Download Influence of Dietary Fat, Caloric Restriction, and

(CANCER RESEARCH 48, 4276-4283, August 1, 1988]
Influence of Dietary Fat, Caloric Restriction, and Voluntary Exercise on
7V-Nitrosomethylurea-induced Mammary Tumorigenesis in Rats1
Leonard A. Cohen,2 Keewhan Choi, and Chung-Xiou Wang
Division of Nutrition and Endocrinology, Naylor Dana Institute for Disease Prevention, Valhalla, New York 10595 [L. A. C., C-X. W.], and Department of Biostatistics,
City of Hope National Medical Center, Duarte, California 91010 [K. CJ
ABSTRACT
The effect of dietary fat, energy restriction, and exercise on /V-nitrosomethylurea (NMU:CAS:684-93-5)-induced
mammary tumorigenesis in
female F344 rats was investigated. Rats were fed the NIH-07 diet until
/V-nitrosomethylurea administration on Day 50 of age, when they were
transferred to six treatment groups. Three sedentary groups were fed
either high-fat (20%, w/w), medium-fat (10%), or low-fat (5%) diets ad
libitum (HFAL, MFAL, LFAL, respectively); two sedentary groups were
fed high fat and medium fat diets restricted to 75% of the food consumed
by their ad libitum counterparts (HFR, MFR), and one group was fed a
HFAL diet but allowed free access to an activity wheel (HFEX). Tumor
yields among the three ad libitum sedentary groups were significantly
greater in the HFAL and MFAL groups when compared to the LFAL
group. Dietary restriction reduced tumor yields by more than 90% of ad
libitum controls regardless of fat intake. Voluntary exercise reduced
tumor yields and delayed time of tumor appearance in HFEX animals to
levels similar to those found in LFAL animals. Animals with voluntary
access to exercise wheels averaged between 1.03 and 2.85 miles/day,
consumed more food (+18%), and exhibited greater weight gain (+13%)
than their sedentary counterparts. Restricted animals exhibited signifi
cantly decreased body weight gains (-15%) compared to their ad libitum
counterparts, but no differences in weight gains were detected among the
HFAL, MFAL, and LFAL groups, despite widely varying amounts of fat
intake. Body composition studies indicated that body fat content was not
influenced by the quantity of fat consumed in the diet, but was signifi
cantly reduced by caloric restriction (—20to 26%) and exercise (—20%).
While the precise mechanisms underlying the tumor-promoting effects
of HFAL diets and the antipromoting effects of energy restriction and
exercise remain to be elucidated, available evidence suggests that these
effects are not due to alterations in energy homeostasis per se, but may
instead be exerted indirectly, and perhaps independently via endocrine,
paracrine, or neurohormonal mechanisms.
INTRODUCTION
Currently, there is considerable debate over the relative im
portance of dietary fat and total caloric intake as modulators
of mammary carcinogenesis (1-5). Over the past half-century,
a number of investigators have demonstrated that high-fat
intake stimulates the development of experimental mammary
cancer in rodents (6-9). These studies have shown that dietary
fat exerts its most pronounced effects on the promotion phase
and that there is a requirement for a minimum level of the
EFA,31 indicate (10). Moreover, the results of laboratory animal
studies are consistent with a substantial body of epidemiological
evidence, suggesting that risk for breast cancer is correlated
with fat intake, especially in the postmenopausal age groups
Received 11/3/87; revised 4/15/88; accepted 5/3/88.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1Supported by a grant from Best Foods, a Division of CPC International, Inc.,
Union, NJ. Presented in part at the 78th Annual Meeting of the American
Association for Cancer Research, Atlanta, GA, March 1987 (77).
1To whom requests for reprints should be addressed.
' The abbreviations used are: EFA, essential fatty acid; ANOVA, analysis of
variance; DMBA, 7,12-dimethylbenz(a)anthracene; HF, high fat; HFAL, high fat
ad libitum sedentary; LFAL, low fat ad libitum sedentary; MF, medium fat;
MFAL, medium fat ad libitum, sedentary; MFR, medium fat sedentary restricted;
NMU, A'-nitrosomethylurea; HFEX, high fat, ad libitum, active, HFR, high fat,
restricted, sedentary.
(11, 12). Another set of data with an equally long history
indicates that tumor incidence is correlated with daily caloric
intake and body weight (3, 4, 13-16). Since fat is the most
calorically dense nutrient, high calorie diets are often synonomous with high fat diets. Hence, it has been difficult to
determine whether fat and calories exert a common or inde
pendent effect on mammary tumorigenesis (5). Recently, using
the DMBA model, Kritchevsky et al. (2, 17) and Pariza (3, 18)
have reemphasized the importance of caloric intake and have
suggested that fat exerts its promoting effects by virtue of its
being more efficiently utilized as an energy source than either
protein or carbohydrate.
The resurgence of interest in the role of reduced energy intake
in cancer prevention has triggered interest in the opposite side
of the energy equation, namely the role of increased energy
expenditure in mammary tumorigenesis. The rationale behind
such an approach is that, if decreased energy intake acts to
suppress tumorigenesis, then its converse, i.e., increased energy
expenditure, should exert a similar suppressive effect. Support
ing this view are the early experimental studies of Rusch and
Kline (19) and Moore and Tittle (20) and the more recent
epidemiological studies of Frisch and coworkers (21, 22) which
suggest that breast cancer incidence is lower in women athletes
than nonathletes. Few laboratory animal experiments (19, 20,
23-25) have investigated the effects of exercise on induced
tumor development; in each of these, forced exercise was used,
thus introducing an intervening stress variable (26, 27).
There is evidence to suggest that tumorigenesis may be af
fected differently by nutritional factors, depending upon the
nature of the initiating agent. Pollard and Luckert (28), for
example, reported that caloric restriction exerted an inhibiting
effect on colon tumorigenesis when the initiating agent was an
indirect-acting carcinogen (requiring host activation) but not
when it was a direct-acting carcinogen (NMU). Since most
previous studies (2, 3, 17, 18) on caloric restriction and exper
imental mammary cancer have used DMBA, which requires
host activation, as the initiating agent it is of importance to
determine the effects of energy restriction (and exercise) using
a direct-acting carcinogen such as NMU (20). Further recom
mending the NMU model is the fact that it more closely mimics
human breast cancer both histologically (30) and in endocrine
responsiveness (31, 32) than the DMBA model.
The overall purpose of this study, therefore, was to analyze
the influence of dietary fat level, energy restriction, and in
creased energy expenditure on the promotion phase of NMUinduced mammary tumorigenesis.
MATERIALS AND METHODS
Experimental Procedures. Female inbred F344 rats were purchased
from Charles River Breeding Laboratories (Kingston, NY) at 28 days
of age, fed the NIH-07 diet, and housed 3 to a cage in polyethylene
cages (19 in x IO': in x 8 in) containing hardwood shavings and covered
with filter tops.4 The animal room was controlled for temperature (24
4 Animals were maintained according to the revised "Guide for the Care and
Use of Laboratory Animals, Department of Health, Education, and Welfare
Publication NIH 85-23, revised 1985.
4276
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
FAT, CALORIE RESTRICTION,
EXERCISE, AND MAMMARY CANCER
±2°C,SD), light (12-h cycle), and humidity (50%). On Day 50 of age,
all animals received a single dose of NMU (37.5 mg/kg of body weight)
by tail injection. The NMU (Ash Stevens, Inc., Detroit, MI) was
dissolved in 5 to 10 drops of 3% acetic acid and brought up to volume
in distilled I LO yielding a stock solution of 10 mg/ml administered
within 3 h of formulation.
