Carcinogenesis vol.26 no.6 pp.1077--1084, 2005
doi:10.1093/carcin/bgi051
Advance Access publication March 3, 2005
Prevention of mouse skin tumor promotion by dietary energy restriction requires
an intact adrenal gland and glucocorticoid supplementation restores inhibition
Jeanne W.Stewart1, Ken Koehler2, William Jackson1,
Jacqueline Hawley1, Weiqun Wang1,4, Angela Au1,
Ron Myers3 and Diane F.Birt1,
1
Department of Food Science and Human Nutrition, 2Department of Statistics
and 3Department of Veterinary Pathology, Iowa State University,
Ames, IA 50011, USA
Email: [email protected]
Our laboratory has demonstrated in the previous studies
that dietary energy restriction (DER) inhibited the promotion of skin tumorigenesis and others have found that
adrenalectomy may reverse that inhibition. The purpose
of the research reported here was to determine if circulating corticosterone (CCS) may be the adrenal hormone responsible for DER prevention of skin carcinogenesis. Female
SENCAR mice were initiated with 7,12-dimethylbenzanthracene (DMBA) and promoted with 12-O-tetradecanoylphorbol-13-acetate (TPA) in either sham-operated or
adrenalectomized (ADX) mice fed ad libitum (AL) or
energy restricted diets. DER was 60% of the AL calorie
intake with the removal of energy from fat and carbohydrate. CCS, the main glucocorticoid hormone secreted by
the murine adrenal gland, was added to the drinking water
of AL/ADX and DER/ADX groups to determine the role of
CCS in the DER inhibition of tumor development. In
sham-operated groups, DER compared with AL-fed mice
experienced significantly decreased papilloma incidence
and multiplicity (P 5 0.0001). ADX did not alter papilloma
incidence or multiplicity in AL-fed mice but ADX partially
reversed the inhibition of papilloma multiplicity and incidence in DER mice. CCS supplementation to both DER/ADX
and AL/ADX mice resulted in reduced papilloma incidence
and multiplicity. In DER/ADX mice, CCS dramatically
reduced papilloma rates while in AL/ADX mice CCS
reduced the papilloma rates to those seen in the DER sham
group. DER significantly reduced carcinoma multiplicity
mean counts per effective animal (P 5 0.0001) compared
with AL-fed groups in sham and ADX/CCS groups. DER/
ADX mice lost the carcinoma multiplicity protection seen in
sham/DER mice. CCS treatment of ADX mice significantly
decreased total carcinoma (in situ and invasive) incidence
rates per effective animal (P 5 0.0003). ADX followed by
CCS treatment in the DER mice resulted in the lowest
carcinoma incidence and multiplicity. Thus, DER-inhibition of skin tumorigenesis was mediated at least in part
Abbreviations: AP-1, activator protein-1; AL, ad libitum; ADX,
adrenalectomized; CCS, circulating corticosterone; DER, dietary energy
restriction; DHEA, dehydroepiandrosterone; DMBA, 7,12-dimethylbenzanthracene; ERK, extracellular signal regulated kinase; IGF-1, insulin
growth factor-1; TPA, 12-O-tetradecanoylphorbol-13-acetate.
Carcinogenesis vol.26 no.6 # Oxford University Press 2005; all rights reserved.
Introduction
With the growing evidence that obesity is a risk factor for
human cancer (1) and the increasing rate of obesity throughout
the world (2), determining the mechanisms whereby dietary
intake modulates cancer has taken on a new significance.
There has long been evidence that restricting dietary intake
could significantly reduce cancer rates. In particular, studies
demonstrated that when mice were fed a calorie restricted
(60% of control calories) diet instead of an ad libitum (AL)
diet the mice exhibited ~40% decreased tumor incidence in
skin and mammary tissues (3,4). In addition, a more moderate
restriction (80% of control intake) was also effective in skin
cancer prevention (5). Since these early findings, studies have
demonstrated dietary or calorie restriction are effective in the
prevention of liver, lung, colon, breast and skin cancers (6--9).
