorimotor, cognitive and social function.
Brain Behav 2, 142-154.
[61] Filali M, Lalonde R, Rivest S (2011) Anomalies in social
behaviors and exploratory activities in an APPswe/PS1 mouse
model of Alzheimer’s disease. Physiol Behav 104, 880-885.
[62] Lipp HP, Wolfer DP (1998) Genetically modified mice and
cognition. Curr Opin Neurobiol 8, 272-280.
[63] Campos AC, Guimaraes FS (2008) Involvement of 5HT1A
receptors in the anxiolytic-like effects of cannabidiol injected
into the dorsolateral periaqueductal gray of rats. Psychophar-
macology (Berl) 199, 223-230.
[64] Campos AC, Ortega Z, Palazuelos J, Fogaca MV, Aguiar DC,
Diaz-Alonso J, Ortega-Gutierrez S, Vazquez-Villa H, Moreira
FA, Guzman M, Galve-Roperh I, Guimaraes FS (2013) The
anxiolytic effect of cannabidiol on chronically stressed mice
depends on hippocampal neurogenesis: Involvement of the
endocannabinoid system. Int J Neuropsychopharmacol 16,
1407-1419.
[65] Guimaraes FS, Chiaretti TM, Graeff FG, Zuardi AW (1990)
Antianxiety effect of cannabidiol in the elevated plus-maze.
Psychopharmacology (Berl) 100, 558-559.
[66] Moreira FA, Aguiar DC, Guimaraes FS (2006) Anxiolytic-
like effect of cannabidiol in the rat Vogel conflict test. Prog
Neuropsychopharmacol Biol Psychiatry 30, 1466-1471.
[67] Onaivi ES, Green MR, Martin BR (1990) Pharmacological
characterization of cannabinoids in the elevated plus maze. J
Pharmacol Exp Ther 253, 1002-1009.
[68] Koivisto H, Grimm MO, Rothhaar TL, Berkecz R, Lutjohann
DD, Giniatullina R, Takalo M, Miettinen PO, Lahtinen HM,
Giniatullin R, Penke B, Janaky T, Broersen LM, Hartmann T,
Tanila H (2014) Special lipid-based diets alleviate cognitive
deficits in the APPswe/PS1dE9 transgenic mouse model of
Alzheimer’s disease independent of brain amyloid deposition.
J Nutr Biochem 25, 157-169.
[69] Scuderi C, Steardo L, Esposito G (2013) Cannabidiol
promotes amyloid precursor protein ubiquitination and
reduction of beta amyloid expression in SHSY5Y cells
through PPARgamma involvement. Phytother Res, doi:
10.1002/ptr.5095.
[70] Morrow JD (2005) Quantification of isoprostanes as indices
of oxidant stress and the risk of atherosclerosis in humans.
Arterioscler Thromb Vasc Biol 25, 279-286.
[71] Pratico D, Lawson JA, Rokach J, FitzGerald GA (2001) The
isoprostanes in biology and medicine. Trends Endocrinol
Metab 12, 243-247.
[72] Pratico D, Rokach J, Lawson J, FitzGerald GA (2004) F2-
isoprostanes as indices of lipid peroxidation in inflammatory
diseases. Chem Phys Lipids 128, 165-171.
[73] Hamelink C, Hampson A, Wink DA, Eiden LE, Eskay RL
(2005) Comparison of cannabidiol, antioxidants, and diuretics
in reversing binge ethanol-induced neurotoxicity. J Pharma-
col Exp Ther 314, 780-788.
[74] Rajesh M, Mukhopadhyay P, Batkai S, Patel V, Saito K,
Matsumoto S, Kashiwaya Y, Horvath B, Mukhopadhyay
B, Becker L, Hasko G, Liaudet L, Wink DA, Veves A,
Mechoulam R, Pacher P (2010) Cannabidiol attenuates car-
diac dysfunction, oxidative stress, fibrosis, and inflammatory
and cell death signaling pathways in diabetic cardiomyopathy.
J Am Coll Cardiol 56, 2115-2125.
[75] Hamilton A, Holscher C (2012) The effect of ageing on neu-
rogenesis and oxidative stress in the APP(swe)/PS1(deltaE9)
mouse model of Alzheimer’s disease. Brain Res 1449,
83-93.
[76] Koudinov AR, Koudinova NV (2001) Essential role for
cholesterol in synaptic plasticity and neuronal degeneration.
FASEB J 15, 1858-1860.
[77] Lutjohann D, Breuer O, Ahlborg G, Nennesmo I, Siden A,
Diczfalusy U, Bjorkhem I (1996) Cholesterol homeostasis in
human brain: Evidence for an age-dependent flux of 24S-
hydroxycholesterol from the brain into the circulation. Proc
Natl Acad SciUSA93, 9799-9804.
