Journal of Cancer Therapy, 2012, 3, 718-730
http://dx.doi.org/10.4236/jct.2012.325091 Published Online October 2012 (http://www.SciRP.org/journal/jct)
The Anti-Tumor Effect of a High Caloric Diet in the PyMT
Mouse Breast Cancer Model Is Initiated by an Increase in
Metabolic Rate
Linda C. Enns, Warren C. Ladiges
Department of Comparative Medicine, University of Washington, Seattle, USA.
Received August 30th, 2012; revised September 29th, 2012; accepted October 13th, 2012
ABSTRACT
Obesity is associated with an increased risk of mortality from certain types of cancer, including cancer of the breast.
Because obesity is associated with multiple risk factors, however, the exact reasons remain unclear. The objective of
this study was to determine which of the risk factors associated with obesity are related to enhanced tumor development.
The MMTV-PyMT mouse model develops mammary tumors which share numerous characteristics with those of hu-
mans. We challenged these mice with a high fat/high carbohydrate, high caloric (HC) diet, and looked for relationships
between enhanced primary tumor development and adiposity, various aspects of glucose homeostasis, and metabolic
factors. The HC diet enhanced tumor progression in PyMT mice. While mice on the HC diet also developed increased
adiposity, hyperglycemia and hyperinsulinemia, none of these risk factors was found to be associated with the observed
increases in tumor growth. Rather, we found that while overall, tumor growth was enhanced in HC diet-fed mice com-
pared to those maintained on a regular diet, it was attenuated in individuals by an HC diet-induced increase in metabolic
rate and decrease in respiratory exchange ratio. Tumor size in HC diet-fed mice was directly related to p38 phosphory-
lation and Bcl-2 inhibition, and the degree of vascularization of these tumors was closely and indirectly related to the
rate of mouse oxygen consumption. The data suggest that an increase in metabolic rate and oxygen consumption, in-
duced by the introduction of a high caloric diet, has a protective effect against tumor growth by increasing the activity
levels of the tumor suppressor p38 and decreasing the activity of the antiapoptotic protein Bcl-2, as well as by reducing
hypoxia-induced tumor vascularization.
Keywords: Obesity; Breast Cancer; Metabolism; Oxygen; ROS; p38; Hypoxia
1. Introduction
With numerous studies documenting an increase in
obesity worldwide [1], obesity and its associated diseases
are rapidly becoming a central health problem of this
century. Abdominal obesity has been found to be
associated with an enhanced risk of cancer mortality in
humans [2,3]; in fact, a 2007 report by the World Cancer
Research Fund and the American Institute of Cancer
Research has suggested that the maintenance of a healthy
body weight throughout life is one of the most important
ways to protect against cancer [4]. In addition to being at
increased risk of developing certain cancers compared to
people who are lean, obese individuals have a higher
likelihood of dying from those cancers once they occur
[5]. One cancer in particular that shows a high cor-
relation between obesity and mortality is cancer of the
breast, with an estimated 30% to 50% of deaths caused by
breast cancer thought to be associated with obesity [6].
While the epidemiology supporting the link between
obesity and cancer-caused mortality is compelling, the
mechanisms behind this association are not clear, in
part because obesity is associated with multiple physio-
logical risk factors. This study was designed to de-
termine which of these different risk factors could be
associated with tumor growth in a high fat/high carbo-
hydrate, high caloric (HC) diet-challenged mammary
cancer mouse model, in an effort to assess the im-
portance of these individual factors in their contribution
to enhanced cancer progression. Transgenic mice that
carry the middle T oncogene under the transcriptional
control of the mouse mammary tumor virus promoter/
enhancer (MMTV-PyMT) develop widespread transfor-
mation of the mammary epithelium and the production of
multi- focal mammary adenocarcinomas, with secondary
metastasis to the lungs [7]. We developed a population of
transgenic (MMTV-PyMT)634Mul mice on a C57BL6/J
background, a strain known to be obesity and diabetes

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
719
sensitive when challenged with a high caloric diet [8].
Because we were interested in the role that obesity plays
in tumor progression, rather than its contribution to in-
creased cancer risk, we waited until tumors were pal-
pable before switching mice to an HC diet. In addition to
tumor burden, various physiological parameters of these
mice were monitored over the course of their tumor
progression, including adiposity, blood pressure, blood
glucose and serum insulin, metabolic rate (rate of oxygen
consumption) and respiratory exchange ratio. While the
rate of tumor growth was increased in mice challenged
with the HC diet, specific relationships between various
physiological risk factors and increased tumor growth
were not found; rather, a surprising picture emerged
showing some protective effects of the diet, related to a
nutrient overload-induced metabolic enhancement.
2. Methods
2.1. Mice
FVB/N-Tg (MMTV-PyMT)634Mul/J transgenic males on a
100% FVB background [9] were obtained from Jackson
Labs and crossed with transgenic females on a congenic
C57/BL6(J) background for 9 generations. Experimental
cohorts were then generated by crossing male PyMT
heterozygotes of these mice with congenic C57/BL6(J)
females. Experimental cohorts consisted of female WT
and littermates heterozygous for the PyMT transgene.
Mice were maintained in a 25˚C-specific pathogen-free
barrier facility with 12-hour alternating light and dark
cycles and were given free access to food and water.
All procedures used in this study were approved by
the Animal Care and Use Committee of the University of
Washington.
2.2. Specialized Diet Cohorts
The two diets used in our studies were standard rodent
chow (5053; Picolab, Richmond, IN) containing 20%
(wt/wt) protein, 4.5% fat (ether extract), and 55% carbo-
hydrate (primarily starch), and a high-fat, high-sucrose,
high caloric (HC) diet (S3282; Bio-Serv, Frenchtown,
NJ) containing 20% protein, 36% fat (primarily lard),
and 36% carbohydrate (primarily sucrose). Because we
wanted to look at the effects of a high calorie diet on
tumor growth, as opposed to cancer risk, we chose to
eliminate the variable of tumor incidence by waiting until
tumors were present before administering the dietary
challenge. All mice were maintained on standard rodent
chow until palpable tumors were just detectable (61 days
of age), at which time they were either switched to the
high-caloric diet or continued on the standard chow for
either 79 or 109 days (140 and 170 days of age, respec-
tively). Cohorts were designated as either “140 day” or
“170 day”. At each of these time points, mice were
euthanized and tumors were excised, measured and
weighed. For each genotype, diet and time point, there
were 15 - 20 mice per cohort, housed 5 to a cage with
genotypes randomized.