Previous research in our laboratory that is relevant to assessing the mechanism of dietary energy restriction (DER) prevention of skin cancer, compared dietary restriction (reduction of
all dietary components) and DER (reduction of calories from
fat and/or carbohydrates while maintaining other nutrient
intakes). The study showed that both dietary restriction and
DER reduced papilloma and carcinoma incidence, and multiplicity and also delayed papilloma appearance but that DER
was the most effective strategy (7). Furthermore, our laboratory has shown that the DER inhibition of tumorigenesis occurs
during the promotion stage of 7,12-dimethylbenzanthracene
(DMBA) initiated and 12-O-tetradecanoylphorbol-13-acetate
(TPA) promoted mouse skin tumorigenesis (10). Research in
other laboratories showed that the adrenal gland was required
for diet restriction prevention of skin and lung tumorigenesis
(11,12). The studies of skin cancer determined that feeding less
of the control diet reduced DMBA-initiated and TPA-promoted mouse skin carcinogenesis in intact mice but not in
adrenalectomized (ADX) mice. These studies fed less of all
dietary components (diet restriction), they did not perform
DER as was used in the present investigation. Furthermore,
they did not assess the key hormone from the adrenal gland
that was responsible for the dependence.
The adrenal gland cortex produces three major classes of
lipid soluble hormones derived from cholesterol: mineralocorticoids (controls levels of sodium and potassium), glucocorticoids (controls general cell function and especially maintains
glucose homeostasis during food deprivation) and androgens
(similar hormonal function as the male sex hormones) (12,13).
The adrenal medulla produces the catecholamines: epinephrine and norepinephrine (13). Studies reported by Pashko
and Schwartz demonstrated that adrenectomy reverses the
1077
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
4
Present address: Department of Human Nutrition, Kansas State University,
Manhattan, KS 66506, USA
To whom correspondence should be addressed
through CCS. However, CCS was more effective in preventing papillomas and carcinomas in DER/ADX mice than
in AL/ADX mice, suggesting that other factors may also be
involved in the DER prevention of tumor formation.
J.W.Stewart et al.
Materials and methods
Animals and diets
The Iowa State University Committee on Animal Care approved the procedures used for the two-stage tumorigenesis model in mice. Six-week-old
SENCAR female animals were obtained from the National Institute of Health
(NIH) in Frederick, MD. Animals were allowed a 1-week acclimatization
period on an AL-fed, control diet (AIN-76/93) (21) as seen in Table I (custom
made by Harlan Teklad, Madison, WI) before surgeries were performed:
animals were then either sham operated or bilateral adrenalectomy was performed (17,22) as indicated in Figure 1. The mice were housed individually in
humidity and temperature controlled rooms with a 12 h light/dark cycle. The
mice were allowed a 2-week recovery period before treatment began. A dorsal
Table I. Experimental diets fed to SENCAR mice
Diet components
Corn oil
Casein
DL-Methionine
Dextrose
Dextrin
Fiber
AIN-93 mineral mix
AIN-93 vitamin mix
Choline bitartrate
Total amount of food
g fed: g control consumed
Control (AL)
5.0
20.0
0.3
15.0
49.9
5.0
3.5
1.0
0.3
100
1.0
40% DER
As formulated
As given
3.8
31.1
0.5
11.5
38.0
7.8
5.4
1.5
0.4
100
2.4
20.0
0.3
7.4
24.3
5.0
3.5
1.0
0.3
64.0
0.64
Pelleted diets were custom made by Harlan Teklad, Madison, WI. The
catalogue numbers were TD99433 for the control (AL) diet and TD99467 for
the DER diet. The amount of DER diet that was offered was calculated
weekly as 64% of the weight consumed by the control mice which provided
the DER mice a 40% reduction of calories from fat and carbohydrate.
1078
patch of skin of each mouse was shaved and tumor development was initiated
on mice that did not have hair regrowth 48 h later with 10 nmol DMBA/200 ml
acetone. Control mice were treated with acetone alone. One half of the treated
groups of animals were placed on 20% DER in the first 2 weeks and then
switched to 40% DER for the rest of the experiment as seen in Table I (Harlan
Teklad, Madison, WI). We pair-fed DER mice at 17:00 h since Gallo and
Weinberg (23) found that mice fed on a food restricted diet (75 % of AL,
control diet) at 17:00 h exhibited a plasma corticosterone pattern similar to
mice fed AL, control diet. The mice were given a tumor promoter, 3.2 nmol
TPA/200 ml acetone twice a week starting at week 3 after DMBA initiation and
continuing for 15 weeks. Drinking water solutions were varied according to
treatment: mice that were sham operated received tap water with 0.6% ethanol,
while ADX mice received either 0.9% NaCl with 0.6% ethanol or 60 mg/ml
corticosterone (ICN Biomedicals, Aurora, OH), dissolved in ethanol and
diluted with tap water