Mastication of almonds: Effects of lipid bioaccessibility, appetite, and hormone response

Abstract

Epidemiologic and clinical data indicate that nuts can be incorporated into the diet without compromising body weight. This has been attributed to strong satiety properties, increased resting energy expenditure, and limited lipid bioaccessibility.
The role of mastication was explored because of evidence that the availability of nut lipids is largely dependent on the mechanical fracture of their cell walls.
In a randomized, 3-arm, crossover study, 13 healthy adults (body mass index, in kg/m(2): 23.1 +/- 0.4) chewed 55 g almonds 10, 25, or 40 times. Blood was collected and appetite was monitored during the following 3 h. Over the next 4 d, all foods were provided, including 55 g almonds, which were consumed under the same chewing conditions. Complete fecal samples were collected.
Hunger was acutely suppressed below baseline (P < 0.05), and fullness was elevated above baseline longer (P < 0.05) after 40 chews than after 25 chews. Two hours after consumption, fullness levels were significantly lower and hunger levels were significantly higher after 25 chews than after 10 and 40 chews (P < 0.05). Initial postingestive glucagon-like peptide-1 concentrations were significantly lower after 25 chews than after 40 chews (P < 0.05), and insulin concentrations declined more rapidly after 25 and 40 chews than after 10 chews (both P < 0.05). Fecal fat excretion was significantly higher after 10 chews than after 25 and 40 chews (both P < 0.05). All participants had higher fecal energy losses after 10 and 25 chews than after 40 chews (P < 0.005).
The results indicate important differences in appetitive and physiologic responses to masticating nuts and likely other foods and nutrients. This trial was registered at clinicaltrials.gov as NCT00768417.

Full text

Mastication of almonds: effects of lipid bioaccessibility, appetite, and
hormone response
1–3
Bridget A Cassady, James H Hollis, Angie D Fulford, Robert V Considine, and Richard D Mattes
ABSTRACT
Background: Epidemiologic and clinical data indicate that nuts can
be incorporated into the diet without compromising body weight.
This has been attributed to strong satiety properties, increased rest-
ing energy expenditure, and limited lipid bioaccessibility.
Objective: The role of mastication was explored because of evi-
dence that the availability of nut lipids is largely dependent on the
mechanical fracture of their cell walls.
Design: In a randomized, 3-arm, crossover study, 13 healthy adults
(body mass index, in kg/m
2
: 23.1 60.4) chewed 55 g almonds 10,
25, or 40 times. Blood was collected and appetite was monitored
during the following 3 h. Over the next 4 d, all foods were provided,
including 55 g almonds, which were consumed under the same
chewing conditions. Complete fecal samples were collected.
Results: Hunger was acutely suppressed below baseline (P,0.05),
and fullness was elevated above baseline longer (P,0.05) after 40
chews than after 25 chews. Two hours after consumption, fullness
levels were significantly lower and hunger levels were significantly
higher after 25 chews than after 10 and 40 chews (P,0.05). Initial
postingestive glucagon-like peptide-1 concentrations were signifi-
cantly lower after 25 chews than after 40 chews (P,0.05), and
insulin concentrations declined more rapidly after 25 and 40 chews
than after 10 chews (both P,0.05). Fecal fat excretion was sig-
nificantly higher after 10 chews than after 25 and 40 chews (both
P,0.05). All participants had higher fecal energy losses after 10
and 25 chews than after 40 chews (P,0.005).
Conclusion: The results indicate important differences in appetitive
and physiologic responses to masticating nuts and likely other foods
and nutrients. This trial was registered at clinicaltrials.gov as
NCT00768417. Am J Clin Nutr 2009;89:794–800.
INTRODUCTION
The high prevalence of obesity in the United States (1) has
prompted recommendations for adherence to low-energy-dense
diets in an effort to decrease energy intake and manage body
weight (2, 3). Despite the lack of conclusive evidence supporting
the link between energy-dense diets and body weight (4), such
diets exclude most nuts. However, epidemiologic studies in-
dicate an inverse association between the frequency of nut
consumption and body mass index (BMI; in kg/m
2
) (5, 6).
Additionally, clinical trials have shown little or no change in
body weight with regular intake of nuts in free-living pop-
ulations (7–11). This issue is important because there is a qual-
ified health claim linking daily consumption of nuts with
reduced cardiovascular disease risk (12), which has resulted in
an enhanced interest in promoting nut intake. The claim is based
on evidence that nut consumption improves blood lipid profiles
(5, 6, 13–15), but other benefits have also been noted, such as
moderation of postprandial glycemia, reduced risk of diabetes
(16–19) and cancer (20–22), and improved bone health (17).
Much of the energy contributed by nuts is offset by com-
pensatory reductions in energy intake from other sources (8).
However, an additional mechanism responsible for the less-than-
predicted influence of nuts on body weight stems from an esti-
mated 10%–20% of the energy from nuts being lost in the stool
(23). This has been attributed to the resistance of nut paren-
chyma cell walls to microbial and enzyme degradation in the
gastrointestinal (GI) tract (24, 25). Consequently, lipids that are
not liberated through the mechanical disruption of the cell walls
are inaccessible for absorption in the gut. This raises the pos-
sibility that masticatory efficiency influences energy balance
through changes in lipid availability.