[78] Suzuki R, Ferris HA, Chee MJ, Maratos-Flier E, Kahn CR
(2013) Reduction of the cholesterol sensor SCAP in the brains
of mice causes impaired synaptic transmission and altered
cognitive function. PLoS Biol 11, e1001532.
[79] Haque ZU, Mozaffar Z (1992) Importance of dietary choles-
terol for the maturation of mouse brain myelin. Biosci
Biotechnol Biochem 56, 1351-1354.
[80] Leoni V, Caccia C (2011) Oxysterols as biomarkers in neu-
rodegenerative diseases. Chem Phys Lipids 164, 515-524.
[81] Hooijmans CR, Van der Zee CE, Dederen PJ, Brouwer KM,
Reijmer YD, van Groen T, Broersen LM, Lutjohann D, Heer-
schap A, Kiliaan AJ (2009) DHA and cholesterol containing
diets influence Alzheimer-like pathology, cognition and cere-
bral vasculature in APPswe/PS1dE9 mice. Neurobiol Dis 33,
482-498.
[82] Vanmierlo T, Rutten K, van Vark-van der Zee LC, Friedrichs
S, Bloks VW, Blokland A, Ramaekers FC, Sijbrands E,
Steinbusch H, Prickaerts J, Kuipers F, Lutjohann D, Mulder
M (2011) Cerebral accumulation of dietary derivable plant
sterols does not interfere with memory and anxiety related
behavior in Abcg5-/- mice. Plant Foods Hum Nutr 66, 149-
156.
[83] Schiepers OJ, de Groot RH, van Boxtel MP, Jolles J, de Jong
A, Lutjohann D, Plat J, Mensink RP (2009) Consuming func-
tional foods enriched with plant sterol or stanol esters for 85
weeks does not affect neurocognitive functioning or mood in

AUTHOR COPY
1396 D. Cheng et al. / Protective Effects of Cannabidiol
statin-treated hypercholesterolemic individuals. J Nutr 139,
1368-1373.
[84] Park SJ, Kim DH, Jung JM, Kim JM, Cai M, Liu X, Hong JG,
Lee CH, Lee KR, Ryu JH (2012) The ameliorating effects of
stigmasterol on scopolamine-induced memory impairments
in mice. Eur J Pharmacol 676, 64-70.
[85] Kalifa S, Polston EK, Allard JS, Manaye KF (2011)
Distribution patterns of cannabinoid CB1 receptors in the hip-
pocampus of APPswe/PS1DeltaE9 double transgenic mice.
Brain Res 1376, 94-100.
[86] Gallagher JJ, Minogue AM, Lynch MA (2013) Impaired per-
formance of female APP/PS1 mice in the Morris water maze
is coupled with increased Abeta accumulation and microglial
activation. Neurodegener Dis 11, 33-41.
[87] Tariot P, Salloway S, Yardley J, Mackell J, Moline M (2012)
Long-term safety and tolerability of donepezil 23mg in
patients with moderate to severe Alzheimer’s disease. BMC
Res Notes 5, 283.
[88] Consroe P, Laguna J, Allender J, Snider S, Stern L, Sandyk
R, Kennedy K, Schram K (1991) Controlled clinical trial
of cannabidiol in Huntington’s disease. Pharmacol Biochem
Behav 40, 701-708.
[89] Cunha JM, Carlini EA, Pereira AE, Ramos OL, Pimentel C,
Gagliardi R, Sanvito WL, Lander N, Mechoulam R (1980)
Chronic administration of cannabidiol to healthy volunteers
and epileptic patients. Pharmacology 21, 175-185.
[90] Hallak JE, Dursun SM, Bosi DC, de Macedo LR, Machado-
de-Sousa JP, Abrao J, Crippa JA, McGuire P, Krystal JH,
Baker GB, Zuardi AW (2011) The interplay of cannabinoid
and NMDA glutamate receptor systems in humans: Pre-
liminary evidence of interactive effects of cannabidiol and
ketamine in healthy human subjects. Prog Neuropsychophar-
macol Biol Psychiatry 35, 198-202.
[91] Hampson AJ, Grimaldi M, Axelrod J, Wink D (1998)
Cannabidiol and (-)Delta9-tetrahydrocannabinol are neuro-
protective antioxidants. Proc Natl Acad Sci U S A 95,
8268-8273.
[92] Pertwee RG (2008) The diverse CB1 and CB2 receptor
pharmacology of three plant cannabinoids: Delta9-tetrahy-
drocannabinol, cannabidiol and delta9-tetrahydrocanna-
bivarin. Br J Pharmacol 153, 199-215.