Content uploaded by Andrew Scholey
Author content
All content in this area was uploaded by Andrew Scholey
Content may be subject to copyright.
http://jop.sagepub.com/
Journal of Psychopharmacology
http://jop.sagepub.com/content/25/8/1088
The online version of this article can be found at:
DOI: 10.1177/0269881110385594
2011 25: 1088 originally published online 11 October 2010J Psychopharmacol
Crystal F Haskell
David O Kennedy, Fiona L Dodd, Bernadette C Robertson, Edward J Okello, Jonathon L Reay, Andrew B Scholey and
improves cognitive performance and mood in healthy adults
) with cholinesterase inhibiting propertiesSalvia lavandulaefoliaMonoterpenoid extract of sage (
Published by:
http://www.sagepublications.com
On behalf of:
British Association for Psychopharmacology
can be found at:Journal of PsychopharmacologyAdditional services and information for
http://jop.sagepub.com/cgi/alertsEmail Alerts:
http://jop.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
Original Paper
Monoterpenoid extract of sage
(Salvia lavandulaefolia) with
cholinesterase inhibiting properties
improves cognitive performance and
mood in healthy adults
David O Kennedy
1
, Fiona L Dodd
1
, Bernadette C Robertson
1
, Edward
J Okello
2
, Jonathon L Reay
1
, Andrew B Scholey
3
and Crystal F Haskell
1
Abstract
Extracts of sage (Salvia officinalis/lavandulaefolia) with terpenoid constituents have previously been shown to inhibit cholinesterase and improve
cognitive function. The current study combined an in vitro investigation of the cholinesterase inhibitory properties and phytochemical constituents of a
S. lavandulaefolia essential oil, with a double-blind, placebo-controlled, balanced crossover study assessing the effects of a single dose on cognitive
performance and mood. In this latter investigation 36 healthy participants received capsules containing either 50 mL of the essential oil or placebo on
separate occasions, 7 days apart. Cognitive function was assessed using a selection of computerized memory and attention tasks and the Cognitive
Demand Battery before the treatment and 1-h and 4-h post-dose. The essential oil was a potent inhibitor of human acetylcholinesterase (AChE) and
consisted almost exclusively of monoterpenoids. Oral consumption lead to improved performance of secondary memory and attention tasks, most
notably at the 1-h post-dose testing session, and reduced mental fatigue and increased alertness which were more pronounced 4-h post-dose. These
results extend previous observations of improved cognitive performance and mood following AChE inhibitory sage extracts and suggest that the ability
of well-tolerated terpenoid-containing extracts to beneficially modulate cholinergic function and cognitive performance deserves further attention.
Keywords
Acetylcholinesterase, attention, memory, monoterpenoid, mood, sage, salvia, Salvia lavandulaefolia
Introduction
The ability of any plant extract to modulate central nervous
system (CNS) function in humans is almost exclusively due to
the presence of ‘secondary metabolites’: constitutive or
induced compounds that are not involved in the immediate
physiological survival of the plant, but which fill ecological
roles that increase the general ‘survivability’ of the plant over
the longer term (Harborne, 1993; Macias et al., 2007). For
instance, one of the largest structural groups of phytochemi-
cals, the alkaloids, predominantly function as feeding and
allelopathic deterrents, and their biotic interactions include
modulating neurotransmitter systems within their intended
target phytophages (Harborne, 1993; Wink, 2003; Wink
et al., 1998). These effects include interactions with the
cholinergic system, and either directly or indirectly with the
dopaminergic system, in herbivores (Hagen et al., 2009).
Whilst alkaloids are always toxic, to mammals, when
used at lower doses their neuro-toxic properties have the
unintended attribute of making them useful as psychotropic
agents in humans. Plant alkaloids, and derivates thereof,
constitute the active components in many of our social
drugs (Hagen et al., 2009) and have provided a multitude
of medicinal compounds with CNS activity (Samuelsson,
2004). These include a number of potentially toxic cholines-
terase inhibitors initially derived from alkaloid phytochemi-
cals and used to treat dementia, including galantamine,
huperzine, physostigmine and rivastigmine (Mukherjee
et al., 2007).
In contrast to the alkaloids, the terpenoid family of phy-
tochemicals has a spectrum of toxicity extending from highly
toxic to benign. In line with this, they exert a more complex
pattern of ecological effects for the synthesizing plant.
1
Brain, Performance and Nutrition Research Centre, Northumbria
University, Newcastle upon Tyne, UK.
2
School of Agriculture Food and Rural Development, University of
Newcastle, Newcastle upon Tyne, UK.
3
NICM Collaborative Centre in Neurocognition, Brain Sciences Institute,
Swinburne University, Melbourne, Australia.
Corresponding author:
David Kennedy, Brain, Performance and Nutrition Research Centre,
Northumbria University, Newcastle upon Tyne, NE1 8ST, UK
Email: david.kennedy@northumbria.ac.uk
Journal of Psychopharmacology
25(8) 1088–1100
! The Author(s) 2011
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269881110385594
jop.sagepub.com
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
These include toxic deterrence, but also attraction via taste,
scent and visual cues, and direct modulation of animals’ CNS
function. As an example, terpenoids can play a direct defen-
sive role as constitutive toxic deterrents (Wittstock and
Gershenzon, 2002). However, they also function as attrac-
tants for insects, both for pollination (Raguso et al., 2006),
and also as indirect defence agents. In this role they are syn-
thesized and released by the plant in response to the presence
of a herbivore in order to attract its natural predators (Maffei
et al., 2007). Terpenoids are amongst the putative active com-
ponents of many herbal extracts with low levels of side effects
that have been shown to affect CNS function, including
Valeriana officinalis (Valerian), Bacopa monniera, Panax gin-
seng and Ginkgo biloba (Kennedy, 2009).
Extracts or leaves of several edible members of the Salvia
(sage) genus, including Salvia officinalis (SO) and Salvia
lavandulaefolia (SL) have been used by humans for more
than two millennia to improve cognitive function and atten-
uate cognitive decline (Kennedy and Scholey, 2006). Their
putative active components encompass a broad range of
terpenoids, including a-pinene, b-pinene, 1,8-cineole, thujone
(SO only), camphor and geraniol (Perry et al., 2001). The
secondary metabolite environmental roles of these phyto-
chemicals may well underlie demonstrations of both acetyl-
cholinesterase (AChE) and butyrylcholinesterase (BuChE)
inhibition seen in vitro in human brain homogenates and
erythrocytes (Kennedy et al., 2006; Perry et al., 1996, 2000,
2001; Savelev et al., 2003, 2004; Scholey et al., 2008; Tildesley
et al., 2003, 2005) and AChE inhibition seen in vivo in the
rodent brain (Perry et al., 2002).
These cholinergic effects are matched by findings of
improved memory performance in rats (Eidi et al., 2006)
and demonstrations of improved cognitive function in
humans following administration of single doses of SO and
SL with cholinesterase-inhibiting properties. In the first of a
series of double-blind, placebo-controlled, balanced cross-
over studies, Tildesley et al. (2003) demonstrated improve-
ments to immediate and delayed recall tasks in healthy
young humans at 1-h and 2.5-h following a single dose of
50 mL SL essential oil and at 2.5-h following 100 mL. Similar
mnemonic effects were subsequently confirmed following
25 mL and 50 mL of the same essential oil, along with improved
serial subtraction task performance and improved levels of
subjective alertness, calmness and contentment (Tildesley
et al., 2005). A similar pattern of mood effects was seen in
comparison with placebo following 300 mg and 600 mg of
encapsulated dried SO leaf, with reduced anxiety also evident
following the lower dose, with these mood effects abolished
by completion of a laboratory psychological stressor
(Kennedy et al., 2006). Recently, Scholey et al. (2008) also
confirmed the memory-improving effects of the two lowest of
four single doses (167, 333, 666, 1333mg) of an ethanolic
extract of SO administered to healthy elderly volunteers
(>65 years). The 333 mg dose also improved attention task
performance across all but one (2.5-h) of the four post-dose
assessments (1-, 2.5-, 4- and 6-h).