The role that masticatory efficiency plays in energy balance is
complex. Increased chewing could liberate more lipids from the
nut and thereby increase the amount of energy available to the
body, which contributes to positive energy balance. In contrast,
the increased presence of lipids in the small intestine may result
in an increased secretion of several hormones, such as chole-
cystokinin (CCK) (26, 27), glucagon-like peptide-1 (GLP-1)
(27), and peptide YY (PYY) (28). Higher plasma concentrations
of CCK (29), GLP-1 (30), and PYY (31) are associated with
greater sensations of satiety. Consequently, the additional amount
of energy available from the increased liberation of lipids may be
offset by a stronger satiety response.
The greater oral mechanical effort required to prepare whole
nuts for deglutition may enhance satiety through neural mech-
anisms as well. Chewing is a key stimulus of cephalic phase
1
From the Department of Foods and Nutrition, Purdue University, West
Lafayette, IN (BAC, JHH, and RDM), and the Department of Medicine,
Indiana University, Indianapolis, IN (ADF and RVC).
2
Supported by a research grant from the Almond Board of California,
which also donated the almonds.
3
Reprints not available. Address correspondence to RD Mattes, Purdue
University, Department of Foods and Nutrition, Stone Hall, Room 212, 700
W State Street, West Lafayette, IN 47907-2059. E-mail: mattes@purdue.edu.
Received July 9, 2008. Accepted for publication December 4, 2008.
First published online January 14, 2009; doi: 10.3945/ajcn.2008.26669.
794 Am J Clin Nutr 2009;89:794–800. Printed in USA. Ó2009 American Society for Nutrition
Downloaded from https://academic.oup.com/ajcn/article/89/3/794/4596705 by guest on 13 May 2021

responses (32, 33), and sensory stimulation may promote the
release of appetitive hormones such insulin (34), ghrelin (35),
CCK (36), PYY (37), and GLP-1 (38). Furthermore, studies in
rats suggest that mastication enhances satiation through hista-
minergic activation of the ventromedial hypothalamus and par-
aventricular nucleus (39). The present study sought to explore
the effects of mastication efficiency on lipid bioaccessibility,
satiety, and hormone responses.
SUBJECTS AND METHODS
Subjects
Participants were recruited starting in April 2006 via public
advertisements. Eligibility was established through completion
of screening questionnaires eliciting health and demographic
information. To be eligible, subjects had to be nonsmokers, have
a BMI of 20–25, be between 18 and 50 y of age, have a full set of
healthy teeth, not be pregnant or lactating, have low dietary
restraint (3-factor eating questionnaire restraint score 13) (40),
have no allergy to nuts, have no endocrine or eating disorders,
be weight stable (,3 kg change over the past 3 mo), and not be
taking medications likely to confound study outcomes. All par-
ticipants signed an informed consent form approved by the In-
stitutional Review Board and received monetary compensation.
Experimental design and procedures
Testing sessions, appetitive ratings, and blood collection
The study followed a crossover design with 3 treatment periods
of 4 consecutive days. There was a minimum of a 1-wk washout
between treatment periods. On day 1 of each treatment period,
participants reported to the laboratory in the morning after an 8-h
fast. They were required to rate their appetitive sensations using
a visual analogue scale (VAS) presented on a personal data as-
sistant. Standard appetite questions, as described by Hill and
Blundell (41), were used. The VAS had end anchors ranging
from ‘‘not at all’’ to ‘‘extremely’’ and questions such as, ‘‘How
strong is your feeling of hunger (fullness, desire to eat, etc.)
right now?’’
After completion of the VAS, an indwelling catheter was
inserted in a vein in the antecubital space of the arm, and
a baseline blood sample was taken. The participant was then
presented with 55 g (’2 oz; 1324 kJ, 27 g fat) (42) of raw, whole
almonds. Depending on the treatment, participants chewed the
almonds in 5-g portions (’4 almonds) 10, 25, or 40 times before
swallowing. A 15-min time allotment was given to consume the
almonds.
Immediately after almond consumption (time point ¼0),
a 15-mL blood sample was drawn. Blood samples were drawn 15,
30, 45, 60, 90, 120, and 180 min after almond consumption. All
samples were collected into EDTA-coated evacuated tubes,
immediately cooled on ice, and transferred to a refrigerated
centrifuge for separation of plasma before storage at 280°C.
Active plasma ghrelin, total GLP-1, and PYY3-36 were measured
with commercially available radioimmunoassay (RIA) kits
(GHRA-88HK, GLP1T-36HK, and PYY-67HK; Millipore,
Billerica, MA). The ghrelin RIA kit had a lower and upper
detection limit of 7.8 and 2000 pg/mL, respectively. The intra-
assay CV was 6.7%, and the interassay CV was 9.6% at a sample
concentration of 138.6 pg/mL. The lower and upper detection
limits for the GLP-1 assay were 3.0 and 333 pmol/L, respectively,
the intraassay CV was 29%, and the interassay CV was 10% at
a sample concentration of 53 pmol/L. The lower detection limit
for the PYYassay was 20 pg/mL, the upper limit was 1280 pg/mL,
and the intra- and interassay CVs were 11% and 15%, re-
spectively, at a sample concentration of 84 pg/mL. Plasma
glucose and insulin concentrations were assayed with the Cobas
Integra 400 Analyzer and Elecsys 2010 Immunoassay System
(Roche Diagnostics Inc, Summerville, NJ). The insulin assay had
a lower and upper detection limit of 1.4 and 6945 pmol/L,
respectively, and the intraassay CV was 1.9% and the interassay
CV was 2.7%. Before each blood collection, participants again
completed the appetitive questions. After the last blood draw, the
catheter was removed, and the participant was served lunch.
