Eur J Nutr (2012) 51:909–916

DOI 10.1007/s00394-011-0263-7

ORIGINAL CONTRIBUTION

Effect of tart cherry juice (Prunus cerasus) on melatonin levels

and enhanced sleep quality
Glyn Howatson • Phillip G. Bell • Jamie Tallent •

Benita Middleton • Malachy P. McHugh •

Jason Ellis

Received: 7 July 2011 / Accepted: 10 October 2011 / Published online: 30 October 2011
Ó Springer-Verlag 2011

Abstract Trial differences were determined using a repeated measures

Background Tart Montmorency cherries have been ANOVA.
reported to contain high levels of phytochemicals including Results Total melatonin content was significantly ele-
melatonin, a molecule critical in regulating the sleep-wake vated (P \ 0.05) in the cherry juice group, whilst no dif-
cycle in humans. ferences were shown between baseline and placebo trials.
Purpose The aim of our investigation was to ascertain There were significant increases in time in bed, total sleep
whether ingestion of a tart cherry juice concentrate would time and sleep efficiency total (P \ 0.05) with cherry juice
increase the urinary melatonin levels in healthy adults and supplementation. Although there was no difference in
improve sleep quality. timing of the melatonin circardian rhythm, there was a
Methods In a randomised, double-blind, placebo-con- trend to a higher mesor and amplitude.
trolled, crossover design, 20 volunteers consumed either a Conclusions These data suggest that consumption of a
placebo or tart cherry juice concentrate for 7 days. Measures tart cherry juice concentrate provides an increase in
of sleep quality recorded by actigraphy and subjective sleep exogenous melatonin that is beneficial in improving sleep
questionnaires were completed. Sequential urine samples duration and quality in healthy men and women and might
over 48 h were collected and urinary 6-sulfatoxymelatonin be of benefit in managing disturbed sleep.
(major metabolite of melatonin) determined; cosinor anal-
ysis was used to determine melatonin circadian rhythm Keywords Tart cherries
Melatonin
Sleep
Recovery
(mesor, acrophase and amplitude). In addition, total urinary
melatonin content was determined over the sampled period.
Introduction

Tart Montmorency cherries (Prunus cerasus), rich in

G. Howatson (&)
P. G. Bell
J. Tallent
J. Ellis numerous phytochemicals, provide a range of health benefits
School of Life Sciences, Northumbria University, that include reduction in symptoms associated with gout [1],
Northumberland Building, Newcastle upon Tyne NE1 8ST, UK
down-regulation of circulating inflammatory markers [2],
e-mail: glyn.howatson@northumbria.ac.uk
analgesic effects following long-distance running [3],
G. Howatson reduced oxidative stress [4], improved recovery following
Centre for Aquatic Research, Department of Zoology, damaging exercise [5–7] and recently, improved sleep
University of Johannesburg, Johannesburg, South Africa
quality in late-life insomnia [8]. Mechanistically, it is
B. Middleton thought the phenolic compounds within tart cherries act as
Centre for Chronobiology, Faculty of Health and Medical ‘free radical’ scavengers that reduce oxidative stress [2]. In
Sciences, University of Surrey, Guildford, UK addition, the anti-inflammatory properties [9] of tart cherries
have been reported to be at a level comparable to a number of
M. P. McHugh
Nicholas Institute of Sports Medicine and Athletic Trauma, non-steroidal anti-inflammatory drugs [10]. In particular, the
Lenox Hill Hospital, New York, NY, USA anthocyanin content of tart cherries, which compares