Whilst pro-cholinergic effects may well underlie the
improved mood and cognition-enhancing effects evident
in the above studies, it is notable that secondary
metabolites from sage extracts also exhibit anti-oxidant,
anti-inflammatory and oestrogenic properties (Perry et al.,
2003), all of which may be relevant to brain function both in
healthy populations and in sufferers of age-associated cognitive
decline and dementia (Kennedy and Scholey, 2006). Unlike
alkaloid phytochemicals, the terpenoid constituents of sage
extracts are also well tolerated and exhibit few unwanted side
effects at the levels likely to be consumed by humans.
The current, exploratory study assessed both the phyto-
chemical components and behavioural effects of an SL essen-
tial oil that has been found in a pilot investigation to have
particularly potent AChE inhibitory properties (IC
50
at
one-tenth the concentration of any previous extract), with a
view to informing the choice of treatment for a future chronic
dosage, randomized controlled trial. The study therefore con-
firmed the AChE inhibitory properties of the extract and
sought to address several further issues: the characteristics
of the extract in terms of components that might contribute
to its potent cholinesterase inhibiting properties; the extent to
which any cognition and mood effects of the extract are com-
parable with those seen previously with other, less potent sage
extracts; and whether further cognitive/mood effects could be
demonstrated during extended completion of mentally fatigu-
ing, cognitively demanding ‘executive’ tasks that should be
sensitive to cholinergic modulation.
The study therefore comprised an initial in vitro investiga-
tion. This was followed by a double-blind, placebo-
controlled, balanced cross-over study in healthy young
adults that assessed the cognitive/mood effects of a single
dose of the extract, in a quantity of essential oil previously
found to be optimal for improving memory performance
(50 mL; Tildesley et al., 2003, 2005). In order to provide
some comparison with previous studies investigating extracts
with considerably lower AChE inhibitory properties, the
assessment included a number of memory/attention tasks pre-
viously shown to be sensitive to cholinesterase-inhibiting
extracts, and these were augmented by an assessment of sub-
jective mood and cognition during 60 min of performing exec-
utive function/working memory and attention tasks.
Methods and materials
In vitro properties of the essential oil
Cholinesterase inhibition.
The method employed was
based on that developed by Savelev et al. (2004). The ability
of the SL essential oil to inhibit human AChE and BuChE
was assessed in vitro using the colorimetric method of Ellman
(1961), as described previously (Okello et al., 2008). The
essential oil was diluted in 53% ethanol prior to assay.
Serial dilutions were prepared to give final assay concentra-
tions of 0.1–0.001 mg/mL. A typical run consisted of 5 mLof
AChE or BuChE at final assay concentrations of 0.03 U/mL;
220 mL of 0.1 M phosphate buffer pH 8; 5 mL of DTNB
(prepared in 0.1 M phosphate buffer pH 7 with 0.12 M
sodium bicarbonate) at a final assay concentration of
0.3 mM; and 5 mL of test solution. The reactants were mixed
in 96-well plates, covered with the transparent plate sealers to
prevent loss of the volatile oil. The reactants were pre-incu-
bated at 30
C for 30 min. The reaction was initiated by adding
5 mL of substrate (1 mM ATChI or 5 mM BTChI final
Kennedy et al. 1089
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
assay concentrations). Change in absorbance kinetic mode
was measured at 405 nm at 30
C for 6 min (with shaking
every 20 s) using a Thermo Labsystems Multiskan Ascent
TM
plate reader with Ascent
TM
software. There were two controls
to monitor for non-enzymatic hydrolysis in the reaction mix-
ture; control 1 consisted of wells with no substrate (replace
substrate with buffer) and control 2 consisted of wells with no
essential oil (replace oil with 53% ethanol). Assays were per-
formed in triplicate, repeated 2.
Composition. Gas chromatography/Mass spectrometry
(GC/MS) analysis of the essential oil was performed on a
Hewlett-Packard 6890 GC with a split/splitless injector
(280
C) linked to a Hewlett-Packard 5973 mass selective
detector (electron voltage 70 eV, filament current 220 mA,
source temperature 230
C, quad temperature 150
C, multi-
plier voltage 2400 V, interface temperature 300
C). Data
acquisition was controlled by a HP Kayak XA PC chemsta-
tion computer in full scan mode (35–550 amu/s). A sample
(1 mL) in dichloromethane was injected by an HP6890 auto
sampler and the split opened after 1 min. Separation was
performed on CB8 (50 m, 0.25 mm ID, 0.25 (mdf)) capillary
column. The GC was temperature programmed from
40–300
Cat4
C/min and held at a final temperature for
20 min with helium as the carrier gas at a flow rate of 1 mL/
min, initial pressure 11.5 psi, split 60 mL/min.
Design
The study employed a randomized, double-blind, placebo-
controlled, balanced cross-over design.
Participants
In total, 36 young adult participants (male 10/female 26,
mean age 23.8 years, SD 4.38) took part in the study.
Participants self-reported that they were in good health and
that they were taking no medication (other than the contra-
ceptive pill), social drugs, or food supplements. Habitual smo-
kers were excluded from the study, as was anyone who was
pregnant (or seeking to become so). Prior to study days all
participants abstained from caffeine and alcohol for a mini-
mum of 12-h prior to the first testing session of the morning
and throughout the testing session. Participants were also
asked to have the same light breakfast (e.g. toast/cereal)
and lunch (e.g. sandwich) at the same time on each study
day, and to consume nothing else, with the exception of
water. Participants’ food intake on each study day was
recorded on a diary card.
The study received ethical approval from the Northumbria
University School of Psychology and Sport Sciences Ethics
Committee and was conducted according to the Declaration
of Helsinki (1964). All participants gave their informed con-
sent prior to their inclusion in the study.
Treatments
The treatments were supplied as identical capsules by
Pharmaton SA (Lugano) and were only identified by a code
when delivered to the site. A disinterested third party
arranged the allocation of treatments as per the counterbalan-
cing schedule.
Depending on the condition to which they were allocated
on each day, the participants received a single soft gel capsule
containing either 50 mL of SL essential oil plus olive oil, or a
placebo capsule containing olive oil.
Cognitive and mood measures
All tasks were delivered within the Computerized Mental
Performance Assessment System (COMPASS), a purpose-
designed software application for the flexible delivery of
randomly generated parallel versions of standard and novel
cognitive assessment tasks. With the exception of the paper
and pencil tasks (word recall), all responses were made using
the computer keyboard and mouse. In this case the assess-
ment comprised a selection of standard psychometric tasks
with stimuli chosen to possess good face validity in an ‘every-
day’ context. The elements of the cognitive assessment are
described below, and presented in Figure 1 in the order in
which they took place.
Memory and attention tasks
Word list presentation. Participants were presented with
lists of 15 nouns describing groceries and household items.
Words were presented for 1 s with a 1-s inter-stimulus
duration.
Picture presentation. Participants were presented with
15 pictures depicting man-made objects, buildings and
scenes with stimulus duration of 2 s and a 1-s inter-stimulus
duration.
Names–face presentation. Twelve faces were presented with
a first name and surname underneath for 2 s with an inter-
stimulus interval of 1 s.
Immediate word recall. Participants were given 60 s to write
down as many of the previously presented words as possible.
The outcome was the number of items correctly recalled.
Immediate name–face recognition. Participants were
re-presented with the faces that were shown earlier, along
with four possible first names and four possible surnames,
including the correct first name and surname that was origi-
nally associated with the face. The task was repeated at the
end of the test session. The outcomes were the number of
items correct (first name and surname) and speed of response
(ms).