Meal preparation and composition
To accurately determine the effect of mastication on lipid
excretion, participants were fed 3 controlled meals and a snack
each day. The 4-d-cycle menu comprised foods typical of
a Western-diet, but excluded all nuts. Standard meals provided
a mean of ’10,266 kJ and a macronutrient composition of 35%
fat, 15% protein, and 50% carbohydrate (Nutrition Data System
for Research Software 2007; University of Minnesota, Minne-
apolis, MN). Meals were eaten in the laboratory, with no addi-
tional food or beverages allowed outside of the laboratory. All
meals were prepared in the laboratory kitchen, and each portion
was weighed before serving. Participants were required to
consume all provided food and beverages. Duplicate portions of
the menu were homogenized, frozen, freeze-dried, and stored.
The samples were analyzed for gross energy by bomb calorimetry
with a Parr 1281 Bomb Calorimeter (Parr Instruments, Moline,
IL) and total fat content was measured with by automated Soxhlet
extraction with an Ankom XT15 Extraction System (Ankom
Technology, Macedon, NY).
Stool collection
On the first morning of the study, participants ingested 3 cap-
sules of green food coloring with their almond load. On the fourth
morning, participants consumed capsules with red food coloring.
Participants were instructed to collect all stool passed until the red
marker appeared. Samples were pooled by participant and treat-
ment. Fecal composites were made by the addition of 2 parts water
followed by homogenization. Aliquots of the samples were then
frozen, freeze-dried, and stored until analyzed. The energy content
of the samples was determined by bomb calorimetry with a Parr
1281 Bomb Calorimeter (Parr Instruments), and the fat content
was measured by automated Soxhlet extraction (Ankom XT15
Extraction System).
Mastication and almond particle size
On a separate visit, participants masticated 5-g almond por-
tions for a specified number of times (10, 25, or 40), but ex-
pectorated rather than swallowed. They then rinsed their mouths
with three 30-mL portions of deionized water and expectorated
any remaining almond particles. The expectorated samples were
collected through a series of 8 sieves that yielded the following
particle size ranges: .3.35, 3.35–2.00, 1.99–1.00, 0.99–0.50,
MASTICATION AND LIPID BIOACCESSIBILITY OF ALMONDS 795
Downloaded from https://academic.oup.com/ajcn/article/89/3/794/4596705 by guest on 13 May 2021

0.49–0.25, 0.24–0.125, 0.124–0.063, 0.062–0.032, and ,0.032
mm (WS Tyler, Mentor, OH). The expectorated samples were
washed with 250 mL deionized water. The water was allowed to
drain completely through all sieves, and the samples were then
dried at 54°C for 6 h. This method has previously been used to
eliminate water from similarly sized almonds (43, 44). The fully
dried samples from each individual sieve were weighed and
recorded.
Statistical analysis
Statistical analyses were performed by using the Statistical
Package for the Social Sciences (SPSS), version 15.0 (SPSS Inc,
Chicago, IL). The criterion level for statistical significance was
P,0.05 (2-tailed). Treatment effects were tested by repeated-
measures analysis of variance, followed, where appropriate, by
a post hoc Bonferroni test to correct for multiple comparisons.
All data are expressed as means 6SEMs. Sign tests were used
to examine the distributions of fecal energy and fat. Relations
between particle size, fecal output, hormone responses, and
appetitive sensations were determined by Pearson’s correlation
analyses.
RESULTS
Participant characteristics
Twenty participants signed consent forms and began the
protocol; however, individuals who failed to report to the lab-
oratory for meals (n¼2), supplied incomplete fecal collections
(n¼1) and appetite data (n¼2), or who did not finish the
protocol for personal reasons (n¼2) were excluded. Partic-
ipants included in the final analysis (n¼13; 5 women and
8 men) had a mean BMI of 23.1 60.4 (range: 19.6–24.9) and
were 24 61.8 y of age (range: 19–43 y).
Almond particle size assessment
The number of chews was negatively correlated with the total
percentage of recovered particles, as calculated by the percentage
weight of all recovered particles in the sieves relative to the total
weight of the masticated almond (r¼20.57, P,0.05). There
was a significantly higher proportion of recovered particles after
10 chews than after 25 and 40 chews (P,0.001) (Figure 1A).