2.3. Body Composition, Blood Glucose and
Serum Insulin, and Blood Pressure
Body composition, blood glucose, serum insulin and
blood pressure are all measurable non-invasively and
both 140 day and 170 day cohorts of mice were moni-
tored for all of these parameters over the course of the
study. Body composition measurements were measured
using quantitative magnetic resonance imaging (QMR),
with an EchoMRI-100 analyzer (EchoMRI, Houston,
TX). Body composition measurements were performed
on live, unsedated mice. For blood glucose measure-
ments, food was removed from mice 6 hours before
blood was drawn by tail pricking. Analyses were per-
formed using a glucometer and Comfort Curve Test
Strips (Advantage; Accu-Chek, Roche, Basel, Switzer-
land). For serum insulin measurements, blood was col-
lected from the retro-orbital sinus into serum separator
tubes (365,956; Becton Dickinson); after separation,
plasma was either used immediately or stored at –80˚C
until analysis. Plasma insulin was measured using an
ELISA kit (EZRMI-13K; LINCO, St Charles, MO) as
per the manufacturer’s instructions. Systolic and diastolic
blood pressures were measured on unsedated, restrained
individuals using a volume pressure recording sensor and
an occlusion tail-cuff (CODA non-invasive blood pres-
sure system; Kent Scientific, Torrington, CT). Mice were
kept warmed to a temperature of 30˚C on a warming
platform (Kent Scientific), and 60 blood pressure meas-
urements were performed over 20 minutes to ensure ac-
climation, as determined by consistency of measure-
ments.
2.4. Metastasis/Pathology
At the time of necropsy, lungs were removed from the
mice and each lobe was perfused directly with 10% neu-
tral buffered formalin. Pulmonary metastatic foci and
tumor burden were quantified from paraffin-embedded,
H & E stained lung tissue sections that went through at
least 4 of the 5 lobes. Slides were scanned to virtual
digital files using Nanozoomer (Hamamatsu, Hamamatsu
City, Japan). Images were magnified to 2.5×, captured
and used for quantification. The number of metastatic
foci was counted in the entire section and standardized to
lung surface area. The surface area of all tumors ob-
served in the section was combined and divided by the
number of foci for average tumor surface area. Metastatic

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
720
incidence was defined as the percentage of mice present-
ing any pulmonary metastatic foci, determined from the
histological sections.
Histopathologic reviews were conducted of hematoxy-
lin and eosin-(H & E) stained primary tumors for a pre-
liminary examination of qualitative differences between
experimental groups [10]. H & E-stained sections of pri-
mary tumors were evaluated for mitosis, necrosis, and
inflammation at various (20 - 400×) magnifications. Nor-
mal and aberrant mitotic figures, characterized by trira-
dial or circular metaphase plates, were counted in three
random fields at 400× magnification. Necrosis was esti-
mated as a percent of necrotic area within the tumor sec-
tion, that is, areas that were hypocellular and hypo- or
hypereosinophilic with basophilic cellular and nuclear
debris. Inflammation was graded as 0 (minimal), 1 (mild),
2 (moderate), or 3 (marked) at a 40× or 100× magnifica-
tion. Vascularity was scored from 0 - 4; 0 = unremark-
able; 1 = few small vessels at periphery; 2 = clusters of
peripheral vessels; 3 = small vessels within tumor bulk; 4
= prominent (readily identifiable at 4× mag.) clusters
within the bulk of the tumor.
2.5. Metabolic Measurements
Indirect calorimetry is a non-invasive process and the
metabolism of both the 140 and 170 day cohorts of mice
was monitored over the course of the study. Metabolic
parameters were measured using an open-circuit indirect
calorimeter (Oxymax; Columbus Instruments, Columbus,
OH). Measurements were taken of the rates of both oxy-
gen consumption (VO2) and carbon dioxide production
(VCO2), and the respiratory exchange ratio (RER), de-
fined as the ratio of the amount of carbon dioxide pro-
duced to the amount of oxygen consumed. Mice were
measured individually and for a period of 48 hours (2
light and 2 dark cycles). Just prior to taking metabolic
measurements, the lean tissue mass of each mouse was
assessed using quantitative magnetic resonance imaging.
VO2 for each mouse were standardized to its lean mass.
2.6. Immunoblotting
Primary tumors were harvested from mice at 140 days of
age. Tissue was ground in liquid nitrogen followed by
homogenization in 1 mL of a cocktail of 20 mL ice-cold
RIPA, 2 protease inhibitor mini-tabs (11836153001; Roche,
Indianapolis, IN) and 400 ul each of the phosphatase
inhibitor cocktails P2850 and P5726 (Sigma-Aldrich, St
Louis, MO). Twenty micrograms of protein from the
cytosolic fraction was loaded to each lane of a NuPAGE
Novex 4% - 12% Bis-Tris Gel (NP0322BOX; Invitrogen,
Grand Island, NY). Blots were probed with primary an-
tibodies to AMPK-P (Abcam; ab72845; diluted 1/1000),
GLUT 1 (Santa Cruz Biotechnology;sc-7903; diluted
1/200), HIF1α (Novus Biologicals; NB100-134; diluted
1/1000), VEGF (Abcam; ab46154; 0.5μl/mL), P-Bcl-2
(ser87) (Assay Biotechnology; #A0460;1/500), P-p38
(Thr180/Tyr182) (Cell Signaling; cs#9211; diluted 1/
500), p38-total (Cell Signaling; cs#9212; 1/500) and β-
actin (Abcam; ab8227; 1/1000). A goat anti-rabbit, hor-
seradish peroxidase-conjugated secondary antibody (Ab-
cam; ab6721) was used at a dilution of 1/3000, and de-
tection was accomplished using ECL (95038-560; VWR
Scientific, Brisbane, CA).
2.7. Statistics
Error bars, where present, indicate ± standard deviation.
The probability of differences between means was de-
termined using the Student’s t test. Probabilities of indi-
vidual data points being different are indicated where the
P < 0.1. To determine the strength of relationships be-
tween different factors, the coefficient of determination
(R2) was calculated.