Simple reaction time. The participant was instructed to
press a response button as quickly as possible every time a
single stimulus (upwards pointing arrow) was presented on
1090 Journal of Psychopharmacology 25(8)
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
the monitor. In total, 50 stimuli were presented with an inter-
stimulus interval that varied randomly between 1 and 3.5 s. The
outcome was the average reaction time (ms) to the stimuli.
Choice reaction time. The participant was instructed to
press either the left or right arrow button as quickly as pos-
sible in response to stimuli that comprised an arrow pointing
left or right. In total, 50 stimuli were presented with an inter-
stimulus interval that varied randomly between 1 and 3.5s.
The outcomes were the number correct and average reaction
time (ms) to the stimuli.
Telephone numbers working memory task. A nine-digit ‘tele-
phone number’ was presented on screen for 5 s. Participants
were instructed to remember the phone number for 10 s, after
which they could input the number via the linear number
keys on the keyboard. A total of eight ‘telephone numbers’
were presented, with the task outcome derived from a score
for each number incorporating correct digits and correct
order.
Distractor task – multi-tasking framework (Wetherell and
Sidgreaves, 2005). A 12-min distractor task was interposed
between the immediate and delayed elements of the memory
and attention assessment. Previous research has shown that per-
formance of the multi-tasking framework (MTF) engenders
increases in self-ratings of negative mood, anxiety and stress-
related physiological responses (Kennedy and Scholey, 2004;
Kennedy et al., 2006). Mood was therefore assessed with
Bond–Lader mood scales and the State Trait Anxiety
Inventory (STAI) state subscale (see ‘mood scales’ below) imme-
diately before and after the 12-min distractor task. A score rep-
resenting the change in mood during the MTF was then
calculated by subtracting the pre-task from the post-task mood
scores.
Delayed word recall. Participants were given 60 s to write
down as many of the previously presented words as possible.
The outcome was the number of items correctly recalled.
Delayed word recognition. Participants were presented with
the 15 original words from the list, plus 15 novel distractor
words. They responded yes/no as to whether each word was
in the original word list. The outcomes were number of words
correct and average speed of response (ms).
Delayed picture recognition task. Participants were pre-
sented with the 15 original pictures, plus 15 distractors.
They responded yes/no as to whether each picture was one
of those presented earlier. The task was scored for number
correct and average speed of response (ms).
Delayed names-face association task. Participants were
re-presented with the faces that were shown earlier, along
with four possible first names and four possible surnames,
including the correct first name and surname that was origi-
nally associated with the face. The outcomes were the number
of items correct (first name and surname) and average speed
of response (ms).
Bond-Lader mood scales
Bond-Lader mood scales
State-trait anxiety inventor
y
30
Minutes
60
Minutes
Cognitive Demand Battery
6× 10 minutes
Presentation of stimuli
Words/pictures/names-faces
Immediate word recall
Immediate name-face recognition
Simple reaction time
Choice reaction time
Telephone numbers
Distractor task (12 min)
Delayed word recall
Delayed word recognition
Delayed picture recognition
Delayed name-face recognition
Figure 1. Timelines of each cognitive/mood assessment. The pre-dose baseline assessment differed from the post-dose assessments (represented here)
in that it only included a single repetition of the tasks from the Cognitive Demand Battery.
Kennedy et al. 1091
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
Cognitive Demand Battery. The objective of the Cognitive
Demand Battery (CDB) is to assess the impact of treatment
on speed/accuracy and mental fatigue during continuous per-
formance of cognitively demanding tasks. The working
hypothesis underlying this approach is that any psychoactive
properties of a test substance are liable to be more readily
apparent during a period of intense cognitive demand and the
‘mental fatigue’ state elicited by this prolonged task
performance.
Participants complete the 10-min battery of tasks six times
in immediate succession (i.e. for a continuous period of
60 min). Application of this battery has been shown to reli-
ably increase self-ratings of ‘mental fatigue’ and to be sensi-
tive to a number of herbal and natural interventions
(Kennedy and Scholey, 2004; Kennedy et al., 2008; Reay
et al., 2005, 2006). The 10-min battery comprises the follow-
ing tasks.
Serial threes subtraction task (2 min). Computerized ver-
sions of the serial subtraction tasks were implemented using
tests of 2-min duration. Participants were required to count
backwards in threes from a given number as quickly and as
accurately as possible using the number keys to enter each
response. A random starting number between 800 and 999
was presented on the computer screen, which was cleared
by the entry of the first response. The task was scored for
number of correct responses and number of errors. In the
case of incorrect responses, subsequent responses were
scored as positive if they were scored as correct in relation
to the new number.
Serial sevens subtraction task (2 min). This was identical to
the serial threes task with the exception that it involved serial
subtraction of sevens.
Rapid Visual Information Processing task (RVIP – 5 min). The
participant was required to monitor a continuous series of
digits for targets of three consecutive odd or three consecutive
even digits. The digits were presented at the rate of 100/min,
and the participant responded to the detection of a target
string by pressing the ‘space bar’ as quickly as possible.
The task was continuous and lasted for 5 min, with eight
correct target strings being presented in each minute.
The task was scored for percentage of target strings correctly
detected, average reaction time for correct detections, and
number of false alarms.
‘Mental fatigue’ visual analogue scale. Participants rated
their current subjective ‘mental fatigue’ state by making a
mark on a 100 mm line with the end points labelled ‘not at
all’ (left-hand end) and ‘very much so’ (right-hand end).
Mood/well-being measures. Mood was assessed with
Bond–Lader mood scales (Bond and Lader, 1974) immedi-
ately prior to the cognitive assessment, and with
Bond–Lader mood scales and the STAI state subscale
before and after the MTF distractor task.
Bond–Lader mood scales. These scales have been utilized
in numerous pharmacological, psychopharmacological
and medical trials. These scales comprise a total of 16
100-mm lines anchored at either end by antonyms
(e.g. ‘alert–drowsy’, ‘calm–excited’). Participants indicate
their current subjective position between the antonyms on
the line. Outcomes comprise three factor analysis derived
scores: ‘Alertness’, ‘Calmness’ and ‘Contentment’.
STAI – ‘state’ subscale. The STAI state subscale
(Speilberger et al., 1969) was administered as a paper and
pencil questionnaire. The scale contains 20 items assessing
the presence (e.g. ‘I am tense’) and absence (e.g. ‘I feel at
ease’) of symptoms of anxiety at the present moment. Each
item is scored from 1–4 with a total scale score between 20
and 80. A lower score represents lower anxiety.
Procedure
Participants attended the Brain, Performance and Nutrition
Research Centre on three separate occasions at 8.30–9.00 am.
Testing took place in a suite of testing facilities with partici-
pants visually isolated from each other.
The first, introductory visit comprised obtaining of
informed consent; training on the cognitive and mood
measures; health screening; collection of demographic data,
and random allocation to treatment counterbalancing order.
Following the introductory visit participants attended the
laboratory on two further occasions 1 week apart.
Participants attended having consumed no caffeine or alco-
hol for at least 12-h. On arrival on both occasions, partici-
pants undertook an initial cognitive/mood assessment
comprising completion of the Bond–Lader mood scales, fol-
lowed by the immediate elements of the memory and atten-
tion assessment. The distractor task (MTF) was then
completed, with Bond–Lader mood scales and STAI com-
pleted before and after. This was followed by the delayed
elements of the memory and attention assessment.
Participants then made a single completion of the 10-min
CDB tasks (see Figure 1). Once this baseline assessment
was completed participants consumed their treatment for
that day. At 1-h and 4-h after consuming their treatment,
participants repeated the cognitive and mood assessment.
These assessments were identical to the pre-dose assessment,
with the exception that the full 60 min (i.e. six completions of
the 10 minutes of tasks) of the CDB were undertaken.