123
910 Eur J Nutr (2012) 51:909–916

favourably with other fruits such as sweet cherries [11], Methods

seems to be of most interest, and these are likely to be
responsible for the anti-oxidative and anti-inflammatory Participants
effects.
A number of recent studies have shown that consump- Following institutional ethical approval from the School of
tion of tart cherry juice can accelerate recovery following Life Sciences Ethics Committee at Northumbria Univer-
strenuous exercise [5–7], where temporary perturbations in sity, UK in accordance with the Helsinki Declaration, 20
inflammation and oxidative stress can occur. These healthy men (n = 10) and women (n = 10) volunteered to
recovery effects have been attributed to the actions of the participate. The mean (±SD) age, height, body mass and
antioxidant and anti-inflammatory phytochemicals con- BMI were 26.6 (±4.6) years, 1.71 (±0.10) m, 72.5 (±15.0)
tained in tart cherries. Pigeon et al. [8] anecdotally reported kg and 24.7 (±3.5) kg m-2, respectively. The age range
claims of improved sleep with cherry juice supplementa- was restricted to 18–40 years to reduce the potential for
tion in participants from a previous trial [6]. Interestingly, age-related sleep disturbances that have been reported in
in addition to the aforementioned phenolic compounds, tart older adults [13]. In addition, all volunteers were physi-
cherries contain high concentrations of melatonin [12]. cally active and participated in moderate physical exercise
Melatonin has a strong influence on the sleep-wake cycle in for at least 150 min/week. After being informed of the
humans and is associated with sleep-promoting properties experimental procedures, participants were asked to com-
[13]. Physiologically, endogenous melatonin secretion plete a health screening questionnaire to ascertain contra-
adjusts according to the light/dark cycle and can directly indications to participation (prescription medicines, sleep
influence nocturnal core temperature and hence facilitate disturbance, special dietary habits, shift work or underlying
the propensity for sleep [14]. Additionally, a strong posi- medical pathology), and volunteers then provided written
tive relationship between increased melatonin and total informed consent.
sleep time in healthy, young individuals has been previ-
ously demonstrated [15]. Interestingly, the balance of evi- Experimental overview and study design
dence would suggest that exogenous melatonin in the
treatment of insomnia is equivocal at best; however, there Participants were supplemented with a tart cherry juice
is a good body of support for melatonin use in managing concentrate or placebo in a randomised, double-blind,
circadian rhythm disturbance, such as those seen from placebo-controlled, crossover study design, during normal
travelling time zones [16]. daily routine. Dependent variables were urinary 6-sul-
In a recent study, the efficacy of tart cherry juice con- phatoxymelatonin (aMT6s), diet recall, objective activity
sumption on sleep indices in a population with late-life recorded through actigraphy (variables) and subjective
insomnia was examined [8]. They reported modest sleep quality. A schematic of the study design is presented
improvements in subjective quality of sleep; however, no in Fig. 1. Initially, participants provided sequential urinary
objective measures of sleep, such as actigraphy, were voids across a 48 h period (days 1 and 2) in order to ana-
taken, and the potential mechanisms responsible for the lyse baseline measures of aMT6s. Over the same 2-day
reported sleep improvements (e.g. melatonin) were period, participants were issued with an activity monitor
impossible to discern. The authors [8] speculated that and completed online daily diet recalls and a sleep diary
increased dietary melatonin associated with consumption immediately following morning awakening. Participants
of tart cherry juice might be responsible for the changes. then continued to complete the questionnaires and diaries
However, there is an alternative hypothesis; the anti- and wear the activity monitors for the remainder of the trial
inflammatory properties of tart cherries may have some period.
influence on the pro-inflammatory cytokines involved in Following the 48-h baseline period, participants were
sleep regulation [17]. Given the potential benefits of tart randomly assigned to either the tart cherry juice concen-
cherry juice in delivering exogenous melatonin and trate or placebo (starting on day 3) for a period of 7 days;
improving sleep quality, we hypothesised that the con- this was based on loading phases from previous studies
sumption of a tart cherry juice concentrate in young, showing efficacy using cherry juice [5–8]. Participants
healthy adults would increase urinary 6-sulfatoxymelatonin were instructed to consume two servings of either tart
and improve indices of sleep quality. Therefore, the aim cherry juice concentrate or fruit-flavoured cordial each day,
of this investigation was to examine the effects of for 7 days. In the last 48 h of this supplementation period,
tart Montmorency cherry juice concentrate on urinary urine was again collected in an identical manner as pre-
6-sulfatoxymelatonin and sleep quality using a double- viously described for the baseline period. Following a
blind, placebo-controlled, crossover design. 14-day washout period, participants repeated the baseline