3. Results
3.1. PyMT Mice on a High Caloric Diet
Accumulate Body Fat, and Become
Hyperglycemic and Hyperinsulinemic
Mice were put on an HC diet at the time when palpable
tumors were detected (61 days). At 79 and 109 days after
the introduction of the diet (when mice were 140 and 170
days old, respectively), body fat, blood glucose, serum
insulin and systolic and diastolic blood pressure were
measured. The HC diet caused both PyMT mice and WT
controls to accumulate fat over the course of the dietary
challenge, although the rate of fat accumulation was
lower in the presence of tumors (Figure 1(a)). Before the
introduction of the HC diet, both genotypes had an aver-
age body fat weight of between 2 g and 3 g. After 79 and
109 days on the diet, WT mice showed 4 and 6-fold in-
creases in fat mass, respectively. While PyMT mice also
accumulated body fat, they lagged behind the WT mice,
showing 3 and 4-fold increases in fat mass respectively at
the two time points. Both PyMT and WT control mice
experienced similar increases of about 50% in blood
glucose by 79 days after the introduction of the HC diet
(Figure 1(b)). PyMT mice displayed a steady increase in
serum insulin over the course of the dietary challenge
(Figure 1(c)). The HC diet had no effect on blood pres-
sures (data not shown).
3.2. PyMT Mice on an HC Diet Show Enhanced
Primary Tumor Growth
PyMT mice maintained on an HC diet showed enhanced

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
721
(a)
(b)
(c)
Figure 1. A high fat/high carbohydrate (HC) diet causes an
accumulation of body fat, and elevated blood glucose and
serum insulin. (a) Adiposity of both WT and PyMT mice
increased over the course of the HC dietary challenge, al-
though the presence of the PyMT transgene decreased the
rate of fat accumulation; (b) After 79 days on the HC diet,
at 140 days of age, both WT and PyMT mice showed simi-
lar increases in blood glucose of about 60%; (c) PyMT mice
on the HC diet showed an increase in serum insulin that
approached significance after 109 days on the diet, or at 170
days of age. By the end of the challenge, serum insulin had
increased about 3-fold. NS = P > 0.1.
primary tumor growth compared to those fed regular
chow (Figures 2(a) and (b)). Mice have 10 mammary
glands, five on each side, located ventro-laterally. There
are four sites that can be distinguished: cervical (1), tho-
racic (2 and 3), abdominal (4) and inguinal (5). At 61 days
of age, PyMT mice presented palpable tumors mice at
many of these sites, so this age was selected for the be-
ginning of the HC dietary challenge. At 140 and 170
days of age, cohorts of mice were euthanized and tumors
were excised and weighed. Tumor burden is typically
standardized to mouse body weight, but this is not a good
indicator of mouse size in obese animals. The tumor
burden at each site was therefore calculated by standard-
izing tumor weight to mouse tibia length. By 140 days of
age, there was a difference between mice fed an HC diet
and those maintained on a regular diet in tumor burden at
the thoracic, abdominal and inguinal sites. Tumor bur-
dens for mice on both diets were still quite small, how-
ever, with mice on a regular diet showing burdens of less
than 1 g/mm, and those on the HC diet showing burdens
between 1 and 3 g/mm. For all mice, the majority of tu-
mor growth happened in the last month, between 140 and
170 days of age. During this time, tumor burdens in-
creased anywhere from 3 to 6-fold, depending on the site.
At three of the four sites, 170 day-old mice maintained
on the HC diet showed tumor burdens 50 to almost 100%
higher than those fed a regular diet; the only exception
was the cervical site, where tumor burden was similar for
both. At this age, the difference in tumor size at the tho-
racic, abdominal and inguinal sites was visually striking
(Figure 2(b)). The incidence of metastasis to the lungs in
mice maintained on the regular and the HC diet was
found to be 71.4 and 60.9%, respectively. No differences
in either the number of metastatic foci in the lungs or the
average size of metastatic tumors was found between
mice on an HC and regular diet (Fi gur es 2 (c) and (d))
3.3. Body Fat, but Not Blood Glucose or Serum
Insulin, Is Related to Tumor Weight in
PyMT Mice
PyMT mice maintained on an HC diet to 170 days of age
showed a large degree of variation in body weight by the
end of the challenge. A QMR analysis showed that the
lean mass of all of these mice, regardless of their body
weight was similar, falling between 20 and 25 g (Figure
3(a)). Fat mass, on the other hand, was variable, and was
responsible for the majority of the differences in body
weight between individuals. When the body fat of these
same mice at 140 days of age was compared with their
final tumor weights at 170 days an inverse relationship
was found, with the least obese mice ultimately develop-
ing the largest tumors (Figure 3(b)). While there was
variation between individuals in blood glucose and serum
insulin levels at 140 days of age, neither of these factors
was found to relate to tumor growth (data not shown).

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
722
(a)
(d)
(c)
(b)
Figure 2. An HC diet enhances pr imary tumor growth. (a, b) PyMT mice on an HC diet showed significantly larger thoracic,
abdominal and inguinal tumors than PyMT mice of the same age maintained on a regular diet. The majority of tumor growth
in all mice occurred between 140 and 170 days of age, with significantly faster rates of tumor growth in HC diet-challenged
mice compared to mice maintained on a regular diet; (c, d) No differences in either the number of metastatic foci to the lungs,
or the rate of secondary tumor growth were found between HC and regular diet-fed PyMT mice. NS = P > 0.1.
3.4. C57BL/6(J) Mice on an HC Diet Experience
an Increase in Oxygen Consumption (VO2)
as well as a Decrease in Respiratory
Exchange Ratio (RER)
and 79 days following introduction of the HC diet. Both
VO2 and RER of mice of both PyMT and WT mice were
very sensitive to a change in diet, showing a 7% increase
and an 8% decrease respectively within the first 24 hours
of the mice being introduced to the HC diet. When mice
were switched back to a regular diet, VO2 fell back to
regular levels within 24 h, although RER did not recover
to its original higher levels during this time period. The
initial levels to which the VO2 rose and the RER fell
within the first 24 hours of the HC diet switch were in-
dicative of the levels maintained over the course of the
HC dietary challenge; at 79 days after the introduction of
Short term and long term effects of the HC diet on oxy-
gen consumption (VO2) and respiratory exchange ratios
(RER) were measured using indirect calorimetry (Fig-
ures 4(a) and (b)). To measure short term effects, PyMT
and WT control mice were assessed while being
switched between the regular diet and HC diet at 24 hour
intervals. Long term measurements were performed at 35

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
723
(a)
(b)
Figure 3. Body fat correlates indirectly with primary tumor
growth. (a) Mice on the high calorie diet by 170 days of age,
had accumulated varying amounts of body fat; (b) The
amount of body fat in the 170 day cohort, measured at 140
days of age, was prognosticative of final tumor burden, with
the leanest mice showing the highest rates of primary tumor
growth.
the diet (140 days of age), mice showed the same both
elevated rates of VO2 as well as reduced respiratory ex-
change ratios that were observed within the first 24 hours.