A significantly lower proportion of almond particles ,3.35 mm
were recovered after 10 chews (P,0.05) (Figure 1B). Previous
work by our group with this sorting technique resulted in 95–
97% recovery when complete collections were made (43). Thus,
the balance is primarily attributable to particles ,0.032 mm and
free lipid that passed through the small screen. The recovered
particle mass ,0.032 mm was negatively correlated with the
total percentage recovery for all treatments (r¼20.91, P,
0.01). The mass of particles sized .3.35 mm was positively
correlated with the total percentage recovery for all treatments
(r¼0.70, P,0.01) and negatively correlated with particle
sizes ,0.032 mm after 10 (r¼20.65, P¼0.016) and 25
(r¼20.77, P¼0.002) chews. However, there were no sig-
nificant correlations between particle size and appetitive ratings,
fecal fat excretion, or hormone concentrations.
Fecal excretion
There was a significant main effect of chewing on fecal energy
excretion (P¼0.015). Mean energy excretion was significantly
higher after 10 chews than after 40 chews (P¼0.011) (Figure 2).
Similarly, total fecal fat excretion determined by gram weight
was significantly higher after 10 chews than after 25 (P¼0.018)
and 40 (P¼0.044) chews. Total fecal fat excretion as a per-
centage of crude fat was significantly higher after 10 than after
40 chews (P¼0.015). In comparison with the measured energy
content of the diet, there was a significant loss of energy after
10 than after 40 chews (P¼0.01). The energy loss was pri-
marily attributable to increased fecal fat excretion. Relative to
the lipid load, the proportion of lipid lost in the stool after 10
chews (43.7% 64.0%) was greater than the proportion lost after
25 (32.7% 62.7%; P¼0.006) and 40 (30.8% 64.4%; P¼
0.015) chews. In absolute terms, percentage fecal fat excretion
increased by 11.1% 63.4% after 25 chews and by 12.9% 6
4.5% after 40 chews compared with losses after 10 chews.
There was a significant inverse association between the
number of chews and energy/g dry fecal weight (r¼20.53, P,
0.05). Total dry fecal weight was significantly lower after 40
chews (175.1 610.7 g) than after 10 chews (152.8 610.8 g;
P¼0.05). Total fecal energy and fat losses were significantly
FIGURE 1. Mean (6SEM) total percentage recovery (A) and percentage
recovery by size distribution of masticated almonds (B) calculated by
percentage weight of recovered particles relative to the total weight of
almonds before and after 10, 25, or 40 chews. n¼13. Comparisons are based
on repeated-measures ANOVA with post hoc Bonferroni multiple comparison
tests. A: Different lowercase letters indicate significant differences between
the number of chews (P,0.001). B: Different lowercase letters within the
same size range indicate significant differences between the number of chews
(P,0.05).
796 CASSADY ET AL
Downloaded from https://academic.oup.com/ajcn/article/89/3/794/4596705 by guest on 13 May 2021

correlated with the 10 (r¼0.68, P¼0.011), 25 (r¼0.58, P¼
0.039), and 40 (r¼0.73, P¼0.004) chew treatments. Fecal
energy excretion was higher in 10 of 13 participants after 10
than after 25 chews (P,0.05) and in 13 of 13 participants after
10 than after 40 chews (P,0.05). All 13 participants also had
higher fecal energy losses after 25 chews than after 40 chews
(P,0.005).
Appetitive ratings
Baseline appetitive ratings were not significantly different
across treatments. Both the hunger- and fullness-by-time inter-
actions were significant (P¼0.001 and P¼0.021, respectively)
(Figure 3). Data are presented as a change from baseline.
Postprandial subjective hunger ratings (0–90 min after almond
consumption) were suppressed below baseline longer with 40
chews than with 25 chews (P¼0.031) (Figure 3A). These
ratings were significantly different from baseline after 40 chews
(P,0.05). Conversely, fullness remained elevated above
baseline longer with 40 chews than with 25 chews (P¼0.041)
and showed a trend toward significance after 10 chews
(P¼0.054) (Figure 3B). Fullness ratings were also significantly
different from baseline 60 min after almond consumption
(P,0.05). Preprandial subjective fullness ratings (2 h after
almond consumption and preceding the subsequent eating oc-
casion) were significantly lower and hunger levels were sig-
nificantly higher after 25 chews than after 10 and 40 chews
(P,0.05).
Postconsumption hormone responses
There were no significant treatment effects on active plasma
ghrelin or PYY. However, the GLP-1-by-time interaction showed
a trend toward significance (P¼0.055), and initial postingestive
concentrations of GLP-1 were lower after 25 chews than after 40
chews (P¼0.016) and significantly lower than baseline
(P,0.05) (Figure 4). Final concentrations after 10 chews were
significantly lower than baseline (P,0.05). Data are presented
as changes from baseline. Although not significant, rank or-
dering of mean treatment response values showed that GLP-1
concentrations were higher after 40 chews followed by 25 and
10 chews. Whereas no significant treatment effects were ob-
served for plasma glucose, the insulin-by-time interaction was
significant (P¼0.025) (Figure 5A, B). Post hoc analysis
showed a more precipitous decline in insulin concentration from
45 to 180 min after almond consumption after 25 and 40 chews
than after 10 chews (both P,0.05).