A standard light lunch was consumed immediately following
the 1-h post-dose assessment. The task running order and
timing for each of the post-dose assessments are shown
in Figure 1.
Statistics
Data were baseline adjusted (calculated against pre-dose)
prior to analysis.
1092 Journal of Psychopharmacology 25(8)
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
The distribution of data for each measure was checked for
symmetry and equivalence of variance (standard deviations)
between conditions, with an Anderson–Darling test of nor-
mality carried out for those measures that departed from
acceptable criteria.
Statistical analysis for the memory/attention and mood
measures was by repeated measures analysis of variance
(ANOVA) (condition treatment order post-dose assess-
ment (1-h, 4-h)), with repetition included as a further factor
for data from the three mood scale assessments. Planned com-
parisons were made between data from each treatment during
each of the post-dose assessments (1-h, 4-h) utilizing
Bonferonni corrected t-tests calculated using the Mean
Squares Error from the ANOVA (Keppel, 1991). Planned
comparisons are only reported for measures that showed a
significant treatment effect or treatment assessment effect
on the ANOVA.
Analysis of the CDB data was by an initial repeated mea-
sures omnibus ANOVA (condition treatment order post-
dose assessment (1-h, 4-h) repetition (1–6)). Planned
comparisons were then carried out between treatments for
each repetition of the battery during each assessment as
described above, with a Bonferonni correction applied for
multiplicity during each assessment. To further protect against
Type I errors only the planned comparisons for those mea-
sures that showed a significant treatment-related effect on
a reduced model ANOVA (reported in results) are presented.
Power. Previous studies of sage extracts have demonstrated
effect sizes ranging between Cohen’s (Cohen, 1992) definition
of medium to, in some cases, large effect sizes (e.g. Tildesley
et al., 2003). Prior to this study the necessary sample size was
therefore calculated assuming a medium effect size of f ¼ 0.25
using G*Power (Erdfelder et al., 1996). This calculation sug-
gested that for the proposed within-subjects design, with two
post-dose repetitions of the tasks, 90% power at a ¼ 0.05
would be achieved with a sample size of 36.
Results
In vitro assays
Cholinesterase inhibition and essential oil
components.
The SL essential oil was a highly potent selec-
tive inhibitor of AChE, with an IC
50
value of 0.003 mg/mL
(3 mg/mL). The essential oil was therefore 10 times more
active than the most potent SO/SL extract assessed previ-
ously, which had a maximum IC
50
concentration of
0.03 mg/mL (Savelev et al., 2004). BuChE was weakly inhib-
ited in this case, only achieving 22% inhibition at the highest
concentration utilized (0.5 mg/mL). The inhibition curve for
AChE, GC/MS chromatogram of the essential oil and the
constituents of the essential oil are presented in Figure 2.
Behavioural assessment
Memory and attention.
A number of memory and atten-
tion measures evinced significant differences between the
treatment groups.
Simple reaction time. There was a main effect of treatment
on the Simple Reaction Time task (F(1,34) ¼ 5.15, p < 0.05)
with a trend towards an interaction between treatment and
session (F(1,34) ¼ 3.91, p < 0.1). Planned comparisons of data
from the individual assessments revealed that participants
performed significantly faster in the essential oil condition
during the 1-h post-dose assessment (t(34) ¼ 3.99, p < 0.01).
Delayed word recall. Delayed word recall evinced a main
effect of treatment (F(1,34) ¼ 5.49, p < 0.05) with a strong
trend towards an interaction between treatment and session
(F(1,34) ¼ 3.87, p < 0.1). The planned comparisons showed
that following essential oil, participants outperformed pla-
cebo during both the 1-h (t(34) ¼ 5.3, p < 0.01) and 4-h
(t(34) ¼ 2.5, p < 0.05) assessments.
Word recognition task. There was also a significant main
effect on the accuracy of performing (number correct) the
Word Recognition task (F(1,34) ¼ 8.7, p < 0.01), with a
trend towards an interaction between treatment and assess-
ment (F(1,34) ¼ 2.88, p < 0.1). Planned comparisons showed
that the essential oil outperformed placebo during the 1-h
assessment (t (34) ¼ 3.28, p < 0.01), with a trend towards the
same effect at 4-h post-dose (t (34) ¼ 1.83, p < 0.1).
Picture recognition task. Similarly,
following essential oil
participants performed better than following placebo on the
accuracy (number correct) of the Picture Recognition Task
(F (1,34) ¼ 5.65, p < 0.05). The planned comparisons revealed
that this effect was apparent at both 1-h (t (34) ¼ 5.38,
p < 0.01) and 4-h (t (34) ¼ 3.3, p < 0.01).
Mood measures. Data from the three Bond–Lader mood
assessments (completed at the start of memory/attention
assessment and before and after the MTF) showed that
there was a significant interaction between treatment and
assessment on the ‘alert’ factor (F(1,34 ¼ 10.16, p < 0.01).
The planned comparisons showed that whereas the treatment
groups did not differ at the 1-h post-dose assessment, they
rated themselves as significantly more ‘alert’ following essen-
tial oil during the 4-h post-dose assessment (t (34) ¼ 2.85,
p < 0.01). Following an Anderson–Darling test of normality,
the data for this measure were normalized with a square root
transformation, and the resulting ANOVA showed the same
treatment assessment interaction as had been evident when
analysing the raw data (F(1,34) ¼ 5.71, p < 0.05).
There was no significant effect on the STAI state subscale.
There was no treatment-related modulation of the stress
response during the MTF in terms of changes in the scores
on the Bond–Lader mood scales and STAI (data not shown).
Measures from the memory and attention tasks and mood
measures that evinced significant results are presented in
Figure 3. Mean (plus SEM) baseline data and ‘change from
baseline’ data from the post-dose assessments for the
memory, attention and mood measures are presented
in Table 1.
Kennedy et al. 1093
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
Cognitive demand battery
Serial 3s. The only significant differences in performance of
the tasks within the CDB were seen on the total number of
Serial 3s subtractions performed. The ANOVA evinced a sig-
nificant interaction between treatment and assessment
(F(1,748) ¼ 6.13, p < 0.05). Reference to the planned compari-
sons showed that this was manifested as significantly improved
performance in the essential oil condition during the third
(t(748) ¼ 2.58, p ¼ 0.01) and sixth (t(748) ¼ 2.76, p < 0.01) rep-
etitions of the battery during the 4-h post-dose assessment,
with a strong trend towards the same effect during the
second repetition (t(748) ¼ 1.93, p < 0.1). There was no modu-
lation during the earlier (1-h post-dose) assessment, and no
significant differences in the number of errors committed.
Mental fatigue. There was also a trend towards a
main effect of treatment (F(1,748) ¼ 3.67, p < 0.1) and a sig-
nificant interaction between treatment and assessment
(F(1, 748) ¼ 5.9, p < 0.05) on subjective ratings of mental fati-
gue during the 60 min of task performance. Planned compar-
isons of data from both assessments showed that during the
1-h post-dose assessment participants rated themselves as less
fatigued following the fourth (t(748) ¼ 2.77, p < 0.05), fifth
(t(748) ¼ 2.86, p < 0.05) and sixth (t(748) ¼ 3.01, p < 0.05)
repetitions of the battery following the essential oil.
Similarly, at the 4-h post-dose assessment participants
rated themselves as less fatigued following the second
(t(748) ¼ 2.73, p < 0.05), third (t(748) ¼ 4.79, p < 0.01),
fourth (
t(748) ¼ 4.33, p < 0.01),
fifth (t(748) ¼ 4.28, p < 0.01)
and sixth (t(748) ¼ 4.56, p < 0.01) repetitions of the battery
while in the essential oil condition.
Mean (plus SEM) change from baseline data from the
Serial 3s task (total subtractions) and mental fatigue ratings
during each assessment are shown in Figure 4.