The presence of tumors was not found to have any effect
on the ability of the HC diet to enhance metabolic rate or
reduce RER (Figures 4(c) and (d)).
3.5. High VO2 and Low RER for Mice on an HC
Diet at 140 Days of Age Relates to Low Final
Tumor Burdens
A relationship was found between the VO2 of individual
PyMT mice on an HC diet at 140 days of age and their
tumor burdens at both 140 and 170 days of age (Figures
5(a) and (c)). Mice with the highest rates of oxygen con-
sumption at 140 days of age had the lowest rates of tu-
mor growth. The RER of individuals on an HC diet at
140 days of age did not show a relationship with tumor
burden at that age, but was a strong prognosticator of
their tumor burden at 170 days of age (Figures 5(b) and
(d)). Mice with the lowest respiratory exchange rates at 140
days of age also showed the lowest rates of tumor growth.
Neither the VO2 nor RER of individuals was associated
with the amount of body fat (data not shown) and for mice
on a regular diet, there was no relationship between either
VO2 or RER and tumor burden (data not shown).
3.6. Primary Tumors of PyMT Mice on an HC
Diet Show Elevated Glycolysis Markers, and
Their Size Is Related to p38 and Bcl-2
Activity
The VO2 of 140 day-old PyMT mice on an HC diet
showed a strong inverse relationship with abdominal
tumor size (Figure 6(a)), which was not found for mice
maintained on a regular diet (Figure 6(c)). 5 abdominal
primary tumors each of differing sizes were selected at
random from 140 day-old PyMT mice on either an HC or
regular diet, and westerns were done for the glycolysis
markers GLUT1 and AMPK-P, for the hypoxia and an-
giogenesis markers HIF1α and VEGF, for total and
phosphorylated levels of the tumor suppressor p38, and
for phosphorylation levels of the antiapototic protein
Bcl-2 (Figure 6(b)). No overt differences were found for
HIF1α. Overall, primary tumors from mice on an HC diet
showed higher levels of AMPK phosphorylation, GLUT1
and VEGF than those on the regular diet, but none of
these levels were found to have a relationship with tumor
size. While all tumors showed similar levels of total p38,
the amount of phosphorylated p38 showed a strong and
inverse relationship with abdominal tumor size in PyMT
mice on an HC diet, with the largest tumors having the
lowest amounts (Figure 6(d)). No relationship was found
between the amount of phosphorylated p38 and tumor
size for mice maintained on a regular diet, however (data
not shown). For tumors taken from mice on the HC diet,
p38 activity was found to correlate inversely with the
amount of active antiapoptotic protein Bcl-2 (Figure
6(e)). Bcl-2 activity related to tumor size, with the high-
est levels corresponding with the largest tumors (Figure
6(f)). The data suggest that p38 activity is attenuating the
enhanced tumor growth of HC diet-challenged mice via
inhibition of the antiapoptotic protein Bcl-2.
3.7. The VO2 of Mice on an HC Diet Shows an
Inverse Relationship with the Amount of
Primary Tumor Vascularization
Abdominal primary tumors were randomly selected from
7 each of 140 day-old PyMT mice maintained on either
an HC or a regular diet (Figure 7(a)). A complete trans-
verse section from the center of each tumor was H & E
stained, and assessed for a number of qualitative pa-
rameters. No differences were found between experi-
mental cohorts for mitosis, necrosis or inflammation
(data not shown). However, when primary tumors were
graded 0 - 4 for amount of vascularity, with 4 being the
most vascularized (Figure 7(c)), for mice on the HC diet,

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
724
(a) (b)
(c) (d)
Figure 4. An HC diet causes an immediate and significant increase in VO2 as well as decrease in respiratory exchange ratio
(RER), neither of which is affected by the presence of the PyMT transgene. (a) When mice on a regular diet were put on an
HC diet, their rate of oxygen consumption, as measured by indirect calorimetry, increased about 7% over the course of 24
hours. When mice were switched back to a regular diet, their VO2 returned to original levels. Over the course of the dietary
challenge, mice maintained high rates of VO2 that did not differ significantly from the original and immediate increases ob-
served within the first 24 hours of consuming the diet; (b) Within 24 hours of being switched from a regular to an HC diet,
mice experienced an 8% decrease in RER. Mice over the course of the HC dietary challenge maintained lower RER’s that did
not differ significantly from the original drop that was observed following the first 24 hours of being introduced to the diet; (c,
d) The presence of tumors did not influence either the increases in VO2 or the decreases in RER observed in HC
diet-challenged mice. NS = P > 0.1.
VO2 at 140 days of age showed a strong and inverse rela-
tionship with the degree of primary tumor vascularization,
with the tumors from the mice with the highest rates of
oxygen consumption having the least degree of vascu-
larization (Figure 7(b)). There was no relationship be-
tween VO2 and tumor vascularization for mice main
tained on a regular diet (data not shown). The data sug-
gest that the enhanced rate of oxygen consumption by the
HC diet is attenuating tumor vascularization.
4. Discussion
Metabolic syndrome is characterized by abdominal obe-
sity, high blood glucose and insulin resistance, and high
blood pressure. It has been associated with increased
mortality by a number of different types of cancer, but
the relative importance of these different risk factors is
unknown, and the mechanisms by which obesity en-
hances cancer growth are poorly understood. One cancer
in particular that has been associated with obesity is can-
cer of the breast. To study the effects of obesity on
mammary cancer in mice, we used the MMTV-PyMT
mouse model. PyMT is a membrane bound polypeptide
and active analogue of a receptor that harbors docking
sites for a number of effector proteins that stimulate mi-
togenesis. The MMTV-PyMT mouse model shares nu-
merous characteristics with human breast tumors, in that
tumors develop with high penetrance and show gradual
loss of estrogen and progesterone receptors, the full mul-
tistage progression from hyperplasia to full-blown ma-
lignancy and metastasis is represented, and metastatic
potential appears to be independent of hormonal fluctua-
tions with a reproducible and measurable progression
rate [11]. We developed a MMTV-PyMT transgenic line
on a C57BL6/J background, known to be obesity sensi-
tive to a high fat/high carbohydrate, high caloric (HC) diet.