123
Eur J Nutr (2012) 51:909–916 911

Fig. 1 A schematic outlining

the implemented protocol.
Supplementation periods
consisted of two 30-mL
servings per day of either a tart
cherry juice concentrate of
placebo

and experimental period whilst consuming the other In addition, other active compounds contained within the tart
supplement. cherry juice were verified by the aforementioned laboratory
and included anthocyanins such as malvidin, cyanidin, pe-
Dietary control largonidin, peonidin, delphinidin, petunidin (total anthocy-
anin content = 9.117 mg mL-1), vitamin A—as beta-
All volunteers completed a dietary recall throughout the carotene (22.64 IU mL-1) and vitamin C—ascorbic acid
baseline and supplementation periods. Participants were (0.324 mg mL-1).
asked to replicate their diet during the first supplementation The placebo was a commercially available, economy,
period as closely as possible during the second period in mixed fruit cordial (containing less than 5% fruit) that was
order to standardise the dietary intake between trials. reported to contain no melatonin or anthocyanins and a
Additionally, in order to isolate dietary melatonin as clo- trace of vitamin C. Participants were instructed to take the
sely as possible, participants were issued with a list of same dose (30 mL) diluted with *200 mL of water.
foods that are known to contain or influence melatonin and
were subsequently asked to abstain from consuming these Dependent variables
for the duration of the trial. Portions of foods thought to
contain antioxidants were totalled for each day then aver- Urine collection and analysis
aged across the experimental period.
Sequential urinary voids were collected 48-h periods to
Supplementation ensure the entire circadian cycle was captured during
each part of the trial to allow for cosinor analysis that
Prior to starting the experiment, participants were informed provided measures of acrophase, mesor and amplitude.
that the trial was to ascertain the influence of two fruit con- Urine was collected in a sterilised measuring cylinder.
centrates on melatonin levels and sleep quality; however, the Void volume, time and date were recorded, before a
nature of the trial regarding tart cherry juice concentrate was 10 mL aliquot of urine was retained, refrigerated and
only revealed when the study had been completed. A serving returned to the laboratory the following morning for
of 30 mL of tart Montmorency cherry juice (Prunus cerasus) labelling and immediate storage at -80 °C for later anal-
concentrate (Cherry Active, Sunbury, UK) was consumed ysis for urinary aMT6s.
within 30 min of wakening and 30 min before the evening
meal on each of the 7-day supplementation periods. Each Urinary 6-sulphatoxymelatonin, aMT6s
30 mL serving was estimated to contain the equivalent of
approximately 90–100 tart cherries and was diluted with Urinary 6-sulphatoxymelatonin, aMT6s (the major metab-
approximately 200 mL of water. An independent laboratory olite of melatonin) was analysed in duplicate using a
(Atlas Bioscience Inc., Tuscan, AZ) conducted melatonin radioimmunoassay [20]. Samples belonging to the same
analysis of the cherry juice concentrate adapting an estab- participant were measured in the same assay run; the intra-
lished HPLC method [18]. The concentration of melato- assay coefficient of variation was \8%; the limit of
nin was 1.42 lg mL-1, which equates to a dose of detection was 0.25 ng mL-1. Evaluation of aMT6s profiles
*42.6 lg/30 mL serving or *85.2 lg day-1. Literature was performed using cosinor analysis, based upon the least
suggests that daily melatonin doses of *0.5–5 mg confer a square approximation of the time series using a cosine
positive effect in managing disturbed sleep rhythm [19]. function with a period of 24 h [21]. Parameters obtained