Discussion
The SL essential oil under investigation was largely composed
of monoterpenoids and exhibited considerably greater inhibi-
tion of human erythrocyte AChE than the most potent SO/SL
extract previously investigated (Savelev et al., 2004). In terms
of behaviour, administration of a single dose of 50 mLof
essential oil resulted in improved cognitive performance.
This was seen in an attenuation of the decreases seen in atten-
tion and memory task performance, most notably during the
1-h post-dose assessment, with this effect diminishing during
the 4-h post-dose assessment. Performance of the CDB, in
terms of the Serial 3s subtraction task, was also improved
but only at the later (4-h) assessment. This pattern was also
seen in the attenuation of mental fatigue during task perfor-
mance, which was more pronounced during the 4-h post-dose
Acetylcholinesterase
–3 –2 –1 0
% inhibition
30
40
50
60
70
80
90
Concentration (log[mg/mL])
COMPOUND RETENTION
TIME
(MIN)
PERCENTAGE
COMPOSITION
a-Pinene 9:56.04 5.6
Camphene 10:18.99 5.7
Sabinene 10:55.21 0.3
b-Pinene 11:00.31 5.1
b-Myrcene 11:17.65 0.6
p-Cymene 12:05.09 0.5
Limonene 12:11.21 4.2
1,8-Cineole 12:14.78 36.4
Linalool 13:37.41 0.4
Camphor 14:36.07 37.0
endo-Borneol 15:00.04 1.2
a-Terpineol 15:24.52 Tr
Linalyl acetate 16:25.22 0.7
Bornyl acetate 17:02.97 0.7
Unidentified 17:04.50 1.2
Unidentified 19:16.10 0.1
Unidentified 19:46.19
0.1
(A)
(B)
(C)
Figure 2. (A) Inhibition of human erythrocyte acetylcholinesterase. (B) GC/MS total ion chromatogram of essential oil, with chromatography
performed on a 50 m 0.25 mm i.d. 0.25 mm CB8–MS column. (C) Compounds in the Salvia lavandulaefolia essential oil (expressed as percentage
composition calculated from areas of detected peaks).
1094 Journal of Psychopharmacology 25(8)
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
assessment, and subjective ratings of alertness, which were
only significantly increased during this later assessment.
The improvements seen in cognitive performance replicate
the improved secondary memory task performance seen pre-
viously in healthy younger and older adults (Scholey et al.,
2008; Tildesley et al., 2003, 2005), and the improved attention
task performance seen in an elderly cohort (Scholey et al.,
2008) following single doses of sage extracts with less potent
cholinesterase inhibiting properties. Increased ‘alertness’ has
also been demonstrated previously (Kennedy et al., 2006;
Tildesley et al., 2005). The effects of sage extracts on extended
performance of mentally demanding tasks, and the resultant
increase in mental fatigue, has not previously been
investigated.
Naturally, these results might have been as a consequence
of a statistical anomaly. In any experiment it is necessary to
strike a balance between the possibility of ‘false positive’ Type
I errors and the obscuring of genuine findings by being overly
conservative and thereby increasing the likelihood of Type II
errors (Keppel, 1991). Given that this was an exploratory
study, we chose to control for the number of comparisons
per task, but not the error rate across the entire experiment.
Msecs
Number correct
Number correct
20
Simple reaction time Delayed word recall
Word recognition
Picture recognition
10
0
1
0
–1
–2
–10
–4
–3
–2
–1
0
1
Salvia
Placebo
–3
Number correct
1
0
–1
–2
–3
Factor score
3
Alert
2
1
0
–1
–2
–3
1 hour post-dose 4 hours post-dose 1 hour post-dose 4 hours post-dose
1 hour post-dose 4 hours post-dose
1 hour post-dose 4 hours post-dose
1 hour post-dose 4 hours post-dose
Salvia
Placebo
Figure 3. Mean (plus SEM) change from baseline data for the 1-h and 4-h post-treatment assessments on the memory, attention and mood measures
that showed significant effects on the ANOVA. (*p < 0.05, **p < 0.01 from the planned comparisons, made using t-tests with a Bonferroni correction).
Kennedy et al. 1095
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
This raises the possibility of treatment effects arising by
chance. However, of the 25 mood and cognitive performance
measures reported here we found improvements on seven
measures (1.25 Type I errors would be predicted by chance
at a ¼ 0.05). With regards to these effects they were all in a
positive direction, and the findings of improved memory,
attention and alertness replicate the findings in previous sim-
ilar studies (Kennedy et al., 2006; Scholey et al., 2008;
Tildesley et al., 2003, 2005). It therefore seems likely that
the results here represent genuine treatment-related benefits.
The essential oil here was shown to be a considerably more
potent AChE inhibitor than previous extracts, with an IC
50
which represented a concentration a tenth of that of the most
potent extract investigated previously. Increased arousal/
alertness, and improved performance on attention, memory
and ‘executive’ tasks can all be accommodated within a
simple ‘cholinergic’ explanation of the results seen here
(Sarter et al., 2003). However, the observation that beneficial
effects decrease with increased doses of essential oil (Tildesley
et al., 2003), and similar behavioural improvement from
extracts with varying anticholinesterase properties across
studies would tend to suggest that cholinesterase inhibition
only has to exceed a certain threshold to modulate brain func-
tion (Kennedy et al., 2006; Scholey et al., 2008; Tildesley
et al., 2003, 2005).
In terms of the essential oil itself, the constituents were
almost exclusively monoterpenoids, including camphor
(37%), 1,8-cineole (36.4%), camphene, a-pinene, b-pinene,
limonene and endo-borneol. While high levels of camphor
have been identified across SL essential oils (Langa et al.,
2009) the levels of 1,8-cineole were higher than those reported
previously (Langa et al., 2009; Perry et al., 2003). Research
suggests that 1,8-cineole is the most potent AChE inhibitor
among the individual constituent chemicals of SL, and acts in
this respect both synergistically with other constituents and
antagonistically with camphor (Perry et al., 2003; Savelev
et al., 2003, 2004). It may well be the case that the unusually
high ratio of 1,8-cineole in comparison with other monoter-
penes, and camphor in particular, underlies the increased
AChE inhibition seen here.
Table 1. Mean (plus SEM, in italics) baseline scores and change from baseline scores from the 1-h and 4-h assessments for the memory, attention and
mood measures. Data for the mood measures are averages across the three repetitions carried out within each assessment. Data shown in bold are the
measures that showed significant modulation
Post-dose change from baseline
Task Treatment Pre-dose Baseline 1-h 4-h
Telephone number (number) Salvia 3.19 0.31 0.00 0.36 0.25 0.37
Placebo 3.19 0.39 0.19 0.36 0.36 0.28
Simple RT (ms) Salvia 288.47 5.05 7.18 4.43 11.70 5.49
Placebo 281.29 4.63 13.19 4.40 17.83 4.10
Choice RT (% accuracy) Salvia 96.28 0.56 1.00 0.51 2.17 0.64
Placebo 95.28 0.63 0.67 0.52 0.89 0.61
Choice RT (ms) Salvia 405.82 9.44 4.75 5.87 1.03 5.67
Placebo 401.07 7.16 0.56 3.58 5.26 6.23
Imm name/face (number correct) Salvia 6.00 0.41 0.56 0.42 0.39 0.40
Placebo 5.44 0.41 0.06 0.45 0.36 0.33
Del name/face (number correct) Salvia 5.72 0.39 0.47 0.36 0.89 0.41
Placebo 5.25 0.52 0.06 0.47 0.56 0.37
Immediate word Recall
(number) Salvia 8.22 0.37 0.15 0.37 0.94 0.41
Placebo 8.38 0.36 0.39 0.33 0.78 0.39
Delayed word Recall (number) Salvia 7.75 0.37 0.08 0.42 2.33 0.46
Placebo 7.83 0.38 1.71 0.36 3.18 0.49
Word Recognition (number correct) Salvia 25.83 0.48 0.94 0.44 1.25 0.47
Placebo 26.78 0.42 1.36 0.46 2.25 0.49
Word Recognition RT (ms) Salvia 1226.11 44.18 23.17 36.42 26.25 31.43
Placebo 1202.94 40.17 7.14 37.41 36.08 34.37
Picture Recognition (number correct) Salvia 23.64 0.57 0.67 0.60 0.64 0.57
Placebo 24.31 0.57 1.69 0.57 2.42 0.56
Picture Recognition RT (ms) Salvia 1386.06 53.18 45.00 35.47 58.28 33.50
Placebo 1341.06 40.82 44.47 30.58 27.47 32.20
Alert
1
(factor score) Salvia 54.68 2.36 0.17 1.27 1.39 1.37
Placebo 57.17 1.91 0.76 1.31 1.47 1.55
Content
1
(factor score) Salvia 61.80 2.11 0.13 0.92 0.63 0.81
Placebo 63.90 1.75 0.68 1.00 0.86 1.04
Calm
1
(factor score) Salvia 58.62 2.06 0.54 1.15 2.51 1.58
Placebo 59.68 1.93 3.31 1.20 3.68 1.97
1
Data averaged across the three repetitions per assessment.