DISCUSSION
Accumulating evidence indicates that nut consumption may
have various health benefits, yet concern about their impact on
body weight persists. Studies indicate that the incorporation of
nuts into the diet does not promote weight gain because they are
highly satiating, their energy-yielding nutrients have limited
bioaccessibility, and they may promote energy expenditure (7–
11, 15, 24, 25, 45). However, the mechanisms that account for
their satiety and bioaccessibility properties are not clear. It was
hypothesized that masticatory function and the subsequent var-
iation in bioaccessibility may modulate these properties. The
results of the present study indicate that mastication significantly
influences energy absorption and appetitive responses, although
the 2 were not related.
FIGURE 3. Mean (6SEM) changes in hunger (A) and fullness (B)
ratings after almond consumption. The hunger- and fullness-by-time
interactions were significant (P¼0.001 and P¼0.021, respectively;
repeated-measures ANOVA with post hoc Bonferroni multiple comparison
tests). Hunger was suppressed below baseline, whereas fullness was elevated
above baseline longer with 40 chews than with 25 chews (P,0.05 and P¼
0.041, respectively). n¼13. The bracket indicates the 15-min time period
allotted for almond consumption.
N
Significant differences between 25 and
40 chews, P,0.05. *Significant differences between 25 and both 10 and 40
chews, P,0.05.
a,b,c
Significant differences between baseline and 10, 25, or
40 chews, respectively, P,0.05.
FIGURE 2. Mean (6SEM) fecal energy losses over the 4 d of stool
collection by number of chews. n¼13. Different lowercase letters indicate
significant differences between chews (P,0.05; repeated-measures ANOVA
with post hoc Bonferroni multiple comparison tests).
MASTICATION AND LIPID BIOACCESSIBILITY OF ALMONDS 797
Downloaded from https://academic.oup.com/ajcn/article/89/3/794/4596705 by guest on 13 May 2021

With controlled chewing, the present study achieved marked
differences in mechanical disruption of the physical state of the
almonds. This was documented, in part, through differences in
resulting particle sizes. The consequence was a significant dose-
response effect on fecal fat and energy excretion. This finding is
consistent with reports that lipid bioaccessibility is primarily
determined by the degree to which the cell walls of nuts are
ruptured in the oral cavity (24). Limited further extraction of
lipids occurs in the gastric and duodenal phases of digestion
(25). This is not attributable to low bioavailability because en-
ergy absorption is markedly higher with the ingestion of nut
butter or oil (45). In vitro models and mathematical modeling
predict as much as 60% of the lipids in finely ground almonds
and 85% of the lipids in 2 mm of natural almond cubes is not
bioaccessible through processes such as chewing and gastric and
duodenal digestion (25). In the only published human feeding
study permitting such a calculation, fecal fat losses increased
with nut consumption by the equivalent of 30%–40% of the lipid
provided by the nut (45). Fecal fat contributed 29% of the total
fecal energy in the control condition and 33.9% in the peanut
consumption condition. The present study did not include
a control arm to determine the fecal fat concentration after in-
gestion of the diet without almonds. Consequently, an absolute
effect on fecal fat loss could not be determined. However,
chewing the almonds 25 or 40 times led to 25.5% and 29.4%
greater reductions in fecal fat relative to chewing only 10 times.
In absolute terms, fat represented 39.8% of fecal energy ex-
cretion with 10 chews, 35.1% with 25 chews, and 35.3% with 40
chews. Although these substantive losses account, in part, for the
limited effect that nut consumption has on body weight, they are
lower than the modeling predictions. This suggests that further
digestion and absorption occur distal to the duodenum. Fer-
mentation in the colon may account for this discrepancy (46).
The increments in total energy loss were 15.6% and 20.5%
with 25 and 40 chews compared with 10 chews. Thus, fat loss was
proportional to that noted for total energy, which indicated that
the primary effect on energy balance was attributable to the fat
component of the almonds. Other studies have shown that protein
bioaccessibility is limited, comparably to that of fat (25). How-
ever, protein accounts for a markedly smaller energy component
of almonds (’14% of energy from protein compared with ’77%
from fat). Part of the fat loss may also stem from a reduction in
the absorption of free fatty acids because of the high fiber
content of almonds (47).
The 3 levels of chewing tested coincided with observations of
naturalistic eating, ie, 9–65 chews for carrots and 14–44 chews
for Brazil nuts (48). To some degree, this range reflects in-
dividual differences in chewing efficiency. Chewing a fixed
number of times results in particles of notable size difference
(49); however, ad libitum chewed, pre-swallowed particle size is
relatively consistent within an individual (48). Generally, par-
ticipants in the present study indicated that the 25-chew condi-
tion was most comfortable. Under this condition, fat accounted
for ’35% of fecal energy—a value exceeding the measured
30% of energy from fat contributed to the diet ingested.
Requiring participants to chew the almonds 40 times led to the
strongest reduction in hunger and augmentation of fullness. Such
FIGURE 5. Mean (6SEM) changes in insulin (A) and glucose (B)
concentrations after almond consumption. The insulin-by-time interaction
was significant (P¼0.02; repeated-measures ANOVA with post hoc
Bonferroni multiple comparison tests). There was a more rapid decline in
insulin concentrations from 45 to 180 min after almond consumption after 25
and 40 chews than after 10 chews (both P,0.05). n¼13. The bracket
indicates the 15-min time period allotted for almond consumption. There
was a significant difference between 10 and 25 chews (P,0.05).