As expected, mice on this diet developed significant in-
creases in a number of factors associated with metabolic

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
725
(a)
(b)
(c)
(d)
Figure 5. Both HC diet-enhanced increases in VO2 and de-
creases in RER are prognosticative of final tumor burden in
PyMT mice. (a, c) The VO2 of the 140 day cohort, measured
at 140 days, and the VO2 of the 170 day cohort, measured at
170 days, both correlated with tumor burden; (b, d) The
RER of the 140 day cohort, measured at 140 days, did not
correlate with tumor burden, but the RER of the 170 day
cohort, measured at 140 days, was strongly prognosticative
of final tumor burden.
syndrome, including hyperglycemia, elevated serum in-
sulin, and increased adiposity (although not hyperten-
sion). It is already known that a high fat diet increases
primary tumor both incidence and burden in the PyMT
breast cancer mouse model [12]. We also found the rate
of primary tumor growth in these mice to be enhanced,
although the variable of tumor incidence was eliminated
by introducing our dietary challenge at the time where
palpable tumors were already detected. There was a large
degree of variability in the effects of the HC diet on both
different physiological parameters as well as primary
tumor growth. This allowed us to compare the relation-
ship between various metabolic responses of the diet-
challenged mice with their tumor growth, in an effort to
determine the most important contributing factors. Here,
we report those relationships, the most surprising of
which points to a protective effect of the HC diet against
tumor progression.
Over the course of the challenge, mice developed ele-
vated blood glucose and serum insulin levels. Hypergly-
cemia has been known to be associated with cancer for
over a century [13], and the correlation between diabetes
and breast cancer is also an old observation, dating back
to the 1950’s [14]. Neoplastic cells use glucose for pro-
liferation [15] and increases in blood glucose may thus in
part promote cancer growth by directly feeding tumors. It
is also thought that hyperglycemia may accelerate cancer
growth through altering growth hormones, such as IGF-I
and insulin [13]. A direct correlation between serum in-
sulin and both recurrence and death in breast cancer
cases has been reported [16]. We found that HC diet-
challenged mice experienced variable increases in serum
blood glucose and insulin; however, neither was found to
correlate with tumor burden amongst diet-challenged
individuals. In contrast to normal cells, which use oxida-
tive phosphorylation to generate ATP, cancer cells show
an altered metabolism that favors increased glycolysis
[17,18]. This metabolic switch can be observed at the
molecular level by changes in tumor cells of the expres-
sion and activity of a number of different proteins.
AMPK is a major cellular energy sensor [19] and its
phosphorylation reflects the activation of catabolic path-
ways in times of increased energy need [20]. GLUT1, a
glucose transporter, is necessary for uptake of glucose
and its expression is known to be upregulated in breast
tumors [21]. We found significantly enhanced AMPK
phosphorylation and upregulation of GLUT1 in the pri-
mary tumors of HC diet-challenged mice, but there was
little variability in the changes of the levels of activity or
expression of either of these proteins, and neither was
found to correlate with enhanced tumor growth in HC
diet-fed individuals.
Another way in which high calorie diets have been
proposed to enhance tumor development is by the accu-
mulation of adipose tissue, which is known to increase
circulating levels of insulin, IGF-1 and adipokines, as

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
726
(c) (b)
(d) (e) (f)
(
a
)
Figure 6. Primary tumor size in HC diet-fed individuals correlates with the amount of phosphorylation of tumor suppressor
p38 and the antiapoptotic protein Bcl-2. (a) The VO2 of individuals from the 140 day, hf diet-fed cohort correlated with the
weight of their abdominal tumors; mice with the highest VO2 presented with the smallest tumors. This correlation was not
found for the 140 day cohort of regular diet-fed PyMT mice (c); (b) 5 primary tumors were selected from the HC
diet-challenged mice shown in (a) (indicated in grey) and 5 from the regular diet-fed mice shown in (c); HC diet-challenged
mice showed higher levels of phosphorylated AMPK, and of GLUT1 and VEGF than regular diet-fed mice, with the excep-
tion of one tumor (tumor 6). No differences were seen between dietary cohorts for either HIF1α or p38 total. The amount of
p38 phosphorylation correlated strongly and inversely with the size of the tumors for the HC diet-fed cohort, however, with
the smallest tumors showed the highest levels of P-p38 (d); The amount of phosphorylated p38 correlated inversely with Bcl-2
phosphorylation levels (e); For mice on the HF diet, the amount of phosphorylated Bcl-2 correlated directly with tumor size,
with the highest levels found in the largest tumors (f).
well as to increase inflammatory conditions by attracting
immune cells [22]. Surprisingly, we found that in our
PyMT transgenic mouse model of breast cancer, while a
high calorie diet leads to increases in both adiposity and
tumor growth, it is the leanest of the HC diet-challenged
mice that develop the highest tumor loads. There is the
possibility that in addition to the anti-tumor effect pro-
vided by the enhanced metabolism of nutrient-over-
loaded mice, there is protection by the adipose tissue
itself. An inverse association between body mass index
(BMI) and the risk of breast cancer among premeno-
pausal women has been observed in numerous studies,
although the mechanisms behind this are not understood
[23]. It is also possible that the reduced adiposity in mice
with higher rates of tumor growth simply reflects an en-
ergy balance that favors tumor development over fat ac-
cumulation. This idea is supported by our observations
that mice with PyMT-generated tumors accumulate fat
slower than wild-type mice.
The observation that increased metabolic rate in re-
sponse to nutrient overload correlates with a reduced
cancer load is novel. In fact, it has been hypothesized
that metabolic stress is one way by which a high calorie
diet might promote tumor growth, by increasing the gen-
eration of reactive oxygen species (ROS) [22]. While this
is possible, it is also known that the tumor suppressor

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
727
(a) (b)
(c)
Figure 7. The VO2 of HC diet-fed individuals correlates inversely with the degree of vascularization of their primary tumors.