1096 Journal of Psychopharmacology 25(8)
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
Given the potentially pivotal role played here by
1,8-cineole, its secondary metabolite roles as an individual
monoterpenoid phytochemical are of some interest.
Evidence suggests that one role that 1,8-cineole plays is as a
feeding deterrent which is expressed within the plant as a
consequence of damage by the herbivore (Vuorinen et al.,
2004). Similar synergies as those seen with regards to AChE
inhibition exist in terms of these feeding deterrence properties,
with 1,8-cineole being the most potent constituent of essential
oils in terms of deterrence, but with the full monoterpenoid
extract exhibiting properties in excess of the constituent
chemicals (Gonza
´
lez-Coloma et al., 2006). 1,8-cineole also
exhibits direct toxicity in larval and adult beetles (Kordali
et al., 2007; Stamopoulos et al., 2007), but not as a conse-
quence of AChE inhibition (Picollo et al., 2008). In general,
monoterpenoid combinations that include 1,8-cineole have
been shown to be toxic to some taxa, for instance ectopara-
sitic mites and house flies (Damiani et al., 2009; Palacios
et al., 2009) but harmless to others, for instance bees
(Damiani et al., 2009). It may be relevant that 1,8-cineole is
also a fragrant attractant for insect pollination (Raguso et al.,
2006) and may well function as one constituent of the
bouquet of volatile components released by a plant during
herbivorous attack in order to attract the natural insect pred-
ators of the attacking herbivores (Rohloff and Bones, 2005).
These non-lethal and benign roles of monoterpenoids in
plant–insect interactions are particularly pertinent here, as
acetylcholine plays a major role in insect behaviour and
memory. Both nicotinic and muscarinic receptors are
expressed within insect nervous systems, and cholinergic ago-
nists and antagonists have been shown to up-regulate and
down-regulate behavioural parameters, respectively (Dacher
and Gauthier, 2008; Ismail et al., 2008). While insects delib-
erately attracted to a plant for pollination or to function as
agents of indirect defence are unlikely to directly consume
plant tissue, and therefore may avoid elements of toxicity,
emission of a volatile that had severe negative effects on the
cholinergic systems of symbiotes would be disadvantageous.
It therefore seems possible that the interactions of monoter-
penoids, including 1,8-cineole, with the cholinergic nervous
system of symbiotic insects must be at least neutral in
nature and may be potentially beneficial to the insect. While
the possibility has yet to be explored, it may well be that
AChE inhibition by secondary metabolites, other than toxic
alkaloids, can be advantageous in behavioural terms to
insects that have a symbiotic relationship with the emitting
plant. Similar cholinergic effects may then extend to other
consuming animals, including humans, due to the similarities
in CNS neurotransmitter systems and functions between
species.
Of course, it is also possible that an unmeasured, non-
cholinergic parameter is responsible for the behavioural
effects evinced here. The results showed a biphasic pattern,
with the simple memory and attention tasks being preferen-
tially improved at the earlier (1-h) assessment, and the more
difficult serial subtraction task, mental fatigue and alertness
being preferentially improved at the later (4-h) assessment.
This pattern could be predicated on an interaction between
the treatment and fatigue/alertness, with participants being
generally less fatigued and more alert, and therefore less
susceptible to improvement in both of these parameters
(and therefore difficult tasks), during the first assessment.
The pattern could also be as a consequence of the differing
absorption profiles of the constituent monoterpenoids. For
instance, whilst there are few data on the absorption of com-
plex mixtures of monoterpenoids, the single constituent
1,8-cineole has been shown to peak in the plasma of possums
at approximately 60 min post-dose (McLean et al., 2008),
whereas limonene peaks in human plasma at 2.5-h post-
dose (Wang et al., 2007). The effects here may therefore be
underpinned by more than one mechanism of action with
differing time profiles of activity.
In the current study we assessed cholinesterase inhibition
by the extract for several specific reasons: previous evidence
shows that sage extracts generally possess this property; pro-
cholinergic treatments are currently prescribed for several
dementias; and this property is potentially relevant to cogni-
tive function, and declining cognitive function as a conse-
quence of both age and dementia. However, we have not
explored other potential mechanisms that might be equally
relevant. The possibility of these extracts possessing further,
complementary, properties requires further attention.
Plant-derived alkaloids, by function and chemical nature,
are toxic to mammals. However, taken in appropriate doses
this broad class of phytochemicals has provided a rich source
Salvia
Placebo
Mental fatigue
Mental fatigue (mm)
Total number of subtractions
40
30
20
10
0
–2
0
2
4
6
8
Serial 3s
123456
123456
123456
Repetitions
1 hour post-dose
4 hours post-dose
123456
Figure 4. Mean (plus SEM) change from baseline data during each rep-
etition at the 1-h and 4-h post-treatment assessments on the measures
from the Cognitive Demand Battery that reached significance on the
ANOVA (assessment treatment in both cases). (t, p < 0.1, *p < 0.05,
**p < 0.01 from the planned comparisons, made using t-tests with a
Bonferroni correction).
Kennedy et al. 1097
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
of drug discovery for medicines, and constitute the majority
of our social drugs (Samuelsson, 2004). Of particular rele-
vance, a number of current treatments for Alzheimer’s disease
are derived from alkaloid secondary metabolites that perform
the function for the plant of interfering with the cholinergic
system of herbivores. These include galantamine, huperzine,
physostigmine and rivastigmine (Mukherjee et al., 2007). In
the UK the National Institute for Clinical Excellence (NICE)
recently recommended against the existing, prescribed cholin-
esterase inhibiting treatments, except for patients suffering
moderate levels of dementia, on the basis of modest efficacy,
high levels of side effects and high cost (NICE, 2009).