123
912 Eur J Nutr (2012) 51:909–916

from this analysis included the following: acrophase time: sleep variables were analysed with a repeated measures
the time of peak aMT6s concentration or the maximum of ANOVA (condition, placebo vs. cherry juice 2; time, pre
the fitted cosinor function; mesor: the mean aMT6s values vs. post). In addition, 95% confidence intervals were also
for all samples included in the cosinor anlaysis; and determined to illustrate the magnitude of change. Finally,
amplitude: the difference between mesor and peak aMT6 baseline measures were examined for differences using a
concentrations. Acrophase time was classed as normal if it paired samples t test. Significance was set at an alpha level
occurred between midnight and 06:00 h [22]. ‘Goodness of of 0.05.
fit’ measures were used to determine the validity of the
cosinor-derived indices: (1) the % rhythm or variability
accounted for by the cosine curve: 100% rhythm = all data
Results
points fall on the cosine curve and 0% rhythm = none of
the data points fall on the cosine curve and (2) the likeli-
Baseline measures taken before the placebo and cherry
hood of data points fitting a straight line as opposed to a
juice trials were not different for any of the dependent
cosine curve, expressed as a P value. Data were considered
variables (P [ 0.05). All variables returned an observed
acceptable if the cosinor fit was significant (P B 0.05) and
power value ranging from 0.156 to 0.999. Although we
the % rhythm C50% [21]. Finally, the total aMT6s
did not quantify all the food consumption using dietary
excreted per 24 h was calculated and a mean for each 48-h
analysis, participants reported a similar diet across the
collection period determined.
trials. In an attempt to quantify this more fully, we
recorded the number of food portions thought to contain
Actigraphy
antioxidants (including melatonin) across the supplemen-
tation periods, 22.8 versus 22.4 for the cherry juice and
Polysomnography (PSG) offers the most accurate assess-
placebo groups, respectively. A paired samples t test
ment of sleep and sleep quality; however, given the nature
showed no significant differences between groups
of the study, we utilised actigraphy that has been shown to
(t = 1.162, P = 0.259).
be reliable and has good agreement with PSG [23]. Par-
ticipants were issued with actigraphy that was worn on the
wrist of the non-dominant arm (Actiwatch 7, CamnTech, Urinary 6-sulphatoxymelatonin (aMT6s)
Cambridge, UK). These were worn for the duration of the
study and removed only for bathing, showering or other The baseline measures preceding the placebo and cherry
aquatic activities. Participants were asked to activate the juice supplements were not different (t = 0.921,
marker function on the watch when getting into bed and P = 0.369). The repeated measures ANOVA showed a
when rising the following morning. Analysis was made on significant trial effect (F = 23.0, P \ 0.001). A significant
sleep efficiency (SE), sleep onset latency (SOL), time in interaction (F = 23.0, P \ 0.001) was also found and
bed (TIB), fragmentation index (FRAGI), total sleep time post hoc analysis revealed that the cherry juice trial was
(TST) and sleep efficiency total (SET); these variables significantly greater than baseline and placebo trials
were calculated using Actiwatch software (Actiwatch, (P \ 0.001; 95% CI = 2,828–5,393 ng and 2,519–5,450 ng,
CamnTech, Cambridge, UK). The mean values for each respectively); there was no difference between baseline or
sample period were used for data analysis. placebo groups (Fig. 2).
Despite the significant increase in total urinary aMT6 s,
Subjective measures the cosinor analysis examining circadian rhythm of aMT6s
showed no differences between trials for any variable
Online subjective sleep diaries were reported immediately (Table 1). Cosinor analysis relies on a significant ‘fit’ of
following awakening each day during both baseline and the melatonin response; of the 80 sets of data collected,
trial periods. This commonly used self-reporting method 22.5% (n = 18) were excluded because they did not sig-
tool has been to be a reliable (90%) measure [24] and nificantly fit according to the cosinor algorithm. Of the
allowed calculation of the following: SE, SOL, wake after remaining data, there were small non-significant rises in the
sleep onset (WASO), napping (NAP), TST and SET [25]. amplitude and mesor in the cherry juice trial, whilst the
acrophase remained largely unchanged throughout. A
Statistical analyses representative example where urinary voids were collected
for a single participant at similar times of the day between
Values are reported as mean (±SD), unless otherwise sta- the placebo and cherry juice trials is presented in Fig. 3;
ted. The cosinor data, total aMT6s and all quantitative these data span the circadian changes in aMT6s over
(actigraphy)- and qualitative (questionnaire)-dependent sequential 48-h periods.

123
Eur J Nutr (2012) 51:909–916 913

Fig. 3 A representative example of a single subject’s circadian

rhythm for urinary melatonin (aMT6s) during the placebo and cherry
juice trials over sequential 48-h periods

Fig. 2 Mean (±SEE) urinary melatonin (aMT6) secretion for the

group following baseline placebo (control), placebo, baseline cherry
cherry juice trial was greater than the baseline and placebo
juice (control) and cherry juice trials.Asterisk denotes that cherry
juice supplementation resulted in significantly greater aMT6s than trials (P B 0.017; 95% CI = 2.1–7.5 and 0.5–9.4, respec-
baseline and placebo trials (P B 0.05) tively). A summary of these data is presented in Table 3.