(a) Abdominal tumors were selected from 7 of the 140 day HC diet-fed cohort (indicated in green) and from 7 of the regular
diet-fed cohort (not shown); (b) The degree of primary tumor vascularization, graded from 0 - 4 with 4 being the most vascu-
larized, was found to correlate inversely with the HC diet-enhanced VO2 of the indi vid ual. This corr ela tio n was n ot found for
regular diet-fed mice (data not shown); (c) Representative images of H & E-stained sections of primary mammary neoplasias,
magnification 20×. Vascularity was scored on a scale of 0 - 4 as detailed in the methods. Vascularity scores are indicated in
the bottom right corner of panels. Blood vessels are highlighted by white arrows.
p38 is activated in response to increases in ROS [24], and
this ROS-induced p38 phosphorylation has been linked
to the down-regulation of several anti-apoptotic proteins
including members of the Bcl-2 family [25-29]. Our
findings that the level of p38 activity and Bcl-2 inhibition
are highly correlative both with each other as well as
with tumor size in HC diet-fed, but not regular diet-fed
mice, indicates that the metabolic boost experienced by
mice on the HC diet may be causing an anti-tumor effect
by decreasing the inhibition of apoptosis through down-
regulation of antiapototic proteins such as Bcl-2, via in-
creased activity of p38.
The presence of increased oxygen (O2) itself may play
a role in suppressing tumor growth. In addition to the HC
diet-enhanced rate of oxygen consumption, the degree of
the drop in respiratory exchange rate of HC diet-fed mice
was also found to be an early and excellent prognostica-
tor of tumor growth. The respiratory exchange rate, or
RER, is calculated as CO2 breathed out/O2 taken in; the
lower the RER, the more oxygen being consumed. One
potential way whereby enhanced oxygen consumption
might attenuate tumor growth is by inhibition of tumor
vascularization, or angiogenesis. We found that the rate
of oxygen consumption in HC diet-fed mice correlated

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
728
strongly and inversely with the degree to which tumors
were vascularized; mice with higher rates of oxygen
consumption bore tumors with less prominent vasculari-
zation. When tumors are microscopic, their cells are
oxygenated by simple diffusion of oxygen from the tu-
mor microenvironment. As they grow, lack of oxygen,
or hypoxia, stimulates the development of vasculature
which is in turn necessary for further tumor growth [30].
One way in which hypoxic regions in tumors can be re-
duced is by increasing blood flow rate and blood oxygen
content [31]. Higher oxygen consumption translates di-
rectly to more oxygen in the blood and an increase in the
oxygen being delivered to tumors. By reducing hypoxia,
enhanced oxygen consumption may attenuate tumor
growth by slowing the development of tumor vasculature.
It is thought that one molecular pathway by which hy-
poxia regulates angiogenesis is through increased ex-
pression of the hypoxia-inducible factor HIF1α, and its
downstream target vascular endothelial growth factor
(VEGF) [32]. We did not find strong differences in
HIF1α levels between regular and HC diet-fed mice.
VEGF levels were clearly higher in the latter, but we did
not find correlations between metabolic rate and either
HIF1α or VEGF levels. It is, however, already known
that the molecular evidence supporting the relationship
between hypoxia and angiogenesis is contentious [33]. It
is possible that there is a spatial aspect to the colocaliza-
tion of hypoxia and angiogenesis; in other words, looking
at protein extracts from whole tumors may obscure rela-
tionships that could be found in select areas of the tumor.
Overall, our study validates the ability of an HC diet to
enhance tumor growth in a mouse cancer model. But our
data clearly show HC diet-related beneficial effects that
attenuate this effect. The beneficial effects of the high
calorie diet, namely the metabolic enhancement that oc-
curs in lean individuals, could be separated out from
more harmful effects as a means of improving cancer
therapies. Improving blood oxygen supply has long been
recognized as having potential for sensitizing tumors to
radiotherapy. Strategies that have been proposed include
decreasing the oxygen affinity of hemoglobin in the
bloodstream using synthetic agents [34]. Using metabolic
enhancers such as carnitine [35] may also enhance the
effects of different cancer therapies. Intense exercise is
also known to increase VO2, and has been found to sig-
nificantly increase phosphorylation of the tumor sup-
pressor p38 [36]. The PyMT mouse model would be an
excellent model for studying the effects of physical exer-
cise on attenuating tumor growth. Also of extreme inter-
est is the potential prognosticative application of this
study. Within 24 hours of being put on the high calorie
diet, the increases in VO2 and decreases in RER of lean
mice with barely detectable tumors were highly prognos-
ticative of cancer aggressiveness and final tumor burden.
It has been found that overfeeding also causes the resting
metabolic rate of lean humans to increase rapidly [37], so
this finding may be translatable to humans.
Our study isolated the effects of obesity on tumor pro-
gression from its effects on cancer risk by introducing the
dietary challenge when primary tumors were already
palpable. The data show that there are clear anti-tumor
effects of an HC diet on tumors already present in mice.
These mice, however, were lean at the time of tumor
incidence and HC diet administration. It is unknown how
the metabolism of a long-term obese mouse or human
would respond to an HC diet, and these findings may
only be applicable to individuals who are metabolically
healthy.
This study shows that an increase in metabolic rate and
oxygen consumption, induced by the introduction of a
high caloric diet, has a protective effect against tumor
growth in the PyMT murine breast cancer model. Ele-
vated oxygen consumption was associated with an in-
crease in the activity levels of the tumor suppressor p38,
as well as a reduction in tumor vascularization. Overall,
however, the effects of a high fat/high calorie diet still
enhanced tumor progression. It is not our intention to
recommend the use of an HC diet to treat human indi-
viduals newly diagnosed with cancer. Rather, this study
implicates metabolic stimulation as a potentially useful
method for enhancing cancer therapies, and indicates that
for lean individuals, administering a high calorie diet
over the short term in combination with metabolic moni-
toring may provide a highly prognosticative tool.
5. Acknowledgements
WCL and LCE designed research; LCE conducted re-
search and analyzed data; LCE performed statistical
analysis and wrote paper; WCL and LCE had primary
responsibility for final content. Histological preparations
and staining were done by the Histology and Imaging
Core (HIC), UW, Seattle, WA. Histological analyses of
tumor sections were performed by Piper Treuting, Dept
of Comparative Medicine, UW, Seattle, WA. Funding:
R21 CA140916, NIH/NCI; P01 AG01751 NIH/NIA. All
authors read and approved the final manuscript.The au-
thors declare that they have no competing interests.
REFERENCES
[1] R. Samper-Ternent and S. Al Snih, “Obesity in Older
Adults: Epidemiology and Implications for Disability and
Disease,” Reviews in Clinical Gerontology, Vol. 22, No.
1, 2012, pp. 10-34. doi:10.1017/S0959259811000190
[2] J. R. Jaggers, X. Sui, S. P. Hooker, M. J. LaMonte, C. E.
