Consumption of both low and high (-)-epicatechin apple puree attenuates platelet reactivity
and increases plasma concentrations of nitric oxide metabolites: A randomized controlled
trial
Amy Gaspera, Wendy Hollandsa, Amelie Casgraina, Shikha Sahaa, Birgit Teuchera, Jack R.
Daintya, Dini P. Venemab, Peter C. Hollmanb, Maarit J. Reinc, Rebecca Nelsonc, Gary
Williamsonc,d, Paul A. Kroona*.
RUNNING TITLE: Anti-platelet effects of apple puree ingestion
a
Food and Health Programme, Institute of Food Research, Norwich Research Park, Norwich NR4
7UA, UK
b
RIKILT-Institute of Food Safety, Wageningen, The Netherlands.
c
Nestle Research Center, Vers-chez-les-Blanc, 1000 Lausanne 26, Switzerland.
d
School of Food Science and Nutrition, University of Leeds, Leeds, LS2 9JT, UK
*Corresponding author: Dr Paul Kroon, tel:+44 (0) 1603 255236, e-mail: paul.kroon@ifr.ac.uk
1
1
ABSTRACT
2
We hypothesised that consumption of flavanol-containing apple puree would modulate platelet
3
activity and increase nitric oxide metabolite status, and that high flavanol apple puree would exert a
4
greater effect than low flavanol apple puree. 25 subjects consumed 230 g of apple puree containing
5
25 and 100 mg epicatechin (low and high flavanol apple puree, respectively) and aspirin (75 mg) in
6
random order. Measurements were made at baseline, acutely after treatment (2, 6 and 24 h), and
7
after 14 d of treatment. Low flavanol apple puree significantly attenuated ADP and epinephrine-
8
induced integrin-β3 expression 2 h and 6 h after consumption and ADP and epinephrine-induced P-
9
selectin expression within 2 h of consumption. High flavanol apple puree attenuated epinephrine
10
and ADP-induced integrin-β3 expression after 2 and 6 h. ADP and epinephrine-induced integrin-β3
11
expression was significantly attenuated 2, 6 and 24 h after consumption of aspirin, while 14 d
12
aspirin consumption attenuated collagen-induced P-selectin expression only. The plasma total nitric
13
oxide metabolite conc. was significantly increased 6 h after consumption of both low and high
14
flavanol apple purees. In conclusion, consumption of apple purees containing ≥25 or 100 mg
15
flavanols transiently attenuated ex vivo integrin-β3 and P-selectin expression and increased plasma
16
nitric oxide metabolite conc. in healthy subjects, but the effect was not enhanced for the high
17
flavanol apple puree.
18
19
20
Keywords : platelet reactivity, nitric oxide, flavonoids
21
22
23
2
24
Cardiovascular disease is one of the major causes of mortality worldwide; for example in
25
both the US and the UK CVD accounts for around 35% of all deaths. The aetiology of CVD is
26
multi-factorial but there is accumulating evidence from epidemiological investigations to suggest an
27
association between improved cardiovascular health and a diet rich in flavanols (1,2). Flavanols are one
28
of the 6 sub-classes of flavonoids and comprise monomeric flavanols or ‘catechins’ (catechin and
29
epicatechin, and gallo and galloylated derivatives such as epigallocatechin gallate) and oligo-
30
/polymeric flavanols collectively known as proanthocyanidins. Numerous short and long term
31
intervention studies using healthy volunteers and at risk groups have reported on the effects of a
32
variety of flavanol-rich plant foods and extracts on risk markers for CVD, and flavanols have been
33
ascribed with various cardio-protective properties including anti-inflammatory
34
activities
35
meta-analysis of randomized controlled trials designed to test the effects of flavonoids on established
36
markers of CVD risk
37
improve flow mediated dilation (FMD)*(9) of the brachial artery. FMD, a measure of endothelial
38
function in humans, is almost exclusively mediated by nitric oxide (NO)
39
Cocoa consumption has been shown to cause acute increases in endothelial NO production
40
there is evidence to show that this is likely mediated by epicatechin (7).
41
(4,5)
, and the ability to reduce LDL oxidation
(6)
(3)
and anti-platelet
and improve endothelial function
(7)
. In a
(8)
, cocoa and chocolate were shown to significantly lower blood pressure and
(10)
, a potent vasodilator.
(11)
and
Platelet activation and subsequent aggregation is a critical event occurring in the pathogenesis
(12)
42
of CVD
43
activated the integrin beta-3 binds fibrinogen and von Willibrand factor which are secreted in
44
response to endothelial cell damage. The activated platelets in turn synthesize and secrete
45
thromboxane A2 (TXA2) and intracellular ADP which, combined with conformational changes in
46
the glycoprotein complex, result in platelet aggregation. Furthermore, P-selectin (an adhesion
47
molecule situated in the membrane of the platelet α-granule) is mobilised to the platelet surface
48
where it mediates platelet-leukocyte adhesion.
49
suppresses the production of TXA2, and integrin beta-3 inhibitors, are well recognized therapeutic
50
regimes used in the prevention and treatment of cardiovascular disorders. Flavanol rich foods and
51
beverages such as green tea, cocoa (and cocoa products) and grape juice have also been shown to
52
modulate platelet function
*
. Platelet aggregation is mediated by a surface fibrinogen receptor (integrin-β3). Once
Anti-platelet therapies such as aspirin, which
(4,13-17)
. Evidence for the anti-aggregatory effects of flavanol rich foods
Abbreviations : ASA, aspirin; CRP, C-reactive protein; ET-1, endothelin-1; HF-apple puree, high
flavanol apple puree; LF-apple puree, low flavanol apple puree; FMD, flow mediated dilation;
TXA2, thromboxane A2; MPA, meta-phosphoric acid.
3
53
are arguably strongest for cocoa which has been shown to lower ADP and epinephrine activated
54
platelet aggregation within 2-6 h of consumption
55
associated with a reduction in the expression of the surface protein integrin-β3. Flavanols mediate
56
platelet function through a variety of mechanisms. Purple grape juice
57
have been shown to inhibit platelet protein kinase C and human platelet 12-lipoxygenase activity.
58
Additionally, chocolate consumption has been shown to favourably affect eicosanoid synthesis
59
providing evidence that flavanols may possess anti-inflammatory properties (20) .
(15,18)
, the effects of which were shown to be
(17)
and cocoa
(19)
for example
60
Flavanols are the major flavonoids consumed by humans, with their contribution to mean total
61
flavonoid intakes estimated at >80% (21). The main contributors to flavanol intakes in Western Europe
62
are chocolate, apples, red wine and tea. Apples are widely consumed and therefore an important
63
source of flavanols in the diet. In this study we investigated the acute and long term effects of
64
apples (delivered as a puree) containing two different levels of flavanols (25 and 100 mg
65
epicatechin), on platelet function, serum lipid profiles, and plasma concentrations of nitric oxide
66
metabolites, CRP, vitamin C and endothelin-1.
67
68
Experimental methods
69
Unless specifically stated all measurements were conducted at the Institute of Food Research in
70
Norwich.
71
72
Chemicals and reagents.