Two days after NMU administration, rats were allocated randomly,
by weight, to six experimental groups. Three sedentary groups (LFAL,
Group 1; MFAL, Group 2; HFAL, Group 3) of 36 animals each were
fed 20%, 10%, and 5% fat diets ad libitum (Table 1). Two other
sedentary groups (MFR, Group 4; HFR, Group 5) of 36 animals each
were fed 26.6 and 13.3% fat diets restricted to 75% of their respective
ad libitum groups, HFAL and MFAL. One active group (HFEX, Group
6) of 30 rats was housed in wheel-cage units and fed the HF diet ad
libitum.
Animals in the HFAL, MFAL, and LFAL groups were housed 3 to
a cage with the exception that 6 animals each from the HFAL and
MFAL were housed singly to assess average daily food consumption.
The 2 restricted groups were fed 75% by weight of the mean daily
dietary intake of their ad libitum counterparts (calculated over a 3-day
period). The diets fed the 2 restricted groups were adjusted to ensure
that they consumed an amount of fat equal to their ad libitum counter
parts. Animals in the HFR and MFR groups were housed one to a
cage. Rats in the active group were housed in individual wheel-cage
units, type H8002 (Lab Products, Inc., Maywood, NJ), consisting of a
wire-bottomed stainless steel housing cage (10 in x 6 in x 5 in) equipped
with a running wheel (13.5-in diameter and 4.5-in wide) to which
animals had free access. Attached to the running wheel was a mechan
ically coupled odometer which counted each turn regardless of direc
tion. One complete revolution of the wheel is equivalent to a distance
of 3.53 ft; 1 mile equals 1496 revolutions. Each wheel cage unit had a
food tray and water bottle attached.
The High, Medium, and Low Fat Diets. The diets in these experiments
were based on the recommendation of the Committee on Laboratory
Animal Diets of the National Academy of Sciences (33, 34) with minor
modifications (Table 1). It has been found in our laboratory (35) and
those of others (36-38) that, under ad libitum feeding conditions, rats
adjust their food intake so that similar energy intake is maintained
despite the fact that diets may differ substantially in energy density.
Hence, animals eat less of the energy-dense, HFAL diet than the lessenergy-dense MFAL or LFAL diets. Consequently, unless the propor
tions of the other dietary components are adjusted, animals consuming
a HFAL diet will obtain less protein fiber, vitamins, and trace elements
than animals fed MFAL diets. The adjustments recommended by the
Committee on Laboratory Animal Diets (39) were, therefore, incorpo
rated into our diet formulations.
As seen in Table 1, the increase in fat was compensated for by a
decrease in the amount of carbohydrate added to the diet. The HFAL,
MFAL, and LFAL diets contained 20%, 10%, and 5% (w/w) corn oil.
Adjustments were also made for the 2 restricted groups to ensure that,
despite lower overall food intake, the absolute intake of fat, vitamins,
Table 1 Composition of experimental diets (g/100 g)
Diet
trace elements, and fiber per rat was identical in both ad libitum and
calorie-restricted groups.
The HFAL, MFAL, and LFAL diets were designed to mimic (a) the
typical western high-fat diet (38% cal as fat), (b) a reduced-fat diet (20%
cal as fat), and (c) the Japanese diet during the yr 1957-1959, (10 to
13% kcal as fat) (40). Based on an estimated average daily food
consumption of 45 cal, the LFAL diet provided approximately 5 to 6
cal of fat/day, or 12% of total calories; the MFAL diet provided
approximately 11 kcal of fat/day or 22% of total calories; and the
HFAL diet provided approximately 21 cal of fat/day or 40% of total
calories as fat. Diets, prepared by BioServ, Inc. (Frenchtown, NJ), were
packed in 10-kg lots in plastic bags and stored at 4 ( in the dark until
used. Diets were administered in powdered form, and tap water was
available ad libitum. Stainless steel J-type powder feeders (Lab Prod
ucts, Inc., Maywood, NJ) were used to prevent scattering of powdered
food.
Food Consumption Measurements. Food consumption was measured
by calculating the difference between the weight of food added to food
cups and that remaining after a 24-h feeding period. This procedure
was repeated for 3 successive days (6 animals/group). Actual average
daily food consumption was recalculated based on an average measured
15% loss of food due to spillage.
Observation Schedule. At weekly intervals, beginning 4 wk after
NMU injection, each rat was weighed, and the position and date of
appearance of palpable tumor were recorded.
Necropsy and Histopathology. Approximately 20 wk after NMU
administration, all rats were sacrificed by < <> euthanasia. Palpable and
nonpalpable (small lesions seen only at necropsy) tumors were excised,
fixed in buffered formalin, blocked in paraffin, sectioned, and stained
with hematoxylin and eosin. Histológica! diagnosis of mammary tu
mors (by C. X. W.) was based on criteria outlined by Young and
Hallowes(41).
Body Composition. Five to eight animals from each group were
randomly selected for body composition determination. At necropsy,
gut contents were removed, and the entire carcasses were then shipped
in dry ice l j Nutrition International, Inc. (New Brunswick, NJ), where
body composition analyses were performed. All animals were labeled
by code to assure that the analyses were unbiased. Proximate analysis
was conducted by standard methods of the Association of Official
Analytical Chemists. Briefly, the method involved determination of
water content by lyophilization followed by chemical assay of total fat
and protein. Total ash was calculated gravimetrically following thermal
decomposition at >500"C.
Statistical Evaluation of Data. Using a Fortran program provided by
Dr. J. J. Gart (National Cancer Institute) (42), a tumor-free survival
function was estimated separately for each treatment group by the
Kaplan-Meier product limit method (43). The survival functions for
the different treatment groups were then compared by the Cox and the
generalized Kruskal-Wallis tests. The purpose of both methods was to
test the null hypothesis that the tumor-free survival functions of all 6
treatment groups were identical. If either test indicated significant
differences among the 6 treatment groups, pairwise comparisons be
tween specific groups were then conducted.
To determine whether tumor incidences varied as a direct function
of fat intake, differences in tumor incidences among the HFAL, MFAL,
and LFAL groups were compared by Bartholomew's test for ordered
IngredientSucroseCorn
alternatives (44). When the null hypothesis of no difference in tumor
incidence among all 6 treatment groups was rejected, pairwise compar
starchCaseinMethionineCorn
isons were conducted using the Fisher exact test with Bonferroni's
adjustment to the a-value for multiple comparisons.
oilCelluloseMinerals"Vitamins"Choline
The significance of differences in tumor multiplicity among the 6
treatment groups was assessed by the following 4 methods. Differences
in total tumor burden/group were assessed for significance by the
pairwise x2 test. Tests for trends were conducted by Bartholomew's test
bitartrate1SOIS200.35.05.03.51.00.2244.8313.4521.140.3210.05.293.71.060.21334.4610.3423.480.35205.874.11.170.23434.2310.2728.120.4313.37.044.921.410.28520.446.1431.230.4
(44). Multiplicity defined as the number of tumors/tumor-bearing ani
4.10
4.71
4.53
4.08
4.53
kcal/g
3.85
mal or the number of tumors/total number of animals at risk, was
assessed by Student's t test; multiplicity, assessed in terms of the
% of kcal CHO
% of kcal protein
% of kcal fat
•
AIN 76A mixture.
67.5
20.8
11.7
57.3
20.7
22.0
39.6
20.7
39.7
43.4
27.4
29.2
22.7
26.5
50.8
39.6
20.7
39.7
frequency distribution of tumors per animal, was tested for significance
by the x2 test for homogeneity.