Terpenoids, on the other hand, perform a much wider
range of functions for the plant, which can include the sym-
biotic attraction of animals, all of which have cholinergic
components to their CNS. Therefore they exhibit a broad
spectrum of toxicity for mammals, from highly toxic to
being entirely edible. In this respect, sage extracts have a his-
tory spanning several millennia as medicinal treatments for
cognitive deterioration and are associated with very few side
effects (Kennedy and Scholey, 2006), as are a number of other
putatively psychoactive herbal extracts with terpenoid active
constituents, including Ginkgo biolba, Valeriana officinalis and
Panax ginseng (Kennedy, 2009). Unfortunately, little method-
ologically adequate research has been directed towards these
more complex plant extracts, and the question of efficacy in
terms of brain function is still an open question. This com-
parative lack of research interest is partly due to current gaps
in our understanding of the relative contributions of the
complex constituents of plant extracts to any efficacy, and
an inability to adequately standardize the levels of the appro-
priate phytochemicals in any final product. The results seen
here, both in terms of potent in vitro AChE inhibition and
enhancement of mood and cognitive performance, certainly
argue against the suggestion that the structural diversity and
relatively weak anticholinesterase activity of terpenoids
render them unlikely to constitute treatments for cognitive
disorders (Houghton and Seth, 2003). However, they do sug-
gest that more research should be directed towards under-
standing the relative contributions and synergies within
complex chemical cocktails in order to move towards extracts
with more potent medicinal properties. Certainly the overall
and comparative levels of active phytochemicals in plant
material depend on a wide range of parameters that are capa-
ble of manipulation, including climate, soil composition, light
levels and time and season of harvest (Santos-Gomes et al.,
2002), geographic location (Putievsky et al., 1986; Santos-
Gomes et al., 2002), habitat (Ben Farhat et al., 2009), salinity
(Ben Taarit et al., 2009), exposure to phytophages (Dicke
et al., 2009; Vuorinen et al., 2004), pathogens, and ultraviolet
light (Langcake and Pryce, 1976), and hydration status
(Bettaieb et al., 2009). Given the sophistication of contempo-
rary agricultural techniques, it would seem feasible to grow
well-standardized plants that benefit from augmented and
constrained levels of beneficial and detrimental components,
respectively, while retaining the positive synergistic properties
of whole extracts. As a single example, increasing the salinity
of a hydroponic medium has been shown to preferentially
augment the concentration of 1,8-cineole, but not camphor,
within SO essential oil (Ben Taarit et al., 2009), and such
growing conditions may provide replicable extracts with the
high AChE inhibitory properties seen here. However, the pos-
sibility of tailor-made natural extracts rich in the most advan-
tageous constituents requires a fuller understanding of the
interaction and synergy between active components, their
secondary metabolite roles in the plant, and their physiolog-
ical activity and CNS/behavioural effects in mammals.
In conclusion, the extract under investigation was a com-
paratively potent AChE inhibitor that improved cognitive
performance and mood in healthy young humans following
a single dose. The currently prescribed cholinesterase inhibit-
ing treatments for Alzheimer’s disease, a number of which are
based on alkaloid phytochemicals, have been shown to have
limited efficacy and, at more effective doses, engender high
levels of negative side effects. Any potential treatment for
both the symptoms of dementia and the natural decrements
in cognitive performance that come with age, that is well
tolerated due to the chemical structure of its active compo-
nents, deserves further investigation. It is intended that an
extract with a similar monoterpenoid profile and cholinergic
properties as that investigated here will be taken forward into
a chronic study assessing cognitive and mood effects in
the elderly.
Funding
The research described herein was supported by a grant to DK from
Pharmaton SA (Lugano), who also supplied the encapsulated essen-
tial oil and placebos.
References
Ben Farhat M, Jordan MJ, Chaouech-Hamada R, Landoulsi A and
Sotomayor JA (2009) Variations in essential oil, phenolic com-
pounds, and antioxidant activity of Tunisian cultivated Salvia
officinalis L. J Agric Food Chem 57: 10349–10356.
Ben Taarit M, Msaada K, Hosni K, Hammami M, Kchouk ME and
Marzouk B (2009) Plant growth, essential oil yield and composi-
tion of sage (Salvia officinalis L.) fruits cultivated under salt stress
conditions. Ind Crops Prod 30: 333–337.
Bettaieb I, Zakhama N, Wannes WA, Kchouk ME and Marzouk B
(2009) Water deficit effects on Salvia officinalis fatty acids and
essential oils composition. Sci Hortic (Amsterdam) 120: 271–275.
Bond A and Lader M (1974) Use of analog scales in rating subjective
feelings. Br J Med Psychol 47: 211–218.
Cohen J (1992) A power primer. Psychol Bull 112: 155–159.
Dacher M and Gauthier M (2008) Involvement of NO-synthase and
nicotinic receptors in learning in the honey bee. Physiol Behav 95:
200–207.
Damiani N, Gende LB, Bailac P, Marcangeli JA and Eguaras MJ
(2009) Acaricidal and insecticidal activity of essential oils on
Varroa destructor (Acari: Varroidae) and Apis mellifera
(Hymenoptera: Apidae). Parasitol Res 106: 145–152.
Dicke M, van Loon JJA and Soler R (2009) Chemical complexity of
volatiles from plants induced by multiple attack. Nat Chem Biol 5:
317–324.
Eidi M, Eidi A and Bahar M (2006) Effects of Salvia officinalis L.
(sage) leaves on memory retention and its interaction with the
cholinergic system in rats. Nutrition 22: 321–326.
Ellman G (1961) A new and rapid colorimetric determination of
acetylcholinesterase activity. Biochem Pharmacol 7: 88.
Erdfelder E, Faul F and Buchner A (1996) GPOWER: A general
power analysis program. Behav Res Methods Instruments
Comput 28: 1–11.
1098 Journal of Psychopharmacology 25(8)
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
Gonza
´
lez-Coloma A, Martı
´
n-Benito D, Mohamed N, Garcı
´
a-Vallejo
MC and Soria AC (2006) Antifeedant effects and chemical com-
position of essential oils from different populations of Lavandula
luisieri L. Biochem Syst Ecol 34: 609–616.
Hagen EH, Sullivan RJ, Schmidt R, Morris G, Kempter R and
Hammerstein P (2009) Ecology and neurobiology of toxin avoid-
ance and the paradox of drug reward. Neuroscience 160: 69–84.
Harborne JB (1993) Introduction to Ecological Biochemistry , Fourth
edn. London: Elsevier.
Houghton PJ and Seth P (2003) Plants and the central nervous
system. Pharmacol Biochem Behav 75: 497–499.
Ismail N, Christine S, Robinson GE, Fahrbach SE (2008) Pilocarpine
improves recognition of nestmates in young honey bees.
Neuroscience Letters 439: 178–181.
Kennedy D (2009) Herbs Affecting the Central Nervous System.
In: Mukherjee P and Houghton P (eds) Evaluation of Herbal
Medicinal Products – Perspectives of Quality, Safety and
Efficacy. London: Pharmaceutical Press, Royal Pharmaceutical
Society of Great Britain.
Kennedy DO and Scholey AB (2004) A glucose-caffeine ‘energy
drink’ ameliorates subjective and performance deficits during
prolonged cognitive demand. Appetite 42: 331–333.
Kennedy DO and Scholey AB (2006) The psychopharmacology of
European herbs with cognition-enhancing properties. Curr Pharm
Des 12: 4613–4623.
Kennedy DO, Haskell CF, Robertson B, et al. (2008) Improved cog-
nitive performance and mental fatigue following a multi-vitamin
and mineral supplement with added guarana (Paullinia cupana).
Appetite 50: 506–513.
Kennedy DO, Pace S, Haskell C, Okello EJ, Milne A and Scholey AB
(2006) Effects of cholinesterase inhibiting sage (Salvia officinalis)
on mood, anxiety and performance on a psychological stressor
battery. Neuropsychopharmacology 31: 845–852.
Keppel G (1991) Design and Analysis. New Jersey: Prentice Hall.
Kordali S, Kesdek M and Cakir A (2007) Toxicity of monoterpenes
against larvae and adults of Colorado potato beetle, Leptinotarsa
decemlineata Say (Coleoptera: Chrysomelidae). Ind Crops Prod
26: 278–297.
Langa E, Porta GD, Palavra AMF, Urieta JS and Mainar AM (2009)
Supercritical fluid extraction of Spanish sage essential oil:
Optimization of the process parameters and modelling.
J Supercrit Fluids 49: 174–181.