FIGURE 4. Mean (6SEM) changes in glucagon-like peptide 1 (GLP-1)
concentrations after almond consumption. The GLP-1–by-time interaction
showed a trend toward significance (P¼0.055; repeated-measures ANOVA
with post hoc Bonferroni multiple comparison tests). n¼12. The bracket
indicates the 15-min time period allotted for almond consumption.
*Significant differences between 25 and 10 and 40 chews, P,0.05.
a,b
Significant differences between baseline and 10 or 25 chews,
respectively, P,0.05.
798 CASSADY ET AL
Downloaded from https://academic.oup.com/ajcn/article/89/3/794/4596705 by guest on 13 May 2021

an effect was hypothesized based on a predicted greater release of
lipid and protein with consequent secretion of satiety hormones in
response to these nutrients. GLP-1 was measured in this study
because long-chain unsaturated fatty acids are effective stimuli
for its release (50) and it elicits satiety sensations in humans
(30). Although GLP-1 concentrations were consistently higher
over the 90-min postprandial period after 40 chews, the differ-
ence was not significant. Furthermore, there was no significant
correlation between GLP-1 and particle size, fecal lipid content,
or appetite ratings. PYY and ghrelin concentrations were also
not altered by the different chewing conditions. Several alter-
native explanations are possible. First, the hypothesis may hold,
but other unmeasured peptides, such as CCK, may have had
a dominating effect. Second, mastication beyond the point of
customary oral processing of a food may diminish its palat-
ability and, consequently, attenuate hunger ratings because
a direct association between palatability and hunger has been
reported (51). Third, animal studies indicate that the mechanical
act of chewing may augment satiety through neural activation of
central satiety centers (39). Mastication is also an effective (33),
and possibly necessary (32), stimulus for cephalic phase re-
sponses, which are hypothesized to modulate appetite and en-
ergy balance (52). Cephalic or sensory-based releases of insulin
(34), ghrelin (35), CCK (36), PYY (37), and GLP-1 (38) have
been documented. Whereas there are observations consistent
with an effect of chewing in humans (53), this has not been
readily replicated with gum used as a masticatory stimulus (54).
Hunger and fullness ratings returned to baseline values more
quickly and significantly overshot this sensation level during the
later time period with 25 chews, unlike the ratings after 10 and 40
chews. The basis of this difference is not clear, but note that the
25-chew condition was regarded as most closely mimicking
customary oral processing. Whether the more extreme conditions
of 10 and 40 chews led to effects akin to novelty-induced
hypophagia (55) warrants consideration.
Almond ingestion blunts the glycemic response to foods (16),
reportedly because of their low glycemic index value or high fat
content. This property has been posited as a mechanism for the
high satiety value of nuts (19). However, the extent to which the
glycemic index or load value of foods is related to appetitive
sensations is a matter of debate (19). In the present study, al-
mond ingestion elicited weak glucose and insulin responses that
were unrelated to the level of chewing. This is most likely
reflective of their high fat and low available carbohydrate con-
tents.
The present findings do not suggest that individuals concerned
with weight management should chew their food less. Rather,
they highlight the important effects of chewing on various factors
that influence weight management (ie, lipid absorption, release of
gut peptides, and increased satiety). Such effects were primarily
observed after 25 or 40 chews. Whereas these findings are ap-
plicable to the consumption of almonds, and perhaps other nuts,
they may provide insight as to how food form may be manip-
ulated to optimize different properties for given purposes.
Consumption of whole nuts may reduce energy absorption and
augment satiety—properties useful for weight management.
Whereas whole nuts are a rich source of various nutrients,
a greater availability of vitamins, unsaturated fat, protein, and
antioxidants may be achieved with more mechanically processed
nut forms. These components may contribute to a reduced risk of
a wide array of health disorders, including cardiovascular disease,
diabetes, and cancer.
We thank William Horn for the development and adaptation of the Appe-
titeLog VAS software (US Department of Agriculture, Agricultural Research
Service, Western Human Nutrition Research Center, Davis, CA 95616).
The authors’ responsibilities were as follows—BAC: study design, testing,
sample and data analyses, and report generation; JHH: study design, testing,
and report generation; RDM: study design, data analyses, and report gener-
ation; ADF: hormone analysis; and RVC: hormone analysis and report
generation. None of the authors had a personal or financial conflict of interest.
REFERENCES
1. Centers for Disease Control and Prevention. Behavioral Risk Factor
Surveillance System Survey data. Atlanta, GA: US Department of
Health and Human Services, Centers for Disease Control and Pre-
vention, 2006.
2. Rolls B, Drewnowski A, Ledikwe JH. Changing the energy density of
the diet as a strategy for weight management. J Am Diet Assoc 2005;
105:98–103.
3. Ello-Martin JA, Roe LS, Ledikwe JH, Beach AM, Rolls BJ. Dietary
energy density in the treatment of obesity: a year-long trial comparing
2 weight-loss diets. Am J Clin Nutr 2007;85:1465–77.
4. Cox DN, Mela DJ. Determination of energy density of freely selected
diets: methodological issues and implications. Int J Obes Relat Metab
Disord 2000;24:49–54.