Total tumor volume was determined by measuring each tumor in 3
4277
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
FAT, CALORIE RESTRICTION,
EXERCISE, AND MAMMARY CANCER
dimensions and using the formula for an ellipse (4/3irRt2,Rï)to calculate
tumor volumes. Since tumor volume data were not normally distributed,
statistical comparisons were conducted by the nonparametric MannWhitney U test.
Animal weight gains were analyzed using a program for ANOVA
with repeated measures (45). This program tests the main effects of
time, diet, and activity status and their joint effects (interactions) on
weight gain.
RESULTS
Food Consumption. On average, the LFAL group consumed
12.4, the MFAL 11.7, and the HFAL group 9 g of food/day
under ad libitum sedentary conditions. The restricted groups
consumed 8.8 (MFR) and 6.75 (HFR) g/day, and the HFEX
group consumed 11.7 g/day. Based on the calculated energy
densities of the different diets (Table 1), the HFR and MFR
groups consumed 32 and 36 kcal/day compared to 41 and 48
kcal/day for their respective ad libitum counterparts (HFAL,
MFAL). HFEX animals consumed considerably more calories
than their sedentary HFAL counterparts (53 versus 41 kcal/
day).
Tumor Histopathology. With the exception of mammary tu
mors, no gross changes in major organs were seen. Tumors
were either adenocarcinomas, adenomas, or fibroadenomas, the
latter two classes being very rare [3 and 4, respectively, of a
total of 192 mammary tumors (Table 2)]. There were 4,1, 1, 2,
0, and 1 unscheduled terminations in Groups 1 to 6, respec
tively, due to necrotizing tumors. No animals died of extraneous
causes.
Survival Analysis. Analysis of survival curves (Fig. 1) indi
cates that mammary tumors appeared in the order HFAL >
MFAL > HFEX > LFAL > MFR > HFR. Tests for overall
trend were highly significant (P < 0.01) by Cox's analysis for
adjusted trends and the generalized Kruskal-Wallis test (P <
0.01). Pairwise comparisons revealed that tumors appeared
more rapidly in the HFAL than the LFAL groups (P < 0.01).
No difference was observed when the HFAL group was com
pared with the MFAL group. Tumor appearance was dramati
cally delayed in the two restricted groups compared to their ad
libitum counterparts (P < 0.01). Tumor appearance was also
significantly delayed in the active group versus its sedentary
control (P < 0.05). When viewed over time (Fig. 1), it can be
seen that all ad libitum groups, including the active group,
exhibited a similar time to initial tumor (63 to 70 days).
However, the subsequent rate of tumor appearance was in
creased in proportion to the amount of fat in the diet of
sedentary rats; voluntary exercise, in contrast, reduced the
subsequent rate to initial tumor appearance, and the subsequent
rate of tumor appearance in restricted animals was drastically
delayed with, in most rats, a complete absence of tumors at
termination of the experiment (85 days, MFR; 150 days, HFR).
Tumor latency, when assessed in terms of mean (or median)
days until first tumor, also indicated that tumor appearance
was delayed by both energy restriction and increased energy
expenditure. Mean (median) days to first tumor, in order of
earliest appearance, were: MFAL, 102 (106); HFAL, 109 (113);
HFEX, 125 (146); LFAL, 128 (150); MFR, 147 (150); and
HFR, 150(150).
Tumor Incidence. Tumor incidence data were analyzed in
terms of total adenocarcinomas, no npal pable adenocarcinomas,
and total palpable adenocarcinomas (Table 3). Tests for overall
negative trend were significant for the HFAL, MFAL, and
LFAL groups when assessed in terms of either total adenocar
cinomas (P < 0.05) or total palpable adenocarcinomas (P <
0.05) but not nonpalpable adenocarcinomas. Pairwise compar
isons indicated a significant difference between the HFAL and
LFAL groups (P < 0.05) and the MFAL and LFAL groups (P
< 0.05) but not between the HFAL and MFAL groups. Pairwise
comparisons between HFAL and MFAL group and their calo
rie-restricted counterparts (HFR and MFR) revealed a striking
inhibition in tumor incidence in the energy-restricted groups
(>90%). The difference in incidence of palpable adenocarcinoma between the active (HFEX) and sedentary (HFAL) groups
was statistically significant (P< 0.05) (Table 3). It is noteworthy
that the frequency of nonpalpable (microscopic) adenocarcinoma was reduced significantly in the two restricted groups but
not in the LFAL or exercised ad libitum groups when compared
to their respective controls (Table 3).
Tumor Multiplicity. When assessed in terms of total tumor
burden/group (Table 2), an overall negative trend (P = <0.05)
was found among the three ad libitum sedentary groups (HFAL
> MFAL > LFAL). Pairwise comparisons revealed that the
LFAL group exhibited a significantly reduced tumor yield in
comparison to its MFAL and HFAL counterparts (P = <0.05).
HFEX animals exhibited a reduced total tumor burden com
pared to their sedentary controls, but the difference did not
achieve statistical significance. Both calorie-restricted groups
exhibited dramatically reduced tumor yields in comparison to
their ad libitum-fed counterparts.
Examination of the patterns of the frequency distribution of
tumors/animal in the various groups revealed that the high
incidence group HFAL exhibited a pattern characterized by a
higher frequency of animals bearing one or more tumors,
whereas the low incidence groups (LFAL, MFR, HFR, HFEX)
exhibited a pattern in which the distribution as a whole was
shifted to the left (Table 4). The MFAL group exhibited an
intermediate pattern. Tests for trend indicated a significant
negative trend for the 3 ad libitum sedentary groups [Group
HFAL > MFAL > LFAL (P < 0.05)]. Pairwise comparisons
of tumor frequency data (including animals with 0 tumors), by
the x2 test for homogeneity, between the HFAL and HFEX
groups indicated a significant decrease in tumor multiplicity in
Table 2 Total mammary tumors by histopathological classification
nonpalpable)Group1
Tumors (palpable and
fat5
of
effective
rats36
enoma1
cinoma2
carcinoma1
adenocarcinoma25
adenocarcinoma9
mammary
tumors38"
libitum (sedentary)
2
Ad libitum (sedentary)
10
36
0
1
1
43
0
510
Ad libitum (sedentary)
3
20
36
2
2
4
0
414
4
Restrict (sedentary)
13
36
0
0
0
0
1
Restrict (sedentary)
5
26
36
0
0
0
0
2
010Total
Ad libitum (active)%of
6TreatmentAd
20No.
30Adenoma0 1Fibroad
1Adenocar- 2Comedo
0Papillary
24Cribiform
°Test for total overall negative trend: Group 3 > Group 2 > Group 1 (P< 0.05) [Bartholomew's test (44)). Pairwise tests: Group 4 or 5 versus Groups
6 (P < 0.0001) and Group 3 versus Group 1 (/•
< 0.05); Group 2 versus Group 1 (not significant), Group 3 versus Group 6 (not significant) by x2 test.
50
59
5238
1, 2, 3, and
4278
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
FAT, CALORIE RESTRICTION, EXERCISE, AND MAMMARY CANCER
i.u-0.9k.o
waned over time, reaching a plateau around 3 mo. The number
of revolutions/day ranged from a high of 25,000 to a low of
400
with the average number of revolutions/day ranging from
333-666641
2000 to 4000 over the duration of the experiment. On average,
E 0.8•
the animals ran 1.8 mile/day. The maximum number of miles
,
>
6666
3
2
6
III!
run per day by an individual rat was 14, and the minimum was
326
I
!