73
The conjugated antibodies CD61 and CD62 were purchased from Invitrogen (Paisley, UK) and Pac-
74
1 from Becton Dickenson (Oxford, UK). ADP, epinephrine and collagen were supplied by Biodata
75
(Alpha laboratories, Hampshire, UK). The Chemiluminescent ET-1 immunoassay kit was supplied
76
by R&D systems (Abingdon, UK). The enzymes β-glucuronidase (Helix pomatia type H5), and
77
sulfatase (H. pomatia type H1), were purchased from Sigma-Aldrich (Poole, UK) and β-
78
glucuronidase, (type IX-A from E. coli) and sulfatase, (type VIII from Abalone Entrails) from Sigma
79
(St. Louis, MO, USA). The phenolic standards (taxifolin, (-)-epicatechin, sinapic acid, galangin,
80
(+)-catechin, phloretin) were obtained from Extrasynthese (Genay, France). All other chemicals
81
were of analytical or HPLC grade.
82
83
Participants and study design.
84
Forty seven potential participants were assessed for eligibility on the basis of a health questionnaire
85
and the results of clinical laboratory tests. The exclusion criteria were as follows: smoking; medical
86
conditions such as gastrointestinal disease, history of ulcers and gastro-intestinal bleeding, diabetes,
4
87
cancer, heart disease, stroke, asthma or hay fever; regular use of aspirin (at least once a week),
88
antacids or laxatives; dietary supplements (unless prepared to cease for 1 month preceding and
89
throughout the study); clinical results at screening judged to affect the study outcome or be
90
indicative of a health problem. Nineteen of the forty seven potential participants were excluded
91
from this study. Seventeen did not meet the inclusion criteria at eligibility assessment and two, even
92
though eligible, declined to participate with no reason given. Of the twenty eight participants
93
randomized to treatment, three were withdrawn by the researcher during the study (1 because
94
aspirin consumption was contra-indicated, 1 developed anaemia and 1 developed an allergy) (see
95
fig 1 for progress of participants through the phases of the crossover trial). Characteristics of the
96
twenty five participants (13 men and 12 women) who completed the study were (mean ± SD);
97
weight 77.5 ± 13.8 kg (range 54.8 – 108.4 kg), BMI 25.9 ± 4.2 kg/m2 (range 20.9 – 33.5 kg/m2) and
98
age 42 ± 11 years (range 23 – 64 y). This study was conducted in the Human Nutrition Unit at the
99
Institute of Food research, (UK) according to the guidelines laid down in the declaration of Helsinki
100
and all procedures involving human subjects were approved by the Human Research Governance
101
Committee of the Institute of Food Research and the Norfolk Research Ethics Committee (ref no:
102
06/Q0101/22). Each participant gave written informed consent prior to taking part in the trial. The
103
trial is registered with clinicaltrials.gov (NCT00568152).
104
The study was a randomized, three-phase crossover design, investigating the acute and long
105
term effects of consuming apple puree containing different amounts of flavanols on risk markers for
106
cardiovascular disease. An aspirin treatment was used as a positive control. Each of the three test
107
phases comprised a 4-week period of intervention followed by a washout period of at least 2 weeks.
108
During each period of intervention participants excluded from the diet food sources that are rich in
109
flavanols (e.g. cocoa, berries, grapes, red wine, apples and legumes) and limited others (e.g. tea and
110
coffee). In addition, consumption of alcohol and oily fish were also limited. However, these foods
111
and beverages were completely excluded from the diet 48 h before blood sampling. A list of
112
authorized and prohibited foods was given and compliance was monitored with the use of food
113
diary records. Participants were assessed at the start (d 1), middle (d 15) and end (d 29) of the
114
intervention period. On d 1 of each intervention period a fasting blood sample was obtained
115
following which participants commenced the low flavanol diet as described above. On d 15 of the
116
intervention period fasted participants had an intravenous cannula inserted and a baseline blood
117
sample (0 h) was obtained. Participants were given a standard breakfast consisting of 2 slices of
118
white toast (72 g) with spread (10 g) followed by either, 230 g apple puree containing 100 mg
119
epicatechin (HF-apple puree), 230 g apple puree containing 25 mg epicatechin (LF-apple puree) or
120
aspirin (75mg dispersed in 100 ml water). To limit variation in food and drink intakes, participants
5
121
refrained from drinking and eating for 2 h after the treatment. Further blood samples were collected
122
at 2, 6 and 24 h. For 24 h before and after consumption of the apple puree, participants were
123
instructed to collect all urine passed. Daily consumption of the respective treatment was continued
124
until d 29 of the intervention period following which another fasted blood sample and 24 h urine
125
collection was obtained.
126
On d 1, 15 and 29 fasting blood samples were collected for assessment of plasma C-reactive
127
protein (CRP), vitamin C, lipid profile (total cholesterol, HDL, LDL, triglycerides) and whole blood
128
platelet reactivity. Samples for Endothelin-1 (ET-1) and NO metabolites assessment were obtained
129
on d 15 and 29. Platelet reactivity and NO metabolites were also assessed at 2, 6 and 24 h after
130
consumption of the treatment on d 15.
131
Plasma samples from 16 individuals collected at 0, 2, 6, and 24 h on d 15 were also used to
132
quantify epicatechin concentrations. Urine was assessed for excretion of epicatechin before
133
treatment and after acute and long term treatment. Sub-samples of each 24 h collection were stored
134
at -40oC until analysis (see Fig 2 for overview of sample collection time points).
135
136
Apple purée preparation and analysis.
137
Two varieties of apples (Golden Delicious and Mitchalin, provided by Coressence Ltd) selected for
138
their differences in natural levels of flavanols were harvested for this study; the cores were removed
139
(but the skins retained) and the apples sliced and frozen. The frozen apple slices were processed by
140
Campden and Chorleywood Food Research Association to produce food grade apple purees. This
141
involved defrosting and gently heating the apples to cook/soften them, addition of citric acid to
142
achieve pH ≤3.8, passage through a colloidal mill to produce puree and then a period of heating at
143
85°C to pasteurise the puree prior to hot filling of lacquered cans and sealing of the cans. The apple
144
purée was packaged into individual 230 g portions. Quantification of flavanol content of the apple
145
purees was determined at the start and then at regular intervals throughout the study, and was
146
similar to the predicted content based on the flavanol content or raw apples. Samples of apple puree
147
were freeze dried and ground into a powder. Freeze dried powder (40 mg in triplicate) was
148
extracted with 70% methanol (950 µl) at 70oC for 20 min with 50 µl galangin (100 mg/L) added as
149
an internal standard. The supernatant was filtered into auto-sampler vials and analysed by HPLC
150
with an Agilent HP1100 using a Luna, silicon (2) column (250 x 4.60 mm, 5 micron particle size;
151
Phenomenex). The mobile phase consisted of (A) dichloromethane, (B) methanol and (C) aqueous
152
acetic acid (1:1 v/v). Samples were eluted with an increasing gradient of (B) as follows: 0-30 min,
153
14 – 28.4%, 30-45 min, 28.4 – 39.2%, 45 – 55 min, 39.2 – 86% and 55 – 65 min, 86 – 14% at a
6
154
flow rate of 1ml/min. Subsequently, the eluent was passed through a fluorescence detector
155
(excitation wavelength 276 nM, emission 316 nM). The contents of phenolics in the low and high-
156
flavanol apple purees were as follows [all mg (g fresh weight puree)-1]: Total hydroxycinnamates-
157
quinic acid esters, 0.138 and 0.497; Total flavanols, 1.037 and 2.020; Total dihydrochalcones. 0.055
158
and 0.067; Total flavonols, 0.095 and 0.066, respectively.