Matthews, G. A. Hand and S. N. Blair, “Metabolic Syn-

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
729
drome and Risk of Cancer Mortality in Men,” European
Journal of Cancer, Vol. 45, No. 10, 2009, pp. 1831-1838.
doi:10.1016/j.ejca.2009.01.031
[3] P. Pothiwala, S. K. Jain and S. Yaturu, “Metabolic Syn-
drome and Cancer,” Metabolic Syndrome and Related
Disorders, Vol. 7, No. 4, 2009, pp. 279-288.
doi:10.1089/met.2008.0065
[4] World Cancer Research Fund/American Institute for
Cancer Research, “Food, Nutrition, Physical Activity and
the Prevention of Cancer: A Global Perspective,” Wash-
ington DC, 2007.
[5] G. Taubes, “Unraveling the Obesity-Cancer Connection,”
Science, Vol. 335, No. 6066, 2012, pp. 28-32.
doi:10.1126/science.335.6064.28
[6] J. M. Petrelli, E. E. Calle, C. Rodriguez and M. J. Thun,
“Body Mass Index, Height, and Postmenopausal Breast
Cancer Mortality in a Prospective Cohort of US Women,”
Cancer Causes and Control, Vol. 13, No. 4, 2002, pp. 325-
332. doi:10.1023/A:1015288615472
[7] C. T. Guy, R. D. Cardiff and W. J. Muller, “Induction of
Mammary Tumors by Expression of Polyomavirus Mid-
dle T Oncogene: A Transgenic Mouse Model for Metas-
tatic Disease,” Molecular and Cellular Biology, Vol. 12,
No. 3, 1992, pp. 965-961.
[8] L. C. Enns, J. F. Morton, R. S. Mangalindan, G. S.
McKnight, M. W. Schwartz, M. R. Kaeberlein, B. K.
Kennedy, P. S. Rabinovitch and W. C. Ladiges, “At-
tenuation of Age-Related Metabolic Dysfunction in Mice
with a Targeted Disruption of the Cbeta Subunit of Pro-
tein Kinase A,” The Journals of Gerontology Series A:
Biological Sciences and Medical Sciences, Vol. 64, No.
12, 2009, pp. 1221-1231. doi:10.1093/gerona/glp133
[9] E. Y. Lin, J. G. Jones, P. Li, L. Zhu, K. D. Whitney, W. J.
Muller and J. W. Pollard, “Progression to Malignancy in
the Polyoma Middle T Oncoprotein Mouse Breast Cancer
Model Provides a Reliable Model for Human Diseases,”
American Journal of Pathology, Vol. 163, No. 5, pp. 2113-
2126. doi:10.1016/S0002-9440(10)63568-7
[10] P. M. Treuting, L. I. Chen, B. S. Buetow, W. Zeng, T. A.
Birkebak, V. L. Seewaldt, K. M. Sommer, M. Emond, M,
L. Maggio-Price and K. Swisshelm, “Retinoic Acid Re-
ceptor Beta2 Inhibition of Metastasis in Mouse Mammary
Gland Xenografts,” Breast Cancer Research and Treat-
ment, Vol. 72, No. 1, 2002, pp. 79-88.
doi:10.1023/A:1014906529407
[11] J. Goh, L. Enns, S. Fatemie, H. Hopkins, J. Morton, C.
Pettan-Brewer and W. Ladiges, “Mitochondrial Targeted
Catalase Suppresses Invasive Breast Cancer in Mice,”
BMC Cancer, Vol. 11, 2011, p. 191.
doi:10.1186/1471-2407-11-191
[12] G. Llaverias, C. Danilo, I. Mercier, K. Daumer, F. Capo-
zza, T. M. Williams, F. Sotgia, M. P. Lisanti and P. G.
Frank, “Role of Cholesterol in the Development and Pro-
gression of Breast Cancer,” American Journal of Pathol-
ogy, Vol. 178, No. 1, 2011, pp. 402-412.
doi:10.1016/j.ajpath.2010.11.005
[13] F. Xue and K. B. Michels, “Diabetes, Metabolic Syn-
drome, and Breast Cancer: A Review of the Current Evi-
dence,” American Journal of Clinical Nutrition, Vol. 86,
No. 3, 2007, pp. s823-s835.
[14] A. S. Glicksman and R. W. Rawson, “Diabetes and Al-
tered Carbohydrate Metabolism in Patients with Cancer,”
Cancer, Vol. 9, No. 6, 1956, pp. 1127-1134.
doi:10.1002/1097-0142(195611/12)9:6<1127::AID-CNC
R2820090610>3.0.CO;2-4
[15] O. Warburg, “On the Origin of Cancer Cells,” Science,
Vol. 123, No. 3191, 1956, pp. 309-314.
doi:10.1126/science.123.3191.309
[16] P. J. Goodwin, M. Ennis, K. I. Pritchard, M. E. Trudeau, J.
Koo, Y. Madarnas, W. Hartwick, B. Hoffman and N.
Hood, “Fasting Insulin and Outcome in Early-Stage
Breast Cancer: Results of a Prospective Cohort Study,”
Journal of Clinical Oncology, Vol. 20, No. 1, 2002, pp.
42-51. doi:10.1200/JCO.20.1.42
[17] M. Ristow, “Oxidative Metabolism in Cancer Growth,”
Current Opinions in Clinical Nutrition and Metabolic
Care, Vol. 9, No. 4, 2006, pp. 339-345.
doi:10.1097/01.mco.0000232892.43921.98
[18] R. A. Gatenby and R. J. Gillies, “Why Do Cancers Have
High Aerobic Glycolysis?” Nature Reviews Cancer, Vol.
4, No. 11, 2004, pp. 891-899. doi:10.1038/nrc1478
[19] D. G. Hardie, “AMP-Activated/SNF1 Protein Kinases:
Conserved Guardians of Cellular Energy,” Nature Re-
views Molecular Cell Biology, Vol. 8, No. 10, 2007, pp.
774-785. doi:10.1038/nrm2249
[20] K. Inoki, J. Kim and K.-L. Guan, “AMPK and mTOR in
Cellular Energy Homeostasis and Drug Targets,” An-
nual Review of Pharmacology and Toxicology, Vol. 52,
2011, pp. 381-400.
[21] M. Schmidt, M. Kapp, M. Krockenberger, J. Dietl and U.
Kammerer, “Glycolytic Phenotype in Breast Cancer: Ac-
tivation of Akt, Up-Regulation of GLUT1, TKTL1 and
Down-Regulation of M2PK,” Journal of Cancer Re-
search and Clinical Oncology, Vol. 136, No. 2, 2010, pp.