2222
66666666
lull
| 0.7
0 miles. In order to determine whether an association existed
66666666
ÃŽ3SSJS3'
66666666 III
between the degree of activity and tumor incidence, animals in
i 0.6
the active group were segregated into tenues based in their
•2222222
O)
C
33333
22222
fit
relative
degree of activity (Fig. 3). To our surprise, it was found
:> 0.5
3
2
that the lowest tertile (least active rats) exhibited the greatest
3
22222222
22222
3333
reduction in tumor incidence. This relationship held true
<£0.4-1
whether total activity, the early (most active phase), or the later
o
(least active) phase was used as the basis for tertile calculations.
1 0.3
o.
Comparison of mean animal weights by activity tertile revealed
o
no differences in weight at termination between groups despite
a. 0.2 i
considerable differences in activity. This lack of effect on animal
0.1
weight gain persisted regardless of how exercise tertiles were
calculated (data not shown).
0.0
Weight Gain Data. Animal weight gain curves for Groups 1,
0
20
40
60
80
100 120 140 160 2, and 3 were statistically indistinguishable, indicating that,
despite wide differences in the caloric density of their respective
Time (Days)
Fig. 1. Kaplan-Meier life table curves for cumulative mammary tumor inci
diets, the animals in the HFAL, MFAL, and LFAL groups
dence for Groups 1 to 6. Life table data include all palpable tumors (adenocarciconsumed food approximately in an isocaloric manner (Fig. 4).
noma and fibroadenoma). Data points represent total palpable tumors including
Profile analysis based on body weight measures revealed, as
those that repressed during the experiment. Ordinate, proportion of animals
expected, that all groups gained weight as a function of time (/'
surviving per unit of time without a tumor (1.0 represents 100% tumor-free
animals). Abscissa, days past NMU treatment. Test for overall trend: Cox's test
<
0.05), and that there was a significant interaction between
for adjusted trends (P< 0.01) and generalized Kruskal-Wallis analysis (/>< 0.01).
Pairwise comparisons: Cox's test: HFAL versus LFAL (/>< 0.001); MFAL versus
time and calorie restriction (i.e., a depression in the weight gain
LFAL (P < 0.05); HFAL versus MFAL (not significant); HFAL versus HFR (P
profile) and time and exercise (i.e., an elevation in the weight
< 0.001); MFAL versus MFR (P< 0.001); HFAL versus HFEX (P < 0.01).
gain profile) (P < 0.0001 in both cases). At termination, the
mean (±SD)body weights for the LFAL, MFAL, and HFAL
the active versus the sedentary group (P < 0.01). When multi
groups were 191 ±8.5, 193 ±10, 197 ±11 g, respectively.
plicity was expressed in terms of mean number of tumors/
Mean body weights for the MFR and HFR groups were 164 ±
tumor-bearing animal, no significant differences were found
6.5
and 165 ±5.4, respectively. Active animals (HFEX) exhib
between any of the groups. When expressed in terms of mean
ited
the highest mean body weight (217 ±17 g) with a greater
number of tumors/total animals at risk, the MFR and HFR
degree
of individual variation than the sedentary groups. At
exhibited significantly decreased multiplicity; all other compar
termination, energy-restricted groups exhibited a 15% decrease,
isons were nonsignificant due to the highly variable and skewed
and active animals, a 13% increase in body weight when com
nature of the tumor multiplicity data.
pared to their respective ad libitum sedentary controls.
Analysis of tumor volume data revealed a positive linear
Body Composition. Deposition of fat in adipose tissue, ex
relationship between increased tumor volume and increased
pressed as a percentage of body fat, was similar in all three
dietary fat intake among the 3 sedentary ad libitum groups
sedentary, ad libitum-fed groups, indicating that the percentage
(HFAL > MFAL > LFAL) (Table 5). However, no differences
in mean tumor volumes could be found between the HFAL and of body fat was not dependent on fat intake (Table 6). Animals
on energy-restricted diets and active animals exhibited signifi
HFEX groups.
Activity Profiles. The activity profiles are compared for the cantly lower percentages of body fat compared to their ad
most and least active of 30 animals in the activity group; all libitum sedentary counterparts. In general, water content varied
inversely with fat content; protein and ash were similar in all
others fell between these two extremes (Fig. 2). Note that
interest in the running wheel reached a peak at 1 mo and slowly six groups.
IMI•22'
III 1 1 1
666666*
M 11M 1M 1
fat5
Group1
libitum (sedentary)
Ad libitum (sedentary)
2
Ad libitum (sedentary)
34
Restrict (sedentary)
Restrict (sedentary)
5
Ad libitum (active)%of
6TreatmentIli
" Number
* Number
' Number
" Test for
versus Group
of animals with
of animals with
of animals with
overall negative
1 (P < 0.05);
adenocarcinoma)
one or
one or
one or
trend:
Group
10
20
13
26
20No.
Table 3 Mammary tumor incidence
adenocarcinomaNo.13
of
rats36
(%)36
18
36
36
16
36
22
36
30Nonpalpable*
12Incidence
SO
45
6
6
40Palpable*
adenocarci-
adenocarcinomaNo.21
nuil'
nomaNo.12"
(%)33"
21
25
3
1
14Incidence
58
70
8
3
471
(%)58
28
28
5
3
19Incidence
78
78
14
8
63
more nonpalpable adenocarcinomas.
more palpable adenocarcinomas.
more palpable and/or nonpalpable tumors.
Group 3 > Group 2 > Group 1 (P < 0.05) (Bartholomew's
test). Pairwise tests: Group 3 versus Group 1 (P < 0.05); Group 2
3 versus Group 6 (P < 0.05) by Fisher's exact test (palpable adenocarcinoma).
Other comparisons
(nonpalpable
and total
not significant.
4279
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
FAT, CALORIE RESTRICTION, EXERCISE, AND MAMMARY CANCER
Final Tumor Incidence by Level of Exercise
Table 4 Tumor frequency distribution
No. of palpable adenocarcinomas/animal
Group"1
15
2
11
17
3
4
33
3
1
5
35
6015
1617
924
°n = 36 (Groups I to 5), n = 30 (Group 6).
* No. of animals.
c Overall test for significance (comparing all 6
5
50
0531
0
1
0
0
0
1
2
0
0
040
g
100
g
80
l
o
60
f
40
g
10°
8
80
P
60
!
o
|
40
c
Table 5 Tumor volume (mm3)
fat5
T, =1.03 -1.69
T, = 1.70 -2.00
T3 = 2.05 -2.87
groups) by x2 test for homoge
neity (P < 0.05). Pairwise comparisons: Group 1 versus Group 3 (P < 0.05);
Group I versus Group 2 (P < 0.05); Group 3 versus Group 6 (P < 0.01); all group
versus Group 5 (P < 0.001); Group 2 versai Group 3 and Group 4 versus Group
5, not significant, by \2 test.
Group12
Total Exercise (142 Days)
of
rats36
libitum (sedentary')
±
Ad libitum (sedentary) 10
682 ±316 (2-> 14,236) 101177
36
1,212 ±333 (4-> 12,740)
3 Ad libitum (sedentary) 20
36
4 Restrict
354 ±148(99->611)
13
36
351
26
63 ±36 (26->99)
5 Restrict
36
63349
535(2->12,130)Median51
Ad libitum (exercise)%of 20No. 30504 1,554 ±Mean301°(2->8,424)*
6TreatmentAd
" Mean ±SEM.
* Range.
c Test for statistical significance by the Mann-Whitney U test: Group 1 versus
Group 2 (P < 0.0258); Group 2 versus Group 6 (P < 0.0167); Group 1 versus
Group 3 (P < 0.0006); Group 3 versus Group 6 (P < 0.2891); Group 2 versai
Group 3 (P < 0.0323); Group 1 versus Group 6 (P < 0.0006). Statistics for
Groups 4 and 5 not done due to insufficient sample size (Group 4, 3 tumors;
Group 5, 2 tumors).