Sleep indices: subjective questionnaire and actigraphy

Discussion
There was a 100% completion rate for the sleep ques-
tionnaires. There was no differences across the different The aim of this investigation was to ascertain whether the
trials for SET, SOL, TST and WASO; however, napping supplementation of tart Montmorency cherry juice con-
time did show a significant trial and interaction effect centrate would (1) increase the urinary aMT6s content
(F = 5.591, P = 0.029), with significantly less napping and (2) improve the objective and subjective sleep indices
time in the cherry juice trial compared to baseline and of young, healthy individuals. We hypothesised that
the placebo trials (P B 0.031; 95% CI = 0.7–13.6 and urinary aMT6s would rise and that sleep parameters
0.7–11.1 min, respectively), and there were no differences would improve as a consequence. This is the first inves-
between baseline measures and the placebo trial (Table 2). tigation to demonstrate that dietary tart cherry juice
Whilst all participants wore the activity monitor for the concentrate increases urinary melatonin levels and pro-
duration of the trials, 17 out of 80 trials (21.3%) were vides improved sleep time and quality in a healthy adult
excluded from analysis due to missing data. There were no population.
differences in SOL and FRAGI; however, there was a sig- The sleep diary information showed that napping time
nificant trial and interaction effect for TIB (F = 7.056, decreased with the administration of cherry juice, whereas
P = 0.016) where cherry juice significantly increased time the actigraphy showed an increase in TIB, TST and SET, a
in bed compared to both baseline and placebo trails global measure of sleep quality. Notwithstanding the rel-
(P B 0.017; 95% CI = 4.5–45.2 and 4.7–40.2 min, atively low baseline SET observed from the actigraphy in
respectively). Furthermore, TST showed a trial and inter- these apparently good sleepers, the cherry juice nonetheless
action effect (F = 11.189, P = 0.003), where the cherry showed a 5–6% increase in SET, which likely to be
juice trial was significantly greater TST than baseline and influenced by the significant increase in TST. In addition,
placebo trials (P B 0.003; 95% CI = 15.2–39.7, 14.7–63.6, given that napping decreased and total time in bed also
respectively). In addition, SET showed significant trial and increased during the cherry juice trial, this is perhaps
interaction effects (F = 5.410, P = 0.031), where the unsurprising. What is also interesting to note is that there

Table 1 Mean (±SD) cosinor analysis based on melatonin circadian rhythm for all experimental conditions
Baseline placebo Placebo Baseline cherry juice Cherry juice
-1
Mesor (ng 9 h ) 17.98 (6.04) 19.17 (7.37) 18.64 (9.76) 21.59 (6.85)
-1
Amplitude (lg 9 h ) 27.39 (15.78) 27.54 (8.37) 27.05 (10.72) 28.57 (15.01)
Acrophase (time) 4.03 (1.03) 3.55 (1.22) 4.05 (1.40) 4.01 (1.01)
Of the possible 80 data sets, 18 did not significantly fit the cosinor curve and were excluded from the analysis

123
914 Eur J Nutr (2012) 51:909–916

Table 2 Subjective sleep questionnaire variables for all conditions; values are mean (±SD)
Baseline placebo Placebo Baseline cherry juice Cherry juice

SE (%) 89.3 (7.3) 91.7 (4.0) 90.0 (6.2) 91.1 (4.9)

SOL (mins) 40.3 (31.6) 39.5 (23.2) 39.8 (25.6) 34.2 (20.5)
WASO (mins) 36.0 (33.0) 19.2 (31.2) 28.8 (30.6) 27.6 (28.2)
Naps (mins) 9.0 (15.1) 7.8 (10.7) 8.6 (13.2) 1.9 (3.5)*
TST (mins) 447 (60) 476 (31) 452 (49) 475 (30)
SET (%) 88.1 (6.8) 90.4 (4.4) 89.4 (5.8) 90.7 (4.9)
SE sleep efficiency, SOL sleep onset latency, WASO wake after sleep onset, total SET sleep efficiency total, TST total sleep time
* Denotes significantly different from all other conditions (P B 0.05)

Table 3 Actigraphy variables for all conditions; values are mean (± SD)
Baseline placebo Placebo Baseline cherry juice Cherry juice

SE (%) 82.8 (15.7) 84.1 (5.8) 83.9 (7.8) 86.8 (3.6)