5. Fraser GE, Sabate
´J, Beeson WL, Strahan TM. A possible protective
effect of nut consumption on risk of coronary heart disease. The Ad-
ventist Health Study. Arch Intern Med 1992;152:1416–24.
6. Hu FB, Stampfer MJ, Manson JE, et al. Frequent nut consumption and
risk of coronary heart disease in women: prospective cohort study. BMJ
1998;317:1341–5.
7. Jaceldo-Siegl K, Sabate
´J, Rajaram S, Fraser GE. Long-term almond
supplementation without advice on food replacement induces favourable
nutrient modifications to the habitual diets of free-living individuals. Br
J Nutr 2004;92:533–40.
8. Hollis J, Mattes RD. Effect of chronic consumption of almonds on body
weight in healthy humans. Br J Nutr 2007;98:651–6.
9. Wien MA, Sabate
´JM, Ikle DN, Cole SE, Kandeel FR. Almonds vs
complex carbohydrates in a weight reduction program. Int J Obes Relat
Metab Disord 2003;27:1365–72.
10. Alper CM, Mattes RD. Effects of chronic peanut consumption on energy
balance and hedonics. Int J Obes Relat Metab Disord 2002;26:1129–37.
11. Sabate
´J, Cordero-Macintyre Z, Siapco G, Torabian S, Haddad E. Does
regular walnut consumption lead to weight gain? Br J Nutr 2005;94:
859–64.
12. US Food and Drug Administration. Qualified health claims: letter of
enforcement discretion—nuts and coronary heart disease. Rockville,
MD: US Food and Drug Administration, 2003:1–4.
13. Griel AE, Kris-Etherton PM. Tree nuts and the lipid profile: a review of
clinical studies. Br J Nutr 2006;96(suppl 2):S68–78.
14. Sabate
´J, Haddad E, Tanzman JS, Jambazian P, Rajaram S. Serum lipid
response to the graduated enrichment of a Step I diet with almonds:
a randomized feeding trial. Am J Clin Nutr 2003;77:1379–84.
15. Jenkins DJ, Kendall CW, Marchie A, et al. Dose response of almonds
on coronary heart disease risk factors: blood lipids, oxidized low-
density lipopr oteins, lipoprotein(a), ho mocysteine, and pulmonary nitr ic
oxide: a randomized, controlled, crossover trial. Circulation 2002;106:
1327–32.
16. Josse AR, Kendall CWC, Augustin LSA, Ellis PR, Jenkins DJA.
Almonds and postprandial glycemia—a dose-response study. Metabo-
lism 2007;56:400–4.
17. Griel A, Kris-Etherton P, Hilpert K, Zhao G, West S, Corwin R. An
increase in dietary n23 fatty acids decreases a marker of bone re-
sorption in humans. Available fromhttp://www.nutritionj.com/content/
6/1/2(cited 8 July 2008).
18. Jiang R, Manson JE, Stampfer MJ, Liu S, Willett WC, Hu FB. Nut and
peanut butter consumption and risk of type 2 diabetes in women. JAMA
2002;288:2554–60.
19. Rajaram S, Sabate
´J. Nuts, body weight and insulin resistance. Br J Nutr
2006;96:S79–86.
MASTICATION AND LIPID BIOACCESSIBILITY OF ALMONDS 799
Downloaded from https://academic.oup.com/ajcn/article/89/3/794/4596705 by guest on 13 May 2021

20. Davis PA, Iwahashi CK. Whole almonds and almond fractions reduce
aberrant crypt foci in a rat model of colon carcinogenesis. Cancer Lett
2001;165:27–33.
21. Gonza
´lez CA, Salas-Salvado
´J. The potential of nuts in the prevention of
cancer. Br J Nutr 2006;96:S87–94.
22. Ip C, Lisk DJ, Scimeca JA. Potential of food modification in cancer
prevention. Cancer Res 1994;54:S1957–9.
23. Sabate
´J. Nut consumption and body weight. Am J Clin Nutr 2003;
78(suppl):S647–50.
24. Ellis PR, Kendall CWC, Ren Y, et al. Role of cell walls in the bio-
accessibility of lipids in almond seeds. Am J Clin Nutr 2004;80:604–13.
25. Mandalari G, Faulks RM, Rich GT, et al. Release of protein, lipid, and
vitamin E from almond seeds during digestion. J Agric Food Chem
2008;56:3409–16.
26. Beglinger C, Degen L. Fat in the intestine as a regulator of appetite—
role of CCK. Physiol Behav 2004;83:617–21.
27. Feltrin KL, Little TJ, Meyer JH, et al. Effects of intraduodenal fatty
acids on appetite, antropyloroduodenal motility, and plasma CCK and
GLP-1 in humans vary with their chain length. Am J Physiol Regul
Integr Comp Physiol 2004;287:R524–33.
28. Feinle-Bisset C, Patterson M, Ghatei MA, Bloom SR, Horowitz M. Fat
digestion is required for suppression of ghrelin and stimulation of
peptide YY and pancreatic polypeptide secretion by intraduodenal lipid.
Am J Physiol Endocrinol Metab 2005;289:E948–53.
29. Moran TH, Kinzig KP. Gastrointestinal satiety signals II. Cholecysto-
kinin. Am J Physiol Gastrointest Liver Physiol 2004;286:G183–8.