Exercise During Last 79 Days
T, = 0.10 -0.82
T2 = 0.83 -1.14
T3=1.15 -1.88
20
0
? 10°
g" 80
c
2
Exercise During First 63 Days
60
1o «
h
Exercise Tenues (x miles/day)
Activity Profiles of Most and Least
Active Animal in Exercise Group
Fig. 3. Tumor incidence plotted as a function of degree of activity. The ranges
of the exercise tertiles for total exercise (top), early exercise (middle), and later
exercise (bottom) are denoted above. Each exercise fertile consisted of 10 animals
(total n = 30).
Most Active
Least Active
220
210
200
190
„ 180
July
August
September
Time
November
December
.!? 170
5 160
Fig. 2. Plot of activity versus time depicting the time-dependent pattern of
activity for the most and least active animals in wheel cage units. Note that early
interest in wheel-running subsided over time. All animals exhibited a similar
chronological pattern of voluntary activity.
150
140
130
DISCUSSION
120
As expected, animals fed LFAL diets exhibited significantly
lower tumor yields than animals fed HFAL diets (Tables 2 to
5). However, the results of feeding the MF diet were less clear.
For example, when assessed in terms of either tumor incidence
or multiplicity, the MFAL group exhibited a significantly
higher tumor yield than the LF group (Tables 3 and 4), sug
gesting that HFAL and MFAL diets exert similar promoting
effects. When assessed in terms of total mammary tumors/
group, however, the difference between MFAL and LFAL
groups did not achieve statistical significance (Table 2); tests
for trend indicated an ordered sequence (HFAL > MFAL >
LFAL) in all but one case [total adenocarcinoma (Table 3)],
suggesting that the MFAL group occupied an intermediate
110
10
Time
15
20
Fig. 4. Animal weights (g) plotted as a function of time (wk from NMU) for
Groups 1 to 6. A, Group 1; A, Group 2; •,Group 3; •,Group 4; D, Group 5; O,
Group 6. By ANO VA for repeated measures, Groups 1, 2, and 3, not significant.
Curves for the two restricted groups (Nos. 4 and 5) were significantly depressed,
and that for the active group (No. 6) was elevated, as a function of time (P <
0.0001), compared to Groups 1 to 3. Points, mean; bars, SD.
position
previous
of NMU
exhibited
between the HFAL and LFAL groups. In contrast,
studies conducted in our laboratory using a lower dose
(25 mg/kg) indicated than animals fed a MFAL diet
tumor yields significantly lower than those fed a
4280
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
FAT, CALORIE RESTRICTION, EXERCISE, AND MAMMARY CANCER
Table 6 Body composition as a percentage of the total
Statistical analysis was by the Student t test (two-tailed comparison). The percentage of fat: Group 2 versus Group 4, P < 0.002; Group 3 versus Group 5, P <
0.006; Group 3 versus Group 6, P < 0.06 (P < 0.03, one-tailed test). The percentage of moisture: Group 3 versus Groups 4 and 5, P < 0.01; Group 3 versus Group 6,
P < 0.07; (P < 0.035, one-tailed). All other pairwise comparisons were not significant.
Group1
of
animals6
(g)192.8 wt
±3.6°
(21)'
2
34
5
16.5(209)'6No.
Mean ±SD.
* Median.
6
57
7
8Fat19.4
19.0
20.7
14.3
16.5
16.0
±2.0 (17)
18.3
17.8
±(20)
±2.3 (13)
18.7
±2.4 (15)
18.3
±4.7 (14)Protein17.7 18.6
±1.5(17)
±0.8 (18)
56.7
+ 0.3(18)
53.3
±0.8 (18)
59.6
±0.9 (18)
59.5
±1.6(18)Moisture56.8
58.0
±2.8 (56)
±1.5(56)
±4.3 (50)
±2.1 (60)
±2.9 (59)
±4.4 (59)Ash4.4
4.3
4.0
4.9
4.6
4.1
±1.4(4.6)
±0.9 (4.6)
±0.3 (3.8)
±0.6 (4.6)
±0.9 (4.4)
±0.9 (3.7)Body
191.7
197.2
163.0
169.4
212.0
±6.3 (195)
±7.7 (195)
±8.9 (201)
±2.8 (165)
±6.8 (168)
±
HFAL diet (46). The reasons for these incongruities remain to lated it under HFAL conditions; and in marked contrast to our
be determined, but they may be due to (a) limitations in the results, Thompson et al. (24) found that exercised animals
exhibited greater DMBA-induced mammary tumor incidence
resolving power inherent in the model used or (b) differences
in the dose of NMU administered.
than sedentary controls (100% versus 79%).
With regard to caloric reduction, our results unambiguously
As the above indicates, the results of exercise studies to date
show that restricting energy intake to 75% of ad libitum con
are as inconsistent as the exercise protocols used were varied.
trols dramatically reduces tumor yields, regardless of total fat The study of Thompson et al. (24) is perhaps the most instruc
intake. These results are consistent with those reported by Lavik tive, since it underlines what may prove to be a key variable in
and Baumann (47), Kritchevsky (2), Pariza (3), and Thompson
activity studies, namely, the intensity of exercise. When our
(48), but they contradict the earlier studies of Tannenbaum (13, data were grouped into tertiles based on level of activity (Fig.
14) who observed that the influence of fat intake was observable
3), it was found that the least active fertile exhibited the lowest
even in calorie-restricted animals. The reasons for this discrep
tumor incidence; that is, the antipromoting effect of activity
ancy are unclear, but could lie in differences in dietary constit
was inversely proportional to the amount of exercise. The
uents, degree and type of calorie restriction, or in the models
possibility therefore may be considered that intense, forced
used (Tannenbaum used the spontaneous mouse mammary
exercise, typical of treadmill running, may enhance tumor de
tumor model, whereas the others used chemically induced rat velopment (as seen in the Thompson study and in our highest
mammary tumor models).
exercise tertile), while less intense exercise may act to inhibit
It is noteworthy that the degree of tumor inhibition due to tumor development.
caloric restriction was considerably greater in this study than
A second reason for the inconsistencies found in previous
that reported in previous studies (2, 3, 17, 18). The reason for studies may be chronobiological in nature. Under voluntary
this is unclear at present but may be related to the fact that the wheel-running conditions, animals are active during the evening
inbred F344 strain was used in this study, whereas the outbred
hours in keeping with the nocturnal habit of rodents. Treadmill
Sprague-Dawley rat was used in the other studies. Also, in the protocols, however, are, for practical reasons, often conducted
latter, the lipid-soluble procarcinogen DMBA (which requires
during daylight hours when rodents are relatively inactive.
Hence, differences in physiological responses to low and high
host activation) served as the initiating agent in other reported
studies, while in the present study we used the direct-acting,
intensity exercise (49, 50) and to disturbances in circadian light/
water-soluble carcinogen NMU (which does not require host dark rhythms (51) may underlie the variable tumor responses
activation). Hence, despite the fact that diet restriction was noted above.
The quantity and pattern of wheel-running reported here
implemented 2 days after carcinogen administration, differ
ences in the pharmacokinetics and tissue disposition of the two were consistent with those reported by others (52-57). For
example, Tokuyama and Okuda (52) reported that female Wiscarcinogens may have had a bearing on the observed results.
To our knowledge, this is the first report showing that vol
tar rats ran an average of 11,000 revolutions/day. Goodrick
untary exercise can effectively reduce tumor yields and delay (53) reported an average of 9,000 revolutions/day. Moreover,
time of tumor appearance in an experimental tumor model.