159
160
Vitamin C analysis.
161
Whole blood collected into EDTA tubes was immediately centrifuged at 2000 g for 10 min at room
162
temperature. Plasma samples (200 µl) were aliquoted into tubes containing 200 µl MPA to prevent
163
degradation. Prior to analysis samples were vortex mixed and centrifuged at 13,000 g for 5 min.
164
After centrifugation a sub-sample (100 µl) was added to ice-cold 6% MPA solution (300 µl) and
165
centrifuged at 13,000 g for a further 5 min. Supernatant (300 µl) was added to an auto-sampler vial
166
and analysed by HPLC (Agilent HP1100) using a Lichrosphere 100 RP-18 column (250 x 4.6 mm)
167
with a 5 µm particle size; Merck). The mobile phase (water - pH 2.2 acidified with sulphuric acid)
168
was pumped through the column at a flow rate of 1.0 ml/min over 10 min. Due to the labile nature
169
of ascorbic acid all samples were prepared under gold lighting, kept on ice throughout and analysed
170
within 4 h.
171
172
CRP and lipid profile measurements.
173
Whole blood collected into serum separating tubes was allowed to clot for 30 min before
174
centrifugation at 2000 g for 10 min. CRP and lipid (total cholesterol/HDL/LDL/triglycerides)
175
analysis was performed following standard protocols at the pathology department at the SPIRE
176
hospital in Norwich.
177
178
ET-1 and nitric oxide metabolites analysis.
179
For ET-1 analysis whole blood was collected into serum separating tubes and allowed to clot for 30
180
min prior to centrifugation at 2000 g for 10 min. Serum samples were subsequently stored at -40oC
181
until analysis. Samples were assayed in batches using a commercially available chemiluminescent
182
immunoassay kit according to the manufacturer’s instructions. All samples and standards were
183
measured in duplicate and analysed on a luminoskan ascent luminometer (Thermolab systems).
184
Quantification of ET-1 in serum was based on standard curves (range 0.4 - 2.3 ng/L) for each of the
185
plates used.
186
For measurement of NO metabolites in plasma, whole blood samples were collected into
187
EDTA tubes and immediately centrifuged at 1000 g for 10 min. Plasma aliquots were snap frozen
7
188
into cryovials and stored at -40oC until analysis. Analysis of nitric oxide metabolites in plasma was
189
conducted at the RKILT-Institute of Food Safety, Wageningen. The analysis was performed as
190
described by Feelisch et al
191
vessel, kept at 60 ºC, was filled with 20 mL freshly prepared Browns solution. The Browns solution
192
was stored on ice in the dark until use and made as follows: 0.65 M KI and 0.3 M I2 in milli-Q H2O
193
(shaken for 5 min at 250 rpm) was subsequently mixed with glacial acetic acid (1: 12 1/3) followed
194
by ultrasonic treatment for 5 min. Plasma samples, without any pretreatment, were injected into the
195
reaction vessel in triplicate (50 µL) through a septum that was replaced after each series of
196
measurements. Inside the reaction vessel NOx (NO2, NO2-, and nitrosated and nitrosylated species
197
(RNOs/ RSNOs) were reduced to NO (g). Helium functioned as a carrier gas to transport the NO (g)
198
through a condenser (3 ºC) followed by a 1 M NaOH solution, which was kept on ice, to remove
199
traces of acids. Subsequently, the NO (g) passed a 0.22 µm filter before it entered a
200
chemiluminescence detector (CLD 88 et., Eco Physics, Duernten, Switzerland). The detection range
201
was set at 0-50 ppb.
(22)
with only minor modifications. Briefly, a water-jacketed reaction
202
The whole system was kept at a constant overpressure of 1-1.05 bar throughout the
203
measurements. Browns solution was refreshed when peak broadening appeared or big bubbles were
204
generated in the reaction vessel. Calibration curves were made with potassium nitrite (0-1000 nM)
205
solved in a physiologic saline solution (0.9 % NaCl) (R2> 0.99). A standard plasma sample, which
206
was stored in small batches at -80°C, was used to determine the reproducibility of the NO
207
metabolite measurements. This control sample had an average NO metabolite content of 145 nM
208
(14 measurements of duplicates). The coefficients of variation were calculated to be 6.9 % within
209
days (14 duplicates) and 12.4 % between days (n=14). Results of a series of analysis were rejected
210
when the value obtained in this control sample exceeded ± 2 x SD from the average level. All
211
samples from one person were analyzed on the same day
212
213
Assessment of platelet reactivity.
214
Whole blood (3.0 mL) was collected directly into sodium citrate (3.2%) after discarding the first 2.0
215
mL of the draw. Samples were drawn with as little pressure as possible and without the use of a
216
tourniquet. Manipulation of the sample was kept to a minimum and samples were processed within
217
10 min of collection.
218
Whole blood (25 μL) was incubated for 10 min in polystyrene tubes (Sarstedt) at room
219
temperature with Mg-Hepes buffer (PH 7.4 unstimulated control), ADP (final conc. 10 µmol/L),
220
epinephrine (final conc. 10 µmol/L) or collagen (final conc. 4 mg/L). After 10 min samples were
221
suspended in 1.0 mL Mg-Hepes buffer. 100 µl aliquots of each of the activated mixes was then
8
222
transferred into tubes containing saturating concentrations of the fluorescent labelled monoclonal
223
antibodies, Pac-1 - fluorescein isothiocyanate (FITC). CD61 - allophycocyanin (APC) and CD62 –
224
phycoerythrin (PE) and incubated in the dark at room temperature for 20 min. Following antibody
225
binding samples were fixed with 1% paraformaldehyde to prevent further in-vitro platelet
226
activation, and stored in the dark until analysis.
227
The CD61-APC probe was used as a platelet identifier. Pac-1- FITC recognizes the integrin-
228
β3 conformation of the fibrinogen binding receptor on activated platelets. CD62- PE recognizes P-
229
selectin which is a component of the α-granule membrane and is expressed at the surface of the
230
activated platelet. All buffers were filtered and brought to room temperature before use. All
231
analyses were performed in duplicate.
232
All samples were analysed on the day of collection using a Beckman Coulter Cytomics
233
FC500 MPL flow cytometer. Instrument performance was verified using Flow Check Fluorospheres
234
before sample acquisition. The acquisition protocol, including compensation settings, was set up
235
prior to the start of the study using appropriate positive and negative controls. A total of 30,000
236
events were collected and platelets differentiated from other blood cells by their characteristic light
237
scatter profile and by gating for CD61 positive events. Following data acquisition by flow
238
cytometry, data analysis was performed using CXP software (Beckman Coulter V.2.1). Activated
239
platelets were defined as the percentage of CD61 positive events that expressed CD62 or Pac-1.
240
241
Extraction & quantification of epicatechin in plasma.