SOL (mins) 28.9 (21.3) 30.5 (34.8) 29.1 (26.8) 21.4 (11.1)
Time in bed (mins) 491.8 (36.7) 492.2 (40.6) 490.0 (32.9) 514.7 (17.0)*
FRAGI (AU) 36.8 (8.2) 35.2 (9.3) 35.8 (8.9) 34.2 (7.6)
TST (mins) 392 (28) 380 (49) 385 (30) 419 (22)*
SET (%) 77.5 (5.9) 77.4 (8.5) 76.8 (6.9) 82.3 (3.6)*
Of the possible 80 data sets, 17 were excluded due to technical issues
SE sleep efficiency, SOL sleep onset latency, FRAGI fragmentation index, total SET sleep efficiency total, TST total sleep time
* Denotes significantly different from all other conditions (P B 0.05)

were non-significant trends towards decreased SOL, which cherries on melatonin metabolism across the course of the
is also likely to have influenced the SET. day. This is especially important when one considers that
Only one study has investigated tart Montmorency the half-life of melatonin is relatively short and it is pos-
cherry juice and sleep parameters (a fresh pressed cherry sible to miss fluctuations in melatonin throughout the day.
juice blended with apple juice, as opposed to a pure cherry Further, there is no indication in the aforementioned
juice concentrate) [8]. They found that elderly individuals, studies [8, 25] of an experimental control or record of
with moderate/severe insomnia, reported improved sleep dietary intake, which makes interpretation of the data
quality, and it was hypothesised that this was due to the problematic. Although we were not able to quantify the
increased exogenous melatonin content afforded by the exact nutritional content for each subject, food intake was
cherry juice. Unfortunately, they did not measure melato- estimated. Participants replicated diet as closely as possible
nin; however, data from our investigation lend additional from trial to trial. This was based on number of portions
evidence that improved sleep quality is mediated by the thought to contain antioxidants—there were no differences
increase in dietary melatonin contained within the cher- between trials. Support for the efficacy of this approach can
ries. Interestingly, a recent addition to the literature [26] be seen by the fact that aMT6s and sleep parameters were
examined the increase in melatonin content from dietary unchanged in baseline and placebo trials, whereas aMT6s
intake of Jert Valley cherries (seven varieties, none of was significantly elevated in the cherry juice trial. Given the
which were Montmorency tart cherries). They showed that dietary control used in the current investigation, coupled
in a very small population of middle-aged and elderly with the significant changes melatonin, we can add support
volunteers, there was an increase in urinary melatonin and to research showing cherries [26], and specifically in this
some modest improvements in sleep parameters. This case, Montmorency cherries improve sleep parameters in
investigation [26] based its observations on the first healthy individuals, which is likely due to the increase in
morning void only, whereas all urinary voids were captured dietary melatonin. These data support previous work
for a 48-h period during each part of the current trial. This showing improved sleep in healthy younger adults with
approach, whilst still having limitations, allows for cosinor exogenous melatonin supplementation [14]; but addition-
analysis and tracking of the circadian rhythm and provides ally, it provides a potential alternative to traditional mela-
a more comprehensive picture of the dietary effects of tonin supplementation in the form of a functional food.