Langcake P and Pryce RJ (1976) The production of resveratrol by
Vitis vinifera and other members of the Vitaceae as a response to
infection or injury. Physiol Plant Pathol 9: 77–86.
McLean S, Brandon S, Boyle RR and Wiggins NL (2008)
Development of tolerance to the dietary plant secondary metab-
olite 1,8-cineole by the brushtail possum (Trichosurus vulpecula).
J Chem Ecol 34: 672–680.
Macias FA, Galindo JLG and Galindo JCG (2007) Evolution and
current status of ecological phytochemistry. Phytochemistry 68:
2917–2936.
Maffei ME, Mitho
¨
fer A and Boland W (2007) Insects feeding on
plants: Rapid signals and responses preceding the induction of
phytochemical release. Phytochemistry 68: 2946–2959.
Mukherjee PK, Kumar V, Mal M and Houghton PJ (2007)
Acetylcholinesterase inhibitors from plants. Phytomedicine 14:
289–300.
NICE (2009) Technology appraisal guidance 111. Donepezil, galanta-
mine, rivastigmine (review) and memantine for the treatment of
Alzheimer’s disease (amended). UK: NICE.
Okello E, Dimaki C, Howes M-JR, Houghton PJ and Perry EK
(2008) In vitro
inhibition of human acetyl- and
butyryl-cholinesterase
by Narcissus poeticus L. (Amaryllidaceae)
flower absolute. Int J Essent Oil Ther 2: 105–110.
Palacios SM, Bertoni A, Rossi Y, Santander R and Urzua A (2009)
Insecticidal activity of essential oils from native medicinal plants
of Central Argentina against the house fly, Musca domestica (L.).
Parasitol Res 106: 207–212.
Perry NSL, Bollen C, Perry EK and Ballard C (2003) Salvia for
dementia therapy: review of pharmacological activity and pilot
tolerability clinical trial. Pharmacol Biochem Behav 75: 651–659.
Perry N, Court G, Bidet N, Court J and Perry E (1996) European
herbs with cholinergic activities: Potential in dementia therapy.
Int J Geriatr Psychiatry 11: 1063–1069.
Perry NSL, Houghton PJ, Jenner P, Keith A and Perry EK (2002)
Salvia lavandulaefolia essential oil inhibits cholinesterase in vivo.
Phytomedicine 9: 48–51.
Perry NSL, Houghton PJ, Sampson J, et al. (2001) In-vitro activity of
S-lavandulaefolia (Spanish sage) relevant to treatment of
Alzheimer’s disease. J Pharm Pharmacol 53: 1347–1356.
Perry NSL, Houghton PJ, Theobald A, Jenner P and Perry EK
(2000) In-vitro inhibition of human erythrocyte acetylcholinester-
ase by Salvia lavandulaefolia essential oil and constituent ter-
penes. J Pharm Pharmacol 52: 895–902.
Picollo MI, Toloza AC, Mougabure Cueto G, Zygadlo J and Zerba E
(2008) Anticholinesterase and pediculicidal activities of monoter-
penoids. Fitoterapia 79: 271–278.
Putievsky E, Ravid U and Dudai N (1986) The influence of season
and harvest frequency on essential oil and herbal yields from a
pure clone of sage (salvia officinalis) grown under cultivated con-
ditions. J Nat Prod 49: 326–329.
Raguso RA, Schlumpberger BO, Kaczorowski RL and Holtsford TP
(2006) Phylogenetic fragrance patterns in Nicotiana sections
Alatae and Suaveolentes. Phytochemistry 67: 1931–1942.
Reay JL, Kennedy DO and Scholey AB (2005) Single doses of Panax
ginseng (G115) reduce blood glucose levels and improve cognitive
performance during sustained mental activity. J Psychopharmacol
19: 357–365.
Reay JL, Kennedy DO and Scholey AB (2006) Effects of Panax gin-
seng, consumed with and without glucose, on blood glucose levels
and cognitive performance during sustained ‘mentally demanding’
tasks. J Psychopharmacol 20: 771–781.
Rohloff J and Bones A (2005) Volatile profiling of Arabidopsis thali-
ana – Putative olfactory compounds in plant communication.
Phytochemistry 66: 1941–1955.
Samuelsson G (2004) Drugs of Natural Origin: a Textbook of
Pharmacognosy. Stockholm: 5th Swedish Pharmaceutical Press.
Santos-Gomes PC, Seabra RM, Andrade PB and Fernandes-Ferreira
M (2002) Phenolic antioxidant compounds produced by in vitro
shoots of sage (Salvia officinalis L.). Plant Sci 162: 981–987.
Sarter M, Bruno JP and Givens B (2003) Attentional functions of
cortical cholinergic inputs: What does it mean for learning and
memory? Neurobiol Learn Mem 80: 245–256.
Savelev SU, Okello EJ and Perry EK (2004) Butyryl- and acetyl-
cholinesterase inhibitory activities in essential oils of Salvia spe-
cies and their constituents. Phytother Res 18: 315–324.
Savelev S, Okello E, Perry NSL, Wilkins RM and Perry EK (2003)
Synergistic and antagonistic interactions of anticholinesterase
terpenoids in Salvia lavandulaefolia essential oil. Pharmacol
Biochem Behav 75: 661–668.
Scholey AB, Tildesley NTJ, Ballard CG, et al. (2008) An extract of Salvia
(sage) with anticholinesterase properties improves memory and atten-
tion in healthy older volunteers. Psychopharmaco logy 198: 127–139.
Speilberger C, Gorsuch R and Lushene R (1969) The State Trait
Anxiety Inventory Manual. Palo Alto: Consulting Psychologists
Press.
Stamopoulos DC, Damos P and Karagianidou G (2007) Bioactivity
of five monoterpenoid vapours to Tribolium confusum (du Val)
(Coleoptera: Tenebrionidae). J Stored Prod Res
43: 571–577.
Kennedy et al. 1099
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
Tildesley NTJ, Kennedy DO, Perry EK, et al. (2003) Salvia lavandu-
laefolia (Spanish Sage) enhances memory in healthy young vol-
unteers. Pharmacol Biochem Behav 75: 669–674.
Tildesley NTJ, Kennedy DO, Perry EK, Ballard CG, Wesnes KA and
Scholey AB (2005) Positive modulation of mood and cognitive
performance following administration of acute doses of Salvia
lavandulaefolia essential oil to healthy young volunteers. Physiol
Behav 83: 699–709.
Vuorinen T, Reddy GVP, Nerg A-M and Holopainen JK (2004)
Monoterpene and herbivore-induced emissions from cabbage
plants grown at elevated atmospheric CO2 concentration.
Atmos Environ 38: 675–682.
Wang S, Chen Y, Gao Z, Xiong M, Zhong Z and Ye L (2007) Gas
chromatographic-mass spectrometric analysis of d-limonene in
human plasma. J Pharm Biomed Anal 44: 1095–1099.
Wetherell MA and Sidgreaves MC (2005) Short communication:
secretory immunoglobulin – a reactivity following increases in
workload intensity using the defined intensity stressor simulation
(DISS). Stress Health 21: 99–106.
Wink M (2003) Evolution of secondary metabolites from an ecolog-
ical and molecular phylogenetic perspective. Phytochemistry 64:
3–19.
Wink M, Schmeller T and Latz-Bruning B (1998) Modes of action of
allelochemical alkaloids: Interaction with neuroreceptors, DNA,
and other molecular targets. J Chem Ecol 24: 1881–1937.
Wittstock U and Gershenzon J (2002) Constitutive plant toxins and
their role in defense against herbivores and pathogens. Curr Opin
Plant Biol 5: 300–307.
1100 Journal of Psychopharmacology 25(8)
at SWINBURNE UNIV on August 15, 2011jop.sagepub.comDownloaded from