242
Analysis and quantification of epicatechin in plasma was conducted at the Nestle Research Centre,
243
Switzerland. Duplicate aliquots of thawed plasma (300 µL each) were spiked with 20 µL of 15 µM
244
internal standard (sinapic acid) solution (concentration in sample = 1 µM) and 300 µL β-
245
glucuronidase/sulfatase enzyme solution. The enzyme solution was prepared by adding 600 units of
246
β-glucuronidase and 60 units of sulfatase to 300 µL of 0.1 M sodium acetate (pH 5.2) per sample.
247
Samples were incubated for 60 min at 37 ºC with 400 rpm before diluting with water (400 µL). The
248
solid-phase extraction consisted of 4 steps: conditioning with MeOH and water, loading of the
249
plasma samples, washing with 5% MeOH, and eluting with MeOH and ACN. Post elution, samples
250
were dried under a stream of nitrogen, resuspended, and transferred to UPLC certified vials
251
(Waters). The samples were analysed using a method similar to that described by Renouf et al (23).
252
253
Extraction & quantification of epicatechin metabolites in urine.
254
Taxifolin 10 µl (1 mg/L) was added to sub-samples of acidified urine (200 uL) as an internal
255
standard. Samples were incubated with phosphate buffer (100 µl; pH 5.0) and enzymes (β9
256
glucuronidase and aryl-sulfatase) for 2 h. Following incubation, samples were acidified with formic
257
acid (5 µl), centrifuged at 17,000 g for 15 min at 4°C and the supernatant transferred into auto-
258
sampler vials for analysis as described by Saha et al
259
in urine was based on matrix-matched external standard curves of (-)-epicatechin (10-1000 µg
260
epicatechin / L urine) extracted in the same way as the samples and analysed immediately before
261
after each batch of samples. The analytical protocol required that samples were re-extracted and re-
262
analysed if the response factor of the post-sample standard curve was >10% different from the pre-
263
sample standard curve, but this did not occur. Standard curves were linear with regression
264
coefficients >99%. The limit of detection was 1.0 µg epicatechin / L urine.
(24)
. The quantification of total (-)-epicatechin
265
266
Statistical analysis.
267
Data were analysed with linear models using R (R Development Core Team R: A language and
268
environment for statistical computing. R Foundation for Statistical Computing, 2006 Vienna,
269
Austria. ISBN 3-900051-07-0, URL http://www.R-project.org). For all models, regression
270
diagnostics were checked to determine if data transformations, outlier omissions or alternative non-
271
parametric models were required. Data were analysed to assess the effects of time and treatment.
272
Plasma polyphenol data were analysed with a T-test. Results were considered significant if P<0.05
273
for all tests.
274
275
RESULTS
276
Data were analysed to assess the effects of acute and long term consumption of HF and LF-apple
277
puree on platelet function and serum lipid profiles, and plasma concentrations of nitric oxide
278
metabolites, CRP, vitamin C and endothelin-1. (See Fig 2 for details). An aspirin treatment (ASA)
279
was used as a positive control.
280
281
Changes in biomarkers during low flavanol diet run-in periods (d 1-14).
282
Data collected at the start and end of the low flavanol run-in diet of each of the three intervention
283
periods were combined. There were no significant changes in fasting levels of CRP, vitamin C,
284
lipids (total cholesterol, LDL-cholesterol and triglycerides), or platelet reactivity (data not shown)
285
over the 2 week periods. However, there was a significant decrease in plasma HDL-cholesterol
286
(p=0.019).
287
288
289
10
290
Acute effects of apple puree consumption (d15 – t = 0, 2, 6 & 24 h)
291
Platelet reactivity: Neither apple puree treatment attenuated collagen-induced integrin-β3 or p-
292
selectin expression during the acute phase of ingestion (data not shown). However, HF-apple puree
293
did significantly attenuate epinephrine-induced integrin-β3 expression 2 h after ingestion
294
(p=0.0242) and ADP- induced integrin-β3 expression 6 h after ingestion (p=0.0005) (Fig 3A). After
295
24 h the latter was found to be significantly higher than at baseline (p=0.0136). LF-apple puree on
296
the other hand significantly attenuated both ADP and epinephrine-induced integrin-β3 expression 2
297
h (p=<0.0001 and 0.0012 respectively) and 6 h (p=<0.0001 and 0.0426 respectively) post ingestion
298
(Fig 3A). Furthermore, LF- apple puree attenuated both ADP and epinephrine-induced P-selectin
299
expression 2 h after ingestion (p=0.0027 and 0.0120 respectively) (Fig 3B). Whilst both apple
300
puree treatments were shown to be effective at attenuating platelet reactivity 2 and 6 h after
301
ingestion there were no significant differences in platelet reactivity between the two apple puree
302
treatments at either of these time points.
303
304
In this study, ASA ingestion served as a positive control and as expected significantly attenuated
305
both ADP and epinephrine-induced integrin-β3 expression throughout the acute phase of ingestion
306
(p=<0.05 for all time points) (Fig 3A). Collagen-induced integrin-β3 expression was only
307
attenuated 24 h after ingestion (p=0.0015) (data not shown). Similarly, ASA ingestion only
308
attenuated ADP-induced p-selectin expression 2 h after ingestion (p=0.0016) (Fig 3B). The acute
309
effects of apple puree ingestion were compared with the aspirin treatment at the 2 and 6 h time
310
points when the differences were most apparent. Compared with HF-apple puree, ASA was
311
significantly better at attenuating ADP-induced P-selectin expression 2 h after ingestion (p=0.0119).
312
Otherwise, no differences between apple puree and ASA were observed 2 and 6 h post
313
consumption.
314
315
Nitric oxide metabolites: There was a significant increase in plasma total NO metabolites conc. 6 h
316
after consumption of HF-apple puree (170 ± 59 compared with 129 ± 37 nmol/L p=0.0297).
317
Additionally, there was a significant increase in plasma total NO metabolites conc. 6 h after
318
consumption of LF- apple puree (176 ± 99 compared with 116 ± 32 nmol/L p=0.0388) (see Table
319
1). There were no between- treatment differences in plasma NO metabolites at any time point.
320
321
Mean 2 h plasma total monomeric catechins ((-)-epicatechin and (+)-catechin) concentrations were
322
significantly higher after ingestion of the HF-apple puree compared with LF-apple puree (mean ±
11
323
SEM: 181 ± 28 nmol/L compared with 44 ± 12 nmol/L, p=0.007), and at 6h (59 ± 10 nmol/L and
324
nd, p=0.0001). The Tmax of epicatechin of 2 h was consistent with absorption in the small intestine.
325
326
Long term effects of apple puree consumption (d 15-29)
327
Platelet reactivity: LF-apple puree significantly attenuated collagen-induced p-selectin expression
328
after 2 weeks daily ingestion (p=0.0018). No other decreases on platelet activity were observed.
329
However, a significant increase in ADP-induced integrin-β3 and P-selectin expression (p=0.0105
330
and 0.032 respectively) was observed after 2 weeks of HF-apple puree ingestion (data not shown).
331
As mentioned previously ASA ingestion served as a positive control. Only collagen-induced p-
332
selectin expression was attenuated after 2 weeks daily ingestion of ASA (p= 0.0018).