123
Eur J Nutr (2012) 51:909–916 915

The secretion of melatonin is influenced by light/dark cosinor analysis) that despite increased total sleep time and
cycles and ultimately is instrumental in the sleep/wake improved sleep quality, this is not the case in asymptom-
cycle [13]. From a physiological perspective, given that atic, healthy younger adults. Notwithstanding this, aMT6s
endogenous melatonin influences core temperature and levels are increased with tart cherry juice consumption, but
facilitates sleep [14], it makes the expectation tenable that an investigation that examines elderly individuals, perhaps
increased exogenous melatonin will further facilitate with disturbed sleep, that incorporates measures of circa-
changes in core temperature and hence be responsible for dian rhythms and sleep quality would be a valuable addi-
the improvements in sleep quality. Further work examining tion to the literature.
the potential physiological outcomes (such as core tem- In conclusion, this is the first study to show direct evi-
perature and EEG in polysomnographic paradigms) from dence that dietary supplementation with a tart Montmorency
exogenous melatonin, specifically from functional foods, cherry juice concentrate increases circulating melatonin and
would be useful additions to the literature. In an attempt to can provide modest improvements in sleep time and quality
elucidate the relationship between the change in SET and in healthy adults with no reported disturbed sleep. Although
the change in melatonin, we conducted a Pearson’s corre- the interaction of other phytochemicals cannot be com-
lation coefficient analysis and found that there was a pletely ruled out, these data provide a mechanism of action
modest relationship (r = 0.416, P [ 0.05), indicating that for the previously conjectural reports of improved sleep
other factors may influence the variables associated with quality with cherry juice supplementation. Subsequently,
sleep quality. Importantly, a limitation with our data is that Montmorency tart cherry juice concentrate might therefore
*21% of the actigraphy data were missing, which may present a suitable adjunct intervention for disturbed sleep
have influenced this correlation and also the non-significant across a number of scenarios in healthy and symptomatic
trends in other actigraphy-dependent variables. Further- individuals.
more, 50% of our sample was women, and it is conceivable
that they were in different stages of the menstrual cycle, Acknowledgments Gratitude is extended to Kelly Mitcheson for
her help in data collection and to CherryActive (Sunbury, UK) for
which may also influence core temperature and hence the donating the cherry juice concentrate.
propensity for sleep disturbance [27]. Future research
might wish to consider this issue in future experiments.
Melatonin is not the only candidate mechanism, given References
that sleep regulation is also influenced by pro-inflammatory
cytokines [17]. Tart cherries have been shown to contain 1. Jacob RA, Spinozzi GM, Simon VA, Kelley DS, Prior RL, Hess-
numerous phenolic compounds that have anti-inflammatory Pierce B, Kader AA (2003) Consumption of cherries lowers
plasma urate in healthy women. J Nutr 133(6):1826–1829
and antioxidant properties that can increase antioxidant 2. Kelley DS, Rasooly R, Jacob RA, Kader AA, Mackey BE (2006)
capacity [5, 26]. Furthermore, cherry juice has been shown Consumption of Bing sweet cherries lowers circulating concen-
to decrease oxidative stress and inflammation following trations of inflammation markers in healthy men and women.
strenuous exercise [5] making it possible that these anti- J Nutr 136(4):981–986
3. Kuehl K, Perrier E, Elliot D, Chesnutt J (2010) Efficacy of tart
oxidant and/or anti-inflammatory properties modulated cherry juice in reducing muscle pain during running: a random-
indices of sleep in this study, although this remains to be ized controlled trial. J Int Soc Sports Nut 7(1):17
demonstrated in an experimental model. 4. Traustadottir T, Davies SS, Stock AA, Su Y, Heward CB, Roberts
It has been previously speculated that the positive LJ II, Harman SM (2009) Tart cherry juice decreases oxidative
stress in healthy older men and women. J Nutr 139(10):
effects on sleep seen from tart cherries might be due to 1896–1900. doi:10.3945/jn.109.111716
improvements in circadian regulation [8]. We observed no 5. Howatson G, McHugh MP, Hill JA, Brouner J, Jewell AP, Van
changes in mesor, amplitude or acrophase, although there Someren KA, Shave RE, Howatson SA (2010) Influence of tart
was a trend towards a higher mesor (essentially equating to cherry juice on indices of recovery following marathon running.
Scand J Med Sci Sports 20(6):843–852. doi:10.1111/j.1600-
the mean value across the circadian cycle). This is perhaps 0838.2009.01005.x
not surprising given that the total urinary melatonin did 6. Connolly DAJ, McHugh MP, Padilla-Zakour OI (2006) Efficacy
increase with cherry juice supplementation. An obvious of a tart cherry juice blend in preventing the symptoms of muscle
difference with the previous work lies within the subject damage. Br J Sports Med 40(8):679–683. doi:10.1136/bjsm.2005.
025429
populations; Pigeon et al. [8] used older adults suffering 7. Bowtell JL, Sumners DP, Dyer A, Fox P, Mileva K (2011)
from moderate/severe insomnia and did not measure cir- Montmorency cherry juice reduces muscle damage caused by
cadian rhythm or melatonin, whereas the current investi- intensive strength exercise. Med Sci Sports Exerc 43(8):1544–
gation used asymptomatic younger adults (B37 years) and 1551
8. Pigeon WR, Carr M, Gorman C, Perlis ML (2010) Effects of a
did measure circadian rhythm and melatonin. Conceivably, tart cherry juice beverage on the sleep of older adults with
cherries might help regulate circadian rhythm in those with insomnia: a pilot study. J Med Food 13(3):579–583. doi:
disturbed sleep [8]; however, our evidence shows (from 10.1089/jmf.2009.0096