219-225. doi:10.1007/s00432-009-0652-y
[22] M. K. Sung, J. Y. Yeon, S. Y. Park, J. H. Park and M. S.
Choi, “Obesity-Induced Metabolic Stresses in Breast and
Colon Cancer,” Annals of the New York Academy of Sci-
ences, Vol. 1229, 2011, pp. 61-68.
doi:10.1111/j.1749-6632.2011.06094.x
[23] K. B. Michels, K. L. Terry and W. C. Willett, “Longitu-
dinal Study on the Role of Body Size in Premenopausal
Breast Cancer,” Archives of Internal Medicine, Vol. 166,
No. 21, 2006, pp. 2395-2402.
doi:10.1001/archinte.166.21.2395
[24] B. M. Emerling, L. C. Platanias, E. Black, A. R. Nebreda,
R. J. Davis and N. S. Chandel, “Mitochondrial Reactive
Oxygen Species Activation of p38 Mitogen-Activated
Protein Kinase Is Required for Hypoxia Signaling,” Mo-
lecular and Cellular Biology, Vol. 25, No. 12, 2005, pp.
4853-4862. doi:10.1128/MCB.25.12.4853-4862.2005
[25] Y. Ramiro-Cortes, A. Guemez-Gamboa and J. M.
Andrade, “Reactive Oxygen Species Participate in the
p38-Mediated Apoptosis Induced by Potassium Depriva-
tion and Staurosporine in Cerebellar Granule Neurons,”
The International Journal of Biochemistry and Cell Biol-

The Anti-Tumor Effect of a High Caloric Diet in the PyMT Mouse Breast Cancer Model Is Initiated by an
Increase in Metabolic Rate
730
ogy, Vol. 43, No. 9, 2011, pp. 1373-1382.
doi:10.1016/j.biocel.2011.06.001
[26] J. Pan J, Q. Chang, X. Wang, Y. Son, Z. Zhang, G. Chen,
J. Luo, Y. Bi, F. Chen and X. Shi, “Reactive Oxygen Spe-
cies-Activated Akt/ASK1/p38 Signaling Pathway in
Nickel Compound-Induced Apoptosis in BEAS 2B Cells,”
Chemical Research in Toxicology, Vol. 23, No. 3, 2010,
pp. 568-577.
[27] M. Pearl-Yafe, D. Halperin, O. Scheuerman and I. Fabian,
“The p38 Pathway Partially Mediates Caspase-3 Activa-
tion Induced by Reactive Oxygen Species in Fanconi
Anemia C cells,” Biochemical Pharmacology, Vol. 67, No.
3, 2004, pp. 539-546. doi:10.1016/j.bcp.2003.09.024
[28] A. Van Laethem, K. Nys, S. Van Kelst, S. Claerhout, H.
Ichijo, J. Vandenheede, M. Garmyn and P. Agostinis,
“Apoptosis Signal Regulating Kinase-1 Connects Reac-
tive Oxygen Species to p38 MAPK-Induced Mitochon-
drial Apoptosis in UVB-Irradiated Human Keratino-
cytes,” Free Radical Biology and Medicine, Vol. 41, No.
9, 2006, pp. 1361-1371.
doi:10.1016/j.freeradbiomed.2006.07.007
[29] W.-S. Choi, D.-S. Eom, B. S. Han, W. K. Kim, B. H. Han,
E. J. Choi, T. H. Oh, G. J. Markelonis, J. W. Cho and Y. J.
Oh, “Phosphorylation of p38 MAPK Induced by Oxida-
tive Stress Is Linked to Activation of Both Caspase-8-
and -9-Mediated Apoptotic Pathways in Dopaminergic
Neurons,” Journal of Biological Chemistry, Vol. 279, No.
19, 2004, pp. 20451-20460. doi:10.1074/jbc.M311164200
[30] P. Carmeliet and R. K. Jain, “Angiogenesis in Cancer and
Other Diseases,” Nature, Vol. 407, No. 6801, 2000, pp.
249-257. doi:10.1038/35025220
[31] T. W. Secomb, R. Hsu, E. T. Ong, J. F. Gross and M. W.
Dewhirst, “Analysis of the Effects of Oxygen Supply and
Demand on Hypoxic Fraction in Tumors,” Acta Onco-
logica, Vol. 34, No. 3, 1995, pp. 313-316.
doi:10.3109/02841869509093981
[32] G. L. Semenza, “Angiogenesis Ischemic and Neoplastic
Disorders,” Annual Review of Medicine, Vol. 54, 2003,
pp. 17-28. doi:10.1146/annurev.med.54.101601.152418
[33] B. J. Moeller, Y. Cao, Z. Vujaskovic, C. Y. Li, Z. A.
Haroon and M. W. Dewhirst, “The Relationship between
Hypoxia and Angiogenesis,” Seminars in Radiation On-
cology, Vol. 14, No. 3, 2004, pp. 215-221.
doi:10.1016/j.semradonc.2004.04.005
[34] B. D. Kavanagh, T. W. Secomb, R. Hsu, P.-S. Lin, J.
Venitz and M. W. Dewhirst, “A Theoretical Model of the
Effects of Reduced Hemoglobin-Oxygen Affinity on
Tumor Oxygenation,” International Journal of Radiation
Oncology Biology Physics, Vol. 53, No. 1, 2002, pp. 172-
179. doi:10.1016/S0360-3016(02)02740-2
[35] L. L. Spriet, C. G. Perry and J. L. Talanian, “Legal
Pre-Event Nutritional Supplements to Assist Energy Me-
tabolism,” Essays in Biochemistry, Vol. 44, 2008, pp. 27-
43. doi:10.1042/BSE0440027
[36] M. Yu, N. K. Stepto, A. V. Chibalin, L. G. D. Fryer, D.
Carling, A. Krook, J. A. Hawley and J. R. Zierath,
“Metabolic and Mitogenic Signal Transduction in Human
Skeletal Muscle after Intense Cycling Exercise,” Journal
of Physiology, Vol. 546, Pt. 2, 2003, pp. 327-335.
doi:10.1113/jphysiol.2002.034223
[37] A. M. Harris, M. D. Jensen and J. A. Levine, “Weekly
Changes in Basal Metabolic Rate with Eight Weeks of
Overfeeding,” Obesity, Vol. 14, No. 4, 2006, pp. 690-695.
doi:10.1038/oby.2006.78