333
334
Plasma lipids, CRP, vitamin C & serum ET-1: 2 weeks daily consumption of LF-apple puree
335
significantly decreased plasma triglycerides (1.3 ± 0.4 mmol/L (d 15) compared with 1.1 ± 0.4
336
mmol/L (d 29); p=0.002). When comparing the effects of the different treatments after 14-d, plasma
337
conc. of vitamin C were significantly higher in the ASA and HF-apple puree groups compared with
338
the LF-apple puree group (p=0.046 and 0.0087). No other significant changes in plasma CRP,
339
vitamin C, lipids (total cholesterol, LDL and HDL-cholesterol) or serum ET-1 were observed (see
340
Table 1).
341
342
Urinary excretion of epicatechin metabolites.
343
Quantification of epicatechin in urine was undertaken before and after treatment with apple puree
344
only. All samples were treated with a mixture of β-glucuronidase and aryl-sulfatase enzymes prior
345
to extraction and analysis, and as a result the only forms of catechin/epicatechin quantified were the
346
aglycones and their methylated derivatives. In twenty one of the 25 participants no epicatechin was
347
detected in the urine samples prior to treatment at d 15, suggesting good overall compliance with
348
the 2 week low flavanol diet. During the acute phase of treatment (0 - 24 h), mean urinary excretion
349
of epicatechin was 320 ± 292 µg and 2191 ± 2510 µg; low and high flavanol apple puree treatments
350
respectively. After 14-d of treatment mean epicatechin excretion in the 24 h urine collection was
351
lower in the LF-apple puree group (213 ± 146 µg), whereas in the HF-apple puree group it
352
remained the same (2208
353
significantly higher after treatment with HF-apple puree compared with LF-apple puree during both
354
the acute and long term phases of consumption (p=<0.0001).
± 2251 µg). As expected urinary excretion of epicatechin was
355
356
12
357
Discussion
358
In this study, two hypotheses were tested: (1) consumption of flavanol-containing apple puree
359
would attenuate platelet reactivity and increase plasma concentration of nitric oxide metabolites, (2)
360
apple puree with a significantly higher content of flavanols would have a greater effect than apple
361
puree with lower flavanol content. Our data show that consumption of both apple purees attenuated
362
platelet reactivity, as measured by surface expression of two platelet activation markers (the cell
363
adhesion molecule P-selectin and the integrin-β3) in agonist-treated whole blood samples post-
364
consumption compared to baseline, but that the high flavanol apple puree was no more effective
365
than the low-flavanol apple puree. Further, our data clearly show that the effects of apple puree
366
consumption on platelet reactivity are transient, i.e. although effects were observed at 2 h and 6 h
367
post consumption, they were not observed 24 h post-consumption nor after a 14 d period of daily
368
consumption. The changes in platelet reactivity were observed in subjects whose flavanol intake
369
had been restricted prior to the 2-week period of consumption of apple puree, and it is not possible
370
to predict whether or not similar effects would be observed in subjects whose flavanol intake had
371
not been restricted.
372
Previous human and animal intervention studies using flavanol rich extracts and foods such
(4,15,18,25)
(13,26)
373
as cocoa
374
decrease platelet aggregation. Evidence for these anti-aggregatory effects appears to be strongest for
375
cocoa which, consistent with our study, decreases ADP and epinephrine activated platelet
376
aggregation within 2-6 hours post dosing
377
attributable to a decrease in the expression of P-selectin and integrin-β3. By contrast such effects
378
are not seen after consumption of foods and beverages rich in other flavonoids such as orange and
379
grapefruit juice (13,27) and white wine (28).
380
, and grapes
have also shown that flavanols improve vascular function and
Contrary to other published reports
(15,18)
. In these studies the effects were shown to be
(4)
investigating the effects of flavanol containing foods
381
and beverages on platelet activity, we largely did not observe an effect of apple puree on platelet
382
reactivity following 2 weeks daily consumption. In our study long term assessment of platelet
383
function was undertaken following a 12 h overnight fast. Monomeric flavanols are absorbed rapidly
384
with maximum plasma concentrations peaking at about 2 hours after consumption and returning to
385
baseline levels within 24h
386
function) of the circulating flavanols and their phase-2 metabolites are only observed in the first few
387
hours after consumption, and would not be observed in fasted volunteers at the end of the 2 week
388
treatment period. Neither did we observe attenuation of collagen-induced platelet activity in
389
response to acute or long term consumption of apple puree. It is well recognized that different
390
platelet agonists trigger different activation pathways in response to different stimuli. For example,
(29)
. Therefore, it is likely that the biological effects (in terms of platelet
13
(4)
391
Murphy et al
392
not arachidonic acid-induced platelet aggregation after feeding healthy subjects cocoa flavanols and
393
procyanidins. Conversely, Bydlowski and colleagues
394
arachidonic acid-induced platelet aggregation after supplementing animals with a coffee extract but
395
observed no effects on collagen-induced platelet aggregation.
report a significant decrease in ADP and collagen-induced platelet aggregation but
(30)
, report a significant decrease in ADP and
396
It has been reported that the major metabolites arising as a result of colonic fermentation of
397
dietary flavanols are the valerolactones, with the major reported forms being 5-(3’,4’-
398
dihydroxyphenyl)-γ-valerolactone and 5-(3-methoxy-4’-hydroxyphenyl)-γ-valerolactone
399
appear in plasma within about 4-6 h of consumption of flavanol-rich foods/extracts and are excreted
400
over the next 48 h
401
from this study, our data are consistent with hydroxyphenyl-γ-valerolactones not significantly
402
affecting platelet reactivity at the concentrations achieved from a 25-100 mg doses in apple purees
403
because no effects were observed in fasted samples. Further, these observations imply that the
404
effects observed in the acute phase are likely due to flavanols such as (-)-epicatechin and (+)-
405
catechin and their phase-2 conjugates rather than the hydroxyphenyl-γ-valerolactones.
406
(31)
. These
(32-34)
. Although we did not quantify valerolactones in plasma or urine samples
Not all studies using flavanol containing foods and beverages show a positive effect on
(35)
407
platelet function.
408
consumption of black tea in patients with coronary artery disease, although any small effects may
409
have been masked by daily aspirin consumption taken as part of the subjects therapeutic regime.
410
Regardless, similar results have been reported by Hodgson et al
411
cups/day of black tea over a 4 week period. In a separate study the same authors report no effect of
412
black tea on platelet aggregation during the acute phase of consumption
413
for the difference in these findings compared to those described earlier is that grapes, cocoa and
414
apples contain monomeric catechins which in black tea are converted to theaflavins and
415
thearubigins during the fermentation process
416
platelet function.
Duffy et al
for example, report no such effects after acute or chronic
(38)
(36)
after feeding healthy subjects 5
(37)
. One such explanation
thus reducing the biological impact in terms of
417
The data presented here show that consumption of both the high and low flavanol apple
418
purees transiently attenuates platelet reactivity and increases in plasma nitric oxide metabolites.
419
Indeed, the magnitudes of the changes in these markers in response to the apple purees were similar.