30. Flint A, Raben A, Astrup A, Holst JJ. Glucagon-like peptide 1 promotes
satiety and suppresses energy intake in humans. J Clin Invest 1998;101:
515–20.
31. Vincent RP, le Roux CW. The satiety hormone peptide YY as a regulator
of appetite. J Clin Pathol 2008;61:548–52.
32. Teff KL, Devine J, Engelman K. Sweet taste: effect on cephalic phase
insulin release in men. Physiol Behav 1995;57:1089–95.
33. Richardson CT, Feldman M. Salivary response to food in humans and its
effect on gastric acid secretion. Am J Physiol 1986;250:G85–91.
34. Teff KL, Mattes RD, Engelman K. Cephalic phase insulin release in
normal weight males: verification and reliability. Am J Physiol Endo-
crinol Metab 1991;261:E430–6.
35. Heath RB, Jones R, Frayn KN, Robertson MD. Vagal stimulation ex-
aggerates the inhibitory ghrelin response to oral fat in humans. J En-
docrinol 2004;180:273–81.
36. Schafmayer A, Nustede R, Pompino A, Kohler H. Vagal influence on
cholecystokinin and neurotensin release in conscious dogs. Scand J
Gastroenterol 1988;23:315–20.
37. Zhang T, Uchida T, Gomez G, Lluis F, Thompson JC, Greeley GH Jr.
Neural regulation of peptide YY secretion. Regul Pept 1993;48:321–8.
38. Jang H-J, Kokrashvili Z, Theodorakis MJ, et al. Gut-expressed gustducin
and taste receptors regulate secretion of glucagon-like peptide-1. Proc
Natl Acad Sci USA 2007;104:15069–74.
39. Sakata T, Yoshimatsu H, Masaki T, Tsuda K. Anti-obesity actions of
mastication driven by histamine neurons in rats. Exp Biol Med (May-
wood) 2003;228:1106–10.
40. Stunkard AJ, Messick S. The three-factor eating questionnaire to mea-
sure dietary restraint, disinhibition and hunger. J Psychosom Res 1985;
29:71–83.
41. Hill AJ, Blundell JE. Nutrients and behaviour: research strategies for the
investigation of taste characteristics, food preferences, hunger sensations
and eating patterns in man. J Psychiatr Res 1983;17:203–12.
42. US Department of Agriculture, Agricultural Research Service. Nutrient
Data Laboratory homepage. Version current 21 February. 2008. Avail-
able from: http://www.ars.usda.gov/ba/bhnrc/ndl (cited 1 March 2008).
43. Frecka JM, Hollis JH, Mattes RD. Effects of appetite, BMI, food form
and flavor on mastication: almonds as a test food. Eur J Clin Nutr (Epub
ahead of print 18 July 2007.
44. Ow RK, Carlsson GE, Karlsson S. Relationship of masticatory man-
dibular movements to masticatory performance of dentate adults:
a method study. J Oral Rehabil 1998;25:821–9.
45. Traoret CJ, Lokko P, Cruz AC, et al. Peanut digestion and energy bal-
ance. Int J Obes (Lond) 2008;32:322–8.
46. Mallillin AC, Trinidad TP, Raterta R, Dagbay K, Loyola AS. Dietary
fibre and fermentability characteristics of root crops and legumes. Br J
Nutr 2008;100:485–8.
47. Baer DJ, Rumpler WV, Miles CW, Fahey GC Jr. Dietary fiber decreases
the metabolizable energy content and nutrient digestibility of mixed
diets fed to humans. J Nutr 1997;127:579–86.
48. Lucas PW, Luke DA. Is food particle size a criterion for the initiation of
swallowing? J Oral Rehabil 1986;13:127–36.
49. Fontijn-Tekamp FA, van der Bilt A, Abbink JH, Bosman F. Swallowing
threshold and masticatory performance in dentate adults. Physiol Behav
2004;83:431–6.
50. Paniagua JA, de la Sacristana,Sa
´nchez E, et al. A MUFA-rich diet improved
postprandial glucose, lipid and GLP-1 responses in insulin-resistant
subjects. J Am Coll Nutr 2007;26:434–44.
51. Hill AJ, Magson LD, Blundell JE. Hunger and palatability: tracking
ratings of subjective experience before, during and after the consump-
tion of preferred and less preferred food. Appetite 1984;5:361–71.
52. Zafra MA, Molina F, Puerto A. The neural/cephalic phase reflexes in the
physiology of nutrition. Neurosci Biobehav Rev 2006;30:1032–44.
53. Sakata T. A very-low-calorie conventional Japanese diet: its implications
for prevention of obesity. Obes Res 1995;3(suppl 2):S233–9.
54. Julis RA, Mattes RD. Influence of sweetened chewing gum on appetite,
meal patterning and energy intake. Appetite 2007;48:167–75.
55. Dulawa SC, Hen R. Recent advances in animal models of chronic an-
tidepressant effects: the novelty-induced hypophagia test. Neurosci Bi-
obehav Rev 2005;29:771–83.
800 CASSADY ET AL
Downloaded from https://academic.oup.com/ajcn/article/89/3/794/4596705 by guest on 13 May 2021