Russell et al. (54), Shyu et al., and Goodrick (53) all reported
Several investigators have reported the results of forced exercise
exercise patterns similar to ours, namely, an early peak followed
in the form of a motor-driven wheel cage, or treadmill, on by a gradual decline. Compensatory hyperphagia, resulting
tumorigenesis (19, 20, 23-25). Rusch and Kline (19), in an from voluntary wheel activity, has also been observed by others
early study, using a motor-driven rotating cage, reported a (56, 57). While a decrease in body fat content was consistently
decrease in the growth of a transplantable fibrosarcoma and a noted in the above studies, variable results were obtained with
reduction in weight gain in active versus sedentary ABC mice. regard to body weight gain. In one study, body weight gains
Later, Moore and Tittle (20), using motorized wheel-running,
were negligible (57), while a decrease (53) and an increase (56)
reported a complete absence of DMBA-induced mammary tu
were observed in the other 2 studies. It may be noted that the
mors in exercised animals, a reduction in body weight, and a weight differential (HFEX versus HFAL) was manifested over
the last 10 wk of the experiment when activity levelled off (Fig.
decreased body fat content. More recently, three conflicting
reports (23-25) appeared in abstract form using the DMBA
2), suggesting that high caloric intake continued despite a lower
model and varying amounts of forced treadmill running as the level of energy expenditure during this period. Our finding of
form of exercise. Bennink et al. (23) reported that exercise
reduced body fat content in HFEX and MFR and HFR re
caused a modest reduction in tumor incidence (-16%), a greater
stricted animals (Table 5) indicates that both experimental
one in tumor size (—23%),and a 40% reduction in fat stores
conditions act to alter energy homeostasis leading to mobiliza
tion of lipid from adipose tissue to meet energy needs. Ir.
compared to sedentary controls. Yedinak et al. (24) reported
contrast to Boissoneault et al. (18), however, we did not see a
that activity reduced tumor incidence under LFAL, but stimu
4281
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
FAT, CALORIE RESTRICTION,
EXERCISE, AND MAMMARY CANCER
compensatory increase in lean body mass in either the restricted
or active groups. Our results indicate that, while the fat:lean
body mass ratio may serve as an accurate index of tumor
incidence under certain conditions, the correlation cannot be
generalized, since the LFAL group which had the highest ratio
of fat to protein exhibited reduced, rather than enhanced, tumor
yields (Tables 2 and 3).
Based on the work of Forbes (58) and on his own studies
using the DMBA-tumor model (3), Pariza has proposed that
the tumor-promoting effect of HFAL diets is a direct conse
quence of the host's more efficient use of fat calories compared
to calories from other sources. According to this hypothesis,
animals fed HFAL diets would be expected to store more fat
than animals fed a LFAL diet. Supporting this view is the
finding by Boissoneault et al. (18) that calorie-restricted animals
exhibited low tumor yields even when fed a HFAL diet, and
that LFAL animals weighed less, stored less body fat, and
increased their protein content (lean body mass) compared to
HFAL animals. Contrary evidence, however, can also be cited.
For example, under ad libitum feeding conditions, HFAL and
LFAL animals consumed roughly equal amounts of calories,
but exhibited widely divergent tumor yields (Tables 2 and 3).
Also, not all HFAL diets promote mammary tumorigenesis
equally despite the fact that they contain equivalent high caloric
densities. For example, HFAL diets containing olive (59), co
conut (59), or marine oils (60) lack tumor-promoting effects.
Although the reasons for this are uncertain, some evidence
points to a limiting effect of dietary EFA (10). In the present
study it was found, in contrast to Boissoneault's (18) report,
It is noteworthy that our studies in a laboratory animal are
consistent with the results of epidemiological studies by Frisch
and coworkers (21, 22). In these studies, it was reported that
the prevalence (life-time occurrence) rates of cancers of the
reproductive tract (uterus, cervix, and ovary) and breast were
significantly lower in former women college athletes compared
to nonat hie tes (21 ). In addition, women athletes were taller and
heavier and exhibited lower body fat content than nonathletes.
It was also shown that the prevalence rates of benign breast
disease (a risk factor for breast cancer) and benign tumors of
the reproductive tract were significantly lower in women ath
letes compared to nonathletes (22).
In summary, the promoting effect of HFAL diets and the
antipromoting effects of energy restriction-increased
energy
expenditure, observed in this study, did not correlate with either
body weight, caloric intake, or the ratio of fat to lean body
mass, suggesting that these effects may be mediated indirectly
and, possibly, independently via paracrine, endocrine, or neurohormonal mechanisms.
ACKNOWLEDGMENTS
We thank Dr. R. Landers, Dr. M. Bieter, Dr. S. Appleton (at R. J.
Reynolds now), and Dr. M. Norvell (consultant now) of Best Foods for
their many contributions to the design and implementation of this
study; J. Reinhardt and C. Barrili of the Research Animal Facility for
their expert technical assistance; E. Beauvoir for preparation of the
manuscript; and Dr. D. P. Rose for critical review of the manuscript.
REFERENCES
that body weight, percentage of body fat, and lean body mass
were similar in HFAL, MFAL, and LFAL groups (Table 6;
1. Roe, F. J. C. Are nutritionists worried about the epidemic of tumors in
Fig. 4). Our body weight results are in keeping with the majority
laboratory animals? Proc. Nutr. Soc., 40: 57-65, 1981.
2. Kritchevsky, D., and Klurfeld, D. M. Caloric effects in experimental mam
of reports indicating that, under ad libitum conditions, weight
mary tumorigenesis. Am. J. Clin. Nutr., 45: 236-242, 1987.
gain curves from animals fed HFAL, MFAL, or LFAL diets
3. Pariza, M. W., and Boutwell, R. k. Historical perspective: calories and
are indistinguishable in short-terms studies (9).
energy expenditure in carcinogenesis. Am. J. Clin. Nutr., 45:151-156,1987.
4. Albanes, D. Total calories, body weight, and tumor incidence in mice. Cancer
Various mechanisms, other than caloric excess, have been
Res., 47:1987-1992, 1987.
proposed to explain the promoting effect of HFAL diets, in
5. King, M. M., and McCay, P. B. Potentiation of carcinogenesis by dietary fat:
cluding fat-induced alterations in hormone secretion, eicosais it caused by high energy consumption or is it an effect of fat itself? J.
Nutr., 116:2313-2315,1986.
noid metabolism, intracellular communication, and immune
6. Tannenbaum, A. The genesis and growth of tumors. III. Effects of a high fat
functions (9). But none of these has been conclusively demon
diet. Cancer Res., 2:468-475, 1942.
strated. With regard to the tumor-inhibiting effect of energy
7. Boutwell, R. K., Brush, M. K., and Rusch, H. P. The stimulating effect of
dietary fat on carcinogenesis. Cancer Res., 9: 741-746, 1949.
restriction, a number of hypotheses can be considered. Energy
8. Carroll, K. K., Braden, L. M., Bell, J. A., and Kalamegham, R. Fat and
restriction may lead, for example, to reduced availability of
cancer. Cancer (Phila.), 58:1818-1825, 1986.