420
These findings prove our original hypothesis wrong. There are at least 2 plausible explanations of
421
this finding. First, it is possible that other components in the apple puree (not the flavanols) are
422
responsible for the effects seen in platelet reactivity and NO metabolite bioavailability. For
423
example, other apple polyphenols such as chlorogenic acid, quercetin, and phloretin could have
424
caused the physiological changes that were observed. It is worth noting that although the
14
425
chlorogenic acid content of the high flavanol apple puree was significantly higher (3.6-fold) than
426
the low flavanol apple puree, this was not associated with any significant changes in the biomarkers
427
assessed in this study. Although there is increasing evidence that other non-flavanol polyphenols
428
such as quercetin can improve vascular bioavailability of NO
429
substantially higher doses than consumed in the study reported here. A second explanation is that
430
the lower dose of apple puree flavanols (25 mg) was sufficient to induce the maximal decrease in
431
platelet reactivity and increase in NO metabolite bioavailability, and increased doses above 25 mg
432
do not increase the magnitude of the response. It is interesting to note that a systematic analysis of
433
all the data from studies investigating the effects of cocoa flavanols on blood pressure with
434
subsequent fitting of the epicatechin dose and response data to a non-linear meta-regression model
435
(and applying a Bayesian approach using Markov chain Monte Carlo methods to appropriately deal
436
with the non-linearity), indicated that 25 mg of epicatechin was sufficient to induce maximal
437
responses in systolic and diastolic blood pressure
438
significantly different between the apple puree consumption groups. Phloretin, a dihydrochalcone
439
typically found in apples, was present in plasma after apple puree consumption, but not after aspirin
440
consumption, confirming the compliance to apple puree consumption (results not shown). Another
441
possible explanation of our results is that non-phenolic components of apple purees are responsible
442
for the observed effects. For example, apples contain nitrates which are a potential source of nitric
443
oxide that can affect the endothelium
444
typical levels are around 0.3 mg/100g(41). In our study, participants ingested 230 g of apple puree,
445
thus providing the equivalent of ~ 0.69 mg nitrates. Other human intervention studies have
446
demonstrated a positive effect of nitrate containing foods/beverages on cardiovascular health, but
447
the effective doses delivered in these studies (≥350 mg nitrate/day) far exceed that used in our study
448
(42,43)
(9)
, this has only been shown for
(39)
. The plasma levels of epicatechin were
(40)
. We did not measure nitrates in the apple purees, but
.
449
Aspirin is the most commonly administered non-steroidal anti-inflammatory drug that
450
significantly attenuates platelet activity as measured by surface expression of platelet receptors
451
integrin-β3
452
integrin-β3 expression at 2, 6 and 24h post consumption. The effects of aspirin on P-Selectin
453
expression were less profound, with attenuation occurring only 2h after consumption and only in
454
response to ADP. Since the main effects of apple puree on platelet reactivity occurred 2 and 6h after
455
consumption, we compared the differences in ADP and epinephrine–induced platelet reactivity
456
between each of the treatments at these time points. P-selectin expression after ex-vivo stimulation
457
of platelets with ADP was significantly lower in the aspirin treated group compared with the high
458
flavanol apple puree group 2h after consumption. Otherwise, we report no significant differences
(44)
. In our study, aspirin significantly attenuated ADP and epinephrine-stimulated
15
459
between the apple puree treatments and aspirin. This is consistent with the findings of Pearson et al
460
(18)
461
epinephrine activated platelet aggregation similar to that achievable with a low dose of aspirin (81
462
mg).
who after feeding healthy subjects a flavanol rich cocoa beverage, report a decrease in ADP and
463
In addition to inhibiting platelet activity, flavanols have been reported to affect vascular
464
function through other mechanisms. The endothelium plays an integral role in maintaining vascular
465
homeostasis which is partly achieved by balancing the production of vasodilators such as
466
endogenous nitric oxide with vasoconstrictors such as endothelin-1. Previous studies have reported
467
both acute and chronic increases in endothelium-dependant vasodilation after consumption of
468
flavanol rich foods and beverages, as measured by flow mediated dilation (FMD) of the brachial
469
artery
470
beverages have been shown to increase NO production (48,49). Nitric oxide inhibits the production of
471
endothelin-1. In our study plasma NO metabolite levels were significantly higher 6 h after subjects
472
consumed apple puree both low and high in epicatechin, but the effect was short-lived and no
473
change in serum concentrations of ET-1 were seen. Plasma cholesterol and C-reactive protein are
474
predictive biomarkers of cardiovascular disease risk. Flavanol rich foods and beverages have been
475
reported to lower plasma total and LDL-cholesterol
476
not show an effect on plasma lipids after acute or long term intake of apple puree consumption.
477
Consistent with other studies
478
concentrations.
(45-48)
. FMD is almost exclusively mediated by NO
(6,51)
(50)
(10)
and flavanol containing foods and
but contrary to these findings our study did
we also report no effects of a flavanol rich food on plasma CRP
479
In conclusion, our data show that consumption of apple puree containing two different
480
amounts of flavanols transiently attenuated ex vivo integrin-β3 and P-selectin expression in healthy
481
subjects. Further, we provide evidence that apple puree containing higher levels of flavanols are no
482
more effective at attenuating platelet reactivity than apple puree containing low levels of flavanols.
483
These data indicate that consumption of apples may have the potential to reduce cardiovascular
484
disease risk by reducing agonist-induced platelet aggregation.
485
486
Acknowledgements
487
This research was supported by an EU grant (Food–CT–2004–513960) and funding from the UK
488
Biotechnology and Biological Sciences Research Council (BBSRC).
489
P.A.K., GW & B.T. designed the research; A.G., A.C., S.S., and W.J.H. conducted the research and
490
laboratory measurements; D.V., P.C.H., M.J.R., & R.N. conducted laboratory measurements; J.R.D.
491
performed statistical analyses; W.J.H. and P.A.K. wrote the manuscript. All authors read and
492
approved the final manuscript. P.A.K. had primary responsibility for final content. P.A.K has
16
493
received research funding from Coressence for a BBSRC CASE studentship and is an independent
494
member of the Coressence Science Board. P.C.H is an independent member of the Scientific
495
Advisory Board of Barry Callebaut. G.W, R.N & M.J.R currently work or have worked for Nestle.
496
A.G, W.H, A.C, S.S, B.T, J.R.D, D.P.V have no conflicts of interest.
497
498
17
Literature cited
1. Buijsse B, Feskens EJM, Kok FJ et al. (2006) Cocoa intake, blood pressure, and
cardiovascular mortality: the Zutphen Elderly Study. Arch Intern Med 166, 411-417.
2. Ding EL, Hutfless SM, Ding X et al. (2006) Chocolate and prevention of cardiovascular
disease: a systematic review. Nutr Metab (Lond) 3, 2.
3. Cooper KA, Donovan JL, Waterhouse AL et al. (2008) Cocoa and health: a decade of
research. Br J Nutr 99, 1-11.
4. Murphy KJ, Chronopoulos AK, Singh I et al. (2003) Dietary flavanols and procyanidin
oligomers from cocoa (Theobroma cacao) inhibit platelet function. Am J Clin Nutr 77,
1466-1473.
5. Rein D, Paglieroni TG, Pearson DA et al. (2000) Cocoa and wine polyphenols modulate
platelet activation and function. J Nutr 130, 2120S-2126S.