123
916 Eur J Nutr (2012) 51:909–916

9. Wang H, Nair MG, Strasburg GM, Chang Y-C, Booren AM, 19. Herxheimer A, Pertrie KJ (2002) Melatonin for the prevention of
Gray JI, DeWitt DL (1999) Antioxidant and antiinflammatory jet lag. Cochrane Database Syst Rev 9:11–24
activities of anthocyanins and their aglycon, cyanidin, from tart 20. Aldhous ME, Arendt J (1988) Radioimmunoassay for 6-sul-
cherries. J Nat Prod 62(2):294–296. doi:10.1021/np980501m phatoxymelatonin in urine using an iodinated tracer. Ann Clin
10. Kim D-O, Heo HJ, Kim YJ, Yang HS, Lee CY (2005) Sweet and Biochem 25(3):298–303
sour cherry phenolics and their protective effects on neuronal cells. 21. Minors DS, Waterhouse JM (1988) Mathematical and statistical
J Agric Food Chem 53(26):9921–9927. doi:10.1021/jf0518599 analysis of circadian rhythms. Psychoneuroendocrinology
11. Wang H, Nair MG, Iezzoni AF, Strasburg GM, Booren AM, Gray 13(6):443–464
JI (1997) Quantification and characterization of anthocyanins in 22. Lockley SW, Skene DJ, Tabandeh H, Bird AC, Defrance R,
Balaton tart cherries. J Agric Food Chem 45(7):2556–2560. doi: Arendt J (1997) Relationship between napping and melatonin in
10.1021/jf960896k the blind. J Biol Rhythm 12(1):16–25. doi:10.1177/07487304
12. Burkhardt S, Tan DX, Manchester LC, Hardeland Rd, Reiter RJ 9701200104
(2001) Detection and quantification of the antioxidant melatonin 23. Ancoli-Israel S, Cole R, Alessi C, Chambers M, Moorcroft W,
in Montmorency and Balaton tart cherries (Prunus cerasus). Pollak CP (2003) The role of actigraphy in the study of sleep and
J Agric Food Chem 49(10):4898–4902. doi:10.1021/jf010321? circadian rythyms. Sleep 26(3):342–392
13. Hughes RJ, Sack RL, Lewy AJ (1998) The role of melatonin and 24. Rogers AE, Caruso CC, Aldrich MS (1993) Reliability of sleep
circadian phase in age-related sleep-maintenance insomnia: diaries for assessment of sleep/wake patterns. Nurs Res
assessment in a clinical trial of melatonin replacement. Sleep 42(6):368–372
21(1):52–66 25. Lockley SW, Skene DJ, Arendt J (1999) Comparison between
14. Claustrat B, Brun J, Chazot G (2005) The basic physiology and subjective and actigraphic measurement of sleep and sleep
pathophysiology of melatonin. Sleep Med Rev 9(1):11–24 rhythms. J Sleep Res 8(3):175–183. doi:10.1046/j.1365-2869.
15. Morris M, Lack L, Barrett J (1990) The effect of sleep/wake state 1999.00155.x
on nocturnal melatonin excretion. J Pineal Res 9(2):133–138. doi: 26. Garrido M, Paredes SD, Cubero J, Lozano M, Toribio-Delgado
10.1111/j.1600-079X.1990.tb00701.x AF, Muñoz JL, Reiter RJ, Barriga C, Rodrı́guez AB (2010) Jerte
16. Ferguson SA, Rajaratnam SMW, Dawson D (2010) Melatonin valley cherry-enriched diets improve nocturnal rest and increase
agonists and insomnia. Expert Rev Neurother 10(2):305–318 6-sulfatoxymelatonin and total antioxidant capacity in the urine
17. Opp MR (2004) Cytokines and sleep: the first hundred years. of middle-aged and elderly humans. J Gerontol Ser A Biol Sci
Brain Behav Immun 18(4):295–297 Med Sci. doi:10.1093/gerona/glq099
18. Iinuma F, Hamase K, Matsubayashi S, Takahashi M, Watanabe 27. Coyne MD, Kesick CM, Doherty TJ, Kolka MA, Stephenson LA
M, Zaitsu K (1999) Sensitive determination of melatonin by (2000) Circadian rhythm changes in core temperature over the
precolumn derivatization and reversed-phase high-performance menstrual cycle: method for noninvasive monitoring. Am J
liquid chromatography. J Chromatogr A 835(1–2):67–72 Physiol Regul Integr Comp Physiol 279:R1316–R1320

123