9. Cohen, L. A. Dietary fat and mammary cancer. In: B. S. Reddy, and L. A.
nutrients for the growing tumor (5). Energy restriction has also
Cohen (eds.), Diet, Nutrition, and Cancer: A Critical Evaluation, Vol. 1, pp.
been associated with endocrine alterations (61), a decrease in
77-100. Boca Raton, FL: CRC Press, 1986.
circulating proludi n (62), altered EFA content of adipose tissue
10. Ip, C. Fat and essential fatty acid in mammary carcinogenesis. Am. J. Clin.
Nutr., ¥5:218-224, 1987.
stores (63), and enhanced natural killer cell activity (64). More
11. Willett, W. C., and MacMahon, B. Diet and cancer: an overview. N. Engl.
over, since the NMU tumor expresses an activated H-ros on
J. Med., 310:697-703, 1984.
cogene (65), energy restriction may suppress the expression of 12. Wynder, E. L., Rose, D. P., and Cohen, L. A. Diet and breast cancer in
causation and therapy. Cancer (Phila.), 58 (Suppl.): 1804-1813, 1986.
this gene. Dietary energy restriction may also exert an anti13. Tannenbaum, A. Genesis and growth of tumors. II. Effects of caloric restric
tion per se. Cancer Res., 2: 460-467, 1942.
promoting effect by inhibiting overall body weight gain (4). It
14. Tannenbaum, A. The dependence of tumor formation on the composition of
is well known that caloric restriction results in a more aggressive
the calorie-restricted diet as well as on the degree of restriction. Cancer Res.,
and irritable animal (66), a fact noted in this study; hence
5:616-625,1945.
neurochemical changes, associated with energy restriction, may 15. Tucker, M. J. The effect of long-term food restriction on tumours in rodents.
Int. J. Cancer, 23:803-807, 1979.
also be associated with reduced tumor yields (67).
16. Ross, M. H., Lustbader, E. D., and Bras, G. Body weight, dietary practices,
How increased energy expenditure may inhibit mammary
and tumor susceptibility in the rat. J. Nati. Cancer Inst., 71: 1041-1046,
1983.
tumorigenesis is only remotely understood. Most studies on the
17. Kritchevsky, D., Weber, M. M., and Klurfeld, D. M. Dietary fat versus caloric
potential health benefits of exercise have focused on longevity
content in initiation and promotion of 7,12-dimethylbenz(a)anthracene-in(68), circulatory disorders, and related changes in plasma lipoduced mammary tu nu >r¡genesisin rats. Cancer Res., 44: 3174-3177,1984.
18. Boissoneault, G. A., Elson, C. E., and Pariza, M. W. Net energy effects of
proteins (69, 70). There have been some studies linking changes
dietary fat on chemically-induced mammary carcinogenesis in F344 rats. J.
in circulating hormone titers (71, 72), prostanoids (73), ßNati. Cancer Inst., 76: 335-338, 1986.
endorphin levels (74, 75), and enhanced immune functions (76) 19. Rusili, H. P., and Kline, B. E. The effect of exercise on the growth of a
mouse tumor. Cancer Res., 4: 116-118, 1944.
io exercise but, as of the present, it is unclear precisely how 20. Moore, C., and Tittle, P. W. Muscle activity, body fat, and induced rat
mammary tumors. Surgery (St. Louis), 73: 329-332, 1973.
these changes may be related to carcinogenesis.
4282
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
FAT, CALORIE RESTRICTION, EXERCISE, AND MAMMARY CANCER
21. Frisch, R. E., Wyshak, G., Albright, N. L., et al. Lower prevalence of breast
cancer and cancers of the reproductive system among former college athletes
compared to non-athletes. Br. J. Cancer, 52: 885-891, 1985.
22. Wyshak, G., Frisch, R. E., Albright, N. L., Albright, T. E., and Schiff, I.
Lower prevalence of benign diseases of the breast and benign tumors of the
reproductive system among former college athletes compared to non-athletes.
Br. J. Cancer, 54:841-845, 1986.
23. Thompson, H. J., Roñan,A. M.. and Tagliaferro, A. R. Effect of exercise on
the enhancement of mammary tumorigenesis by high dietary fat. Proc. Am.
Assoc. Cancer Res., 28:157, 1987.
24. Bennink, M. R., Palmer, H. J., and Messina, M. J. Exercise and caloric
restriction modify rat mammary carcinogenesis. Fed. Proc., 45:1087, 1986.
25. Yedinak, R. A., Layman, D. K., and Milner, J. A. Influences of dietary fat
and exercise on DMBA-induced mammary tumors. Fed. Proc., 46: 436,
1987.
26. Riley, V. Psychoneuroendocrine influences on immuno-competance and neo
plasia. Science (Wash. DC), 212:1100-1109, 1981.
27. Steplewski, Z., Vogel, W. H., Ehya, H., Poropatich, C., and Smith, J. M.
Effects of restraint stress on inoculated tumor growth and immune response
in rats. Cancer Res., 45:5128-5133,1985.
28. Pollard, M., and Luckert, P. H. Tumorigenic effects of direct- and indirect
acting chemical carcinogens in rats on a restricted diet. J. Nati. Cancer Inst.,
74:1347-1349, 1985.
29. Cullino, P. M., and Pettigrew, H. M., and Grantham, F. H. JV-Nitrosomethylurea as mammary gland carcinogen in rats. J. Nati. Cancer Inst., 54:401414, 1975.
30. Williams, J. C., Gusterson, B., Humphreys, J., et al. W-Methyl-A'-nitrosoureainduced rat mammary tumors. Hormone responsiveness but lack of sponta
neous metastasis. J. Nati. Cancer Inst., 66: 147-151, 1981.
31. Rose, D. P., and Noonan, J. J. Hormone dependence of rat mammary tumors
induced by Ar-nitrosomethylurea. Eur. J. Cancer, 17: 1347-1358, 1981.
32. Aratali. B. M., Finegan, H. M., RÅ“, J., et al. Hormone dependency in Nnitrosomethylurea-induced rat mammary tumors. Endocrinology, 111: 584588, 1982.
33. Bieri, J. (... Stoewsand, G. S.. Briggs, G. M., et al. Report of the American
Institute Ad Hoc Committee on Standards for Nutritional Studies. J. Nutr.,
707:1340-1348, 1977.
34. Bieri, J. G. Second report of the Ad Hoc Committee on Standards for
Nutritional Studies. J. Nutr., 110:1726, 1980.
35. Cohen, L. A., and Thompson, D. O. The influence of dietary medium chain
triglycérideson rat mammary tumor development. Lipids, 22: 455-461,
1987.
36. Chan, P. C., and Dao, T. L. Enhancement of mammary carcinogenesis by a
high fat diet in Fischer, Long-Evans, and Sprague-Dawley rats. Cancer Res.,
47:165-167, 1981.
37. King, M. M., and McCay, P. B. Possibilities of dietary fat and antioxidants
as modulators of mammary carcinogenesis. In: Cancer and the Environment:
Possible Mechanisms of Thresholds for Carcinogens and Other Toxic Sub
stances, pp. 285-306. New York: Mary Ann Lieben, Inc., 1983.
38. ( ¿animal.E. B., Carroll, K. K., and Plunkett, E. R. Effects of dietary fat on
mammary carcinogenesis by 7,12-dimethyl[a]benzanthracene. Cancer Res.,
27:1737-1742, 1967.
39. Newberne, P. M., Bieri, J. G., Briggs, G. M., et al. Control of diets in
laboratory animal experimentation. Anim. Res. News, 21: Al, 1978.
40. Hirayama, T. Epidemiology of breast cancer with special reference to the
role of diet. Prev. Med., 7: 173-195, 1978.
41. Young, S., and Hallowes, R. C. Tumors of the mammary gland. In: V. X.
Turosov (ed.), Pathology of Tumors in Laboratory Animals, Vol. 1, pp. 3174. Lyon: IARC, 1973.
42. Gart, J. J., Chu, K. C., and Tarone, R. E. Statistical issues in interpretation
of chronic bioassay tests for carcinogenicity. J. Nati. Cancer Inst., 62: 957974, 1979.
43. Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete
observation. J. Am. Stat. Assoc., 53: 457-481, 1958.
44. Bartholomew, D. J. A test of homogeneity for ordered alternatives. Biometrika, 46:328-335, 1959.
45. Helwig, J. T., and Council, K. A. (eds.). Statistical Analysis User's Guide, pp.
245-265. Raleigh, NC: SAS Institute, 1979.
46. Cohen, L. A., Choi, K., Weisburger, J. H., and Rose, D. P. Effect of varying
proportions of dietary fat on the development of A'-nitrosomethylurea-induced rat mammary tumors. Anticancer Res., 6: 215-218, 1986.
47. Lavik, P. S., and Baumann, C. A. Further studies on the tumor-promoting
action of fat. Cancer Res., 3: 749-756, 1943.
48. Thompson, H. J., Meeker, L. D., Tagliaferro, A. R., and Roberts, J. S. Effect
of energy intake on the promotion of mammary carcinogenesis by dietary
fat. Nutr. Cancer, 7: 37-41, 1985.
49. Williams, P. T., Wood, P. K., Haskell, W. L., and Vranizan, M. A. The
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
effects of running mileage and duration on plasma lipoprotein levels. J. Am.
Med. Assoc., 247: 2674-2679, 1982.
Chang, F. E., Dodds, W. G., Sullivan, M., Kim, M. H., and Malarkey, W.
B. The acute effects of exercise on prolactin and growth hormone secretion:
comparison between sedentary women and women runners with normal and
abnormal menstrual cycles. J. Clin. Endocrino!. Metab., 62: 551-556, 1986.
Halberg, F. Quo vadis basic and clinical chronobiology: promise for health
maintenance. Am. J. Anat., 168: 543-594, 1983.
Tokuyama, K., and Okuda, H. Fatty acid synthesis in adipose tissues of
physically trained rats in vivo. Am. J. Physiol., 245: E8-E13, 1983.
Goodrick, C. L. Effects of long-term voluntary wheel exercise on male and
female Wistar rats. 1. Longevity, body weight, and metabolic rate. Gerontol
ogy, 26: 22-23, 1980.
Sliyu. B. C., Anderson, S. A., and Thoren, P. Spontaneous running in wheels.
A microprocessor assisted method for measuring physiological parameters
during exercise in rodents. Acta Physiol. Scand., 121:103-109, 1984.
Russell, J. C., Epling, W. F., Perce, D., Amy, R. M., and Boer, D. P.
Induction of voluntary prolonged running by rats. J. Appi. Physiol., 63:
2549-2553, 1985.
Tokuyama, K., Saito, M., and Okuda, H. Effect of wheel running on food
intake and weight gain of male and female rats. Physiol. Behav., 28: 899903, 1982.
Anantharaman-Barr, H. G., and Decombaz, J. The effect of voluntary activity
on growth and energy expenditure of female rats. Int. J. Vit. Nutr. Res., 57:
341, 1987.
Forbes, E. B., Swift, R. W., Elliott, R. F., et al. Relation of fat to economy
of food utilization. I. By the growing albino rat. J. Nutr., 31: 203-212,1946.
Cohen, L. A., Thompson, D. O., Maeura, Y., Choi, K., Blank, M. E., and
Rose, D. P. Dietary fat and mammary cancer. I. Promoting effects of different
dietary fats on A'-nitrosomethylurea-induced rat mammary tumorigenesis. J.
Nati. Cancer Inst., 77: 33-42, 1986.
Karmali, R. A. Eicosanoids in neoplasia. Prev. Med., 16:493-502, 1987.
Yang, H., Crée,
T. C., and Schalch, D. S. Effect of a carbohydrate-restricted,
calorie-reduced diet on the growth of young rats and on serum growth
hormone, somatomedins, total thyroxine and triiodothyronine, free T4 index,
and total corticosterone. Metabolism, 36: 794-798, 1987.
Sarkar, M. H., Fernandez, G., Telang, N. T., Kourides, I. A., and Good, R.
A. Low calorie diet prevents the development of mammary tumors in C3H
mice and reduces circulating prolactin level, murine mammary tumor virus
expression, and proliferation of mammary alveolar cells. Proc. Nati. Acad.
Sci. USA, 79: 7758-7762, 1982.
Pittet, P. G., Jequier, E., and Kriz, E. Food restriction and associated changes
in adipose tissue fatty acid composition, at 6 sites in adult male rats. Int. J.
Vit. Nutr. Res., 52: 219-223, 1982.
Weindruck, R. H., Devens, B. H., Raff, H. U., and Walford, R. L. Influence
of dietary restriction and aging on natural killer cell activity in mice. J.
Immunol., 130: 993-996, 1983.
Zarbl, H., Sukumar, S., Arthur, A. V., et al.. Direct mutagenesis of Ha-ruv1 oncogenes by A'-nitroso-A'-methylurea during initiation of mammary car
cinogenesis in rats. Nature (Lond.), 315: 382-385, 1985.
Lore, R., Gottdiener, C., and Delahunty, M. J. Lean and mean rats: some
effects of acute changes in the food supply upon territorial aggression.
Aggress. Behav., 12:409-415, 1986.
Albee, R. R., Mattsson, J. L., Yano, B. L., and Chang, L. W. Neurobehavioral
effects of dietary restriction in rats. Neurotoxicol. Teratol., 9:203-211,1987.
Holloszy, J. O., and Smith, E. K. The effects of exercise on longevity of rats.
Fed. Proc., 46: 1850-1853, 1987.
Deshaies, Y., Leblanc, J., and Richard, D. Influence of palatable high-fat diet
and exercise training on the high-density lipoprotein to total cholesterol ratio
in the rat. Metabolism, 32:62-65, 1983.
Haskell, W. L. The influence of exercise training on plasma lipids and
lipoproteins in health and disease. Acta Med. Scand. Suppl., 777: 25-37,
1987.
Villanueva, A. L., Schlosser, C., Hopper, B., Liu, J. H., Hoffman, I., and
Robar, R. W. Increased Cortisol production in women runners. J. Clin.
Endocrinol. Metab., 63:133-136, 1986.
Carlberg, K. A., and Fregly, M. J. Disruption of estrous cycles in exercised
trained rats. Proc. Soc. Exp. Biol. Med., 779: 21-24, 1985.
Rauramaa, R. Physical activity and prostanoids. Acta Med. Scand. Suppl.,
7/7: 137-142, 1986.
LeGoaster, J., and LeMagnen, J. Neurobiology and cancer: possible role of
/3-endorphins in cancer growth regulation. Cancer J., 7: 204-207, 1987.
Goldfarb, A. H., Hatfield, B. D., Sforzo, G. A., and Flynn, M. G. Serum ßendorphin levels during a graded exercise test to exhaustion. Med. Sci. Sports
Exer., 19: 78-82, 1986.
Simon, H. B. The immunology of exercise. J. Am. Med. Assoc., 252: 27352738, 1984.
Cohen, L. A. The fat effect on N-Nitrosomethylurea (NMU)-induced fat
mammary carcinogenesis: inhibition by (a) voluntary energy expenditure and
(b) caloric restriction. Proc. Am. Assoc. Cancer Res., 28: 155, 1987.
4283
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.
Influence of Dietary Fat, Caloric Restriction, and Voluntary
Exercise on N-Nitrosomethylurea-induced Mammary
Tumorigenesis in Rats
Leonard A. Cohen, Keewhan Choi and Chung-Xiou Wang
Cancer Res 1988;48:4276-4283.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/48/15/4276
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1988 American Association for Cancer Research.