6. Mathur S, Devaraj S, Grundy SM et al. (2002) Cocoa products decrease low density
lipoprotein oxidative susceptibility but do not affect biomarkers of inflammation in humans.
J Nutr 132, 3663-3667.
7. Loke WM, Hodgson JM, Proudfoot JM et al. (2008) Pure dietary flavonoids quercetin
and (-)-epicatechin augment nitric oxide products and reduce endothelin-1 acutely in
healthy men. Am J Clin Nutr 88, 1018-1025.
8. Hooper L, Kroon PA, Rimm EB et al. (2008) Flavonoids, flavonoid-rich foods, and
cardiovascular risk: a meta-analysis of randomized controlled trials. Am J Clin Nutr 88, 3850.
9. Heiss C, Dejam A, Kleinbongard P et al. (2003) Vascular effects of cocoa rich in flavan3-ols. Jama-Journal of the American Medical Association 290, 1030-1031.
10. Joannides R, Haefeli WE, Linder L et al. (1995) Nitric oxide is responsible for flowdependent dilatation of human peripheral conduit arteries in vivo. Circulation 91, 13141319.
11. Fisher ND, Hughes M, Gerhard-Herman M et al. (2003) Flavanol-rich cocoa induces
nitric-oxide-dependent vasodilation in healthy humans. J Hypertens 21, 2281-2286.
12. Ruggeri ZM (2002) Platelets in atherothrombosis. Nat Med 8, 1227-1234.
13. Keevil JG, Osman HE, Reed JD et al. (2000) Grape juice, but not orange juice or
grapefruit juice, inhibits human platelet aggregation. J Nutr 130, 53-56.
14. Kang WS, Lim IH, Yuk DY et al. (1999) Antithrombotic activities of green tea catechins
and (-)-epigallocatechin gallate. Thromb Res 96, 229-237.
15. Rein D, Paglieroni TG, Wun T et al. (2000) Cocoa inhibits platelet activation and
function. Am J Clin Nutr 72, 30-35.
16. Pace-Asciak CR, Rounova O, Hahn SE et al. (1996) Wines and grape juices as
modulators of platelet aggregation in healthy human subjects. Clin Chim Acta 246, 163182.
17. Freedman JE, Parker C, 3rd, Li L et al. (2001) Select flavonoids and whole juice from
purple grapes inhibit platelet function and enhance nitric oxide release. Circulation 103,
2792-2798.
18. Pearson DA, Paglieroni TG, Rein D et al. (2002) The effects of flavanol-rich cocoa and
aspirin on ex vivo platelet function. Thromb Res 106, 191-197.
19. Schewe T, Sadik C, Klotz LO et al. (2001) Polyphenols of cocoa: inhibition of
mammalian 15-lipoxygenase. Biol Chem 382, 1687-1696.
20. Schramm DD, Wang JF, Holt RR et al. (2001) Chocolate procyanidins decrease the
leukotriene-prostacyclin ratio in humans and human aortic endothelial cells. Am J Clin Nutr
73, 36-40.
21. Chun OK, Chung SJ, Song WO (2007) Estimated dietary flavonoid intake and major
food sources of U.S. adults. J Nutr 137, 1244-1252.
18
22. Feelisch M, Rassaf T, Mnaimneh S et al. (2002) Concomitant S-, N-, and hemenitros(yl)ation in biological tissues and fluids: implications for the fate of NO in vivo. Faseb
J 16, 1775-1785.
23. Renouf M, Guy PA, Marmet C et al. (2010) Measurement of caffeic and ferulic acid
equivalents in plasma after coffee consumption: small intestine and colon are key sites for
coffee metabolism. Mol Nutr Food Res 54, 760-766.
24. Saha S, Hollands W, Needs PW et al. (2012) Human O-sulfated metabolites of (-)epicatechin and methyl-(-)-epicatechin are poor substrates for commercial aryl-sulfatases:
implications for studies concerned with quantifying epicatechin bioavailability. Pharmacol
Res 65, 592-602.
25. Holt RR, Schramm DD, Keen CL et al. (2002) Chocolate consumption and platelet
function. Jama-Journal of the American Medical Association 287, 2212-2213.
26. Polagruto JA, Gross HB, Kamangar F et al. (2007) Platelet reactivity in male smokers
following the acute consumption of a flavanol-rich grapeseed extract. J Med Food 10, 725730.
27. Osman HE, Maalej N, Shanmuganayagam D et al. (1998) Grape juice but not orange
or grapefruit juice inhibits platelet activity in dogs and monkeys. J Nutr 128, 2307-2312.
28. Folts JD (1998) Antithrombotic potential of grape juice and red wine for preventing
heart attacks. Pharmaceutical Biology 36, 21-27.
29. Richelle M, Tavazzi I, Enslen M et al. (1999) Plasma kinetics in man of epicatechin
from black chocolate. Eur J Clin Nutr 53, 22-26.
30. Bydlowski SP, Yunker RL, Rymaszewski Z et al. (1987) COFFEE EXTRACTS INHIBIT
PLATELET-AGGREGATION INVIVO AND INVITRO. International Journal for Vitamin and
Nutrition Research 57, 217-223.
31. Urpi-Sarda M, Monagas M, Khan N et al. (2009) Targeted metabolic profiling of
phenolics in urine and plasma after regular consumption of cocoa by liquid
chromatography-tandem mass spectrometry. J Chromatogr A 1216, 7258-7267.
32. Del Rio D, Calani L, Cordero C et al. (2010) Bioavailability and catabolism of green tea
flavan-3-ols in humans. Nutrition 26, 1110-1116.
33. van der Hooft JJJ, de Vos RCH, Mihaleva V et al. (2012) Structural Elucidation and
Quantification of Phenolic Conjugates Present in Human Urine after Tea Intake. Analytical
Chemistry 84, 7263-7271.
34. Grimm T, Skrabala R, Chovanova Z et al. (2006) Single and multiple dose
pharmacokinetics of maritime pine bark extract (pycnogenol) after oral administration to
healthy volunteers. BMC Clin Pharmacol 6, 4.
35. Duffy SJ, Vita JA, Holbrook M et al. (2001) Effect of acute and chronic tea
consumption on platelet aggregation in patients with coronary artery disease. Arterioscler
Thromb Vasc Biol 21, 1084-1089.
36. Hodgson JM, Puddey IB, Mori TA et al. (2001) Effects of regular ingestion of black tea
on haemostasis and cell adhesion molecules in humans. Eur J Clin Nutr 55, 881-886.
37. Hodgson JM, Puddey IB, Burke V et al. (2002) Acute effects of ingestion of black tea
on postprandial platelet aggregation in human subjects. Br J Nutr 87, 141-145.
38. Menet MC, Sang S, Yang CS et al. (2004) Analysis of theaflavins and thearubigins
from black tea extract by MALDI-TOF mass spectrometry. J Agric Food Chem 52, 24552461.
39. Ellinger S, Reusch A, Stehle P et al. (2012) Epicatechin ingested via cocoa products
reduces blood pressure in humans: a nonlinear regression model with a Bayesian
approach. Am J Clin Nutr 95, 1365-1377.
40. Kleinbongard P, Dejam A, Lauer T et al. (2006) Plasma nitrite concentrations reflect
the degree of endothelial dysfunction in humans. Free Radic Biol Med 40, 295-302.
19
41. Hord NG, Tang Y, Bryan NS (2009) Food sources of nitrates and nitrites: the
physiologic context for potential health benefits. Am J Clin Nutr 90, 1-10.
42. Siervo M, Lara J, Ogbonmwan I et al. (2013) Inorganic nitrate and beetroot juice
supplementation reduces blood pressure in adults: a systematic review and meta-analysis.
J Nutr 143, 818-826.
43. Webb AJ, Patel N, Loukogeorgakis S et al. (2008) Acute blood pressure lowering,
vasoprotective, and antiplatelet properties of dietary nitrate via bioconversion to nitrite.
Hypertension 51, 784-790.
44. McKenzie ME, Malinin AI, Bell CR et al. (2003) Aspirin inhibits surface glycoprotein
IIb/IIIa, P-selectin, CD63, and CD107a receptor expression on human platelets. Blood
Coagul Fibrinolysis 14, 249-253.
45. Hodgson JM, Puddey IB, Burke V et al. (2002) Regular ingestion of black tea improves
brachial artery vasodilator function. Clin Sci (Lond) 102, 195-201.
46. Heiss C, Kleinbongard P, Dejam A et al. (2005) Acute consumption of flavanol-rich
cocoa and the reversal of endothelial dysfunction in smokers. J Am Coll Cardiol 46, 12761283.
47. Engler MB, Engler MM, Chen CY et al. (2004) Flavonoid-rich dark chocolate improves
endothelial function and increases plasma epicatechin concentrations in healthy adults. J
Am Coll Nutr 23, 197-204.
48. Schroeter H, Heiss C, Balzer J et al. (2006) (-)-Epicatechin mediates beneficial effects
of flavanol-rich cocoa on vascular function in humans. Proc Natl Acad Sci U S A 103,
1024-1029.
49. Matsuo S, Nakamura Y, Takahashi M et al. (2001) Effect of red wine and ethanol on
production of nitric oxide in healthy subjects. Am J Cardiol 87, 1029-1031: A1026.
50. Jeong WS, Kong ANT (2004) Biological properties of monomeric and polymeric
catechins: Green tea catechins and procyanidins. Pharmaceutical Biology 42, 84-93.
51. Alexopoulos N, Vlachopoulos C, Aznaouridis K et al. (2008) The acute effect of green
tea consumption on endothelial function in healthy individuals. Eur J Cardiovasc Prev
Rehabil 15, 300-305.
20
Figure legends
Fig 1 Flow diagram of the progress of subjects through 3 phases of a randomized crossover design
study.
Fig 2. Time points for blood and urinary assessment on a randomized 3-phase crossover design
study. During each period of intervention assessments were made before and after 2 week low
flavanol diet (days 1-15), and after acute (day 15 – t= 0, 2, 6 & 24h) and long term (day 15-29)
consumption of apple puree high and low in epicatechin (100 and 25 mg respectively) and aspirin
(75mg). There were no assessments made on ET-1 or NO metabolite levels during the aspirin
phase.
Fig 3. Acute effects of aspirin (
(
), high flavanol apple puree (
) and low flavanol apple puree
) on ADP and epinephrine induced integrin-β3 (A) and P-selectin (B) expression. Data are
presented as box plots and expressed as % change in integrin-β3 and P-selectin expression 2h, 6h
and 24h post ingestion of treatment on d15. * Significantly different from 0h within the treatment
group.
21
Figure 1
Assessed for eligibility
(n= 47)
Enrolment
Excluded (n=19)
n=17 did not meet inclusion
criteria
n=2 declined to participate
post enrolment
Analysis
Follow-up
Allocation
Randomized to treatments
(n=28)
Aspirin intervention
LF-apple puree intervention
HF-apple puree intervention
Completed (n=25)
Completed (n=25)
Completed (n=26)
Did not complete (n=3)
Did not complete (n=3)
Did not complete (n=2)
Lost to follow-up (n=3)
1 developed anaemia
1 developed an allergy
1 aspirin ingestion contra-indicated
Analysed (n=25)
Excluded from analysis (n=3)
1 withdrawn after only completing day 1 of first allocated
intervention (therefore no data available)
2 did not complete all 3 interventions and therefore excluded on
grounds of statistical balance (analysis undertaken on a per
protocol basis which is appropriate for this study)
22
Figure 2
23
Figure 3
24
Table 1. Plasma concentrations of CVD risk markers before and after consumption of 230 g apple puree high and low in epicatechin (100 and 25 mg respectively)
and aspirin (75mg).
D1
Biomarker
High flavanol apple puree
CRP (mg/L)
Vitamin C (μmol/L)
Total cholesterol (mmol/L)
HDL Cholesterol (mmol/L)
LDL Cholesterol (mmol/L)
Triglycerides (mmol/L)
ET-1 (ng/L)
NO metabolites (nmol/L)
Low flavanol apple puree
CRP (mg/L)
Vitamin C (μmol/L)
Total cholesterol (mmol/L)
HDL Cholesterol (mmol/L)
LDL Cholesterol (mmol/L)
Triglycerides (mmol/L)
ET-1 (ng/L)
NO metabolites (nmol/L)
Aspirin
CRP (mg/L)
Vitamin C (μmol/L )
Total cholesterol (mmol/L)
HDL Cholesterol (mmol/L)
LDL Cholesterol (mmol/L)
Triglycerides (mmol/L)
Mean
SD
3.38
67.5
5.35
1.37
3.44
1.24
2.58
41.5
0.77
0.34
0.73
0.45
3.75
59.1
5.11
1.39
3.10
1.16
4.05
61.9
5.32
1.42
3.34
1.23
3.38
23.9
0.91
0.32
0.89
0.46
5.31
25.3
0.99
0.37
0.85
0.53
D15
0h
Mean SD
3.75
61.7
5.14
1.36
3.29
1.19
0.67
129a
3.69
19.1
0.89
0.33
0.80
0.47
0.48
37
3.07
60.8
5.16
1.35
3.24
1.25a
0.71
116a
2.73
18.2
0.89
0.32
0.79
0.44
0.51
32
2.80
73.2
5.22
1.34
3.31
1.23
1.87
49.0
0.92
0.29
0.81
0.46
D29
2h
Mean SD
173
129
90
49
6h
Mean SD
170b
176b
59
98
24h
Mean SD
Mean
SD
147
71
3.86
84.2A
5.07
1.30
3.28
1.19
0.71
150
3.44
76.0
0.81
0.31
0.66
0.49
0.56
102
79
2.89
63.0 AB
5.06
1.54
3.09
1.10b
0.71
127
2.70
36.3
0.80
0.87
0.79
0.44
0.47
69
3.49
66.4 B
5.09
1.32
3.19
1.15
2.48
20.7
0.86
0.28
0.81
0.46
150
Values are mean ± SD. n=25. Data were analysed by 1-way ANOVA.
AB
Within a column means with unlike upper case superscript letters were significantly different (p<0.05)
Ab
Within a row means with unlike lower case superscript letters were significantly different (p<0.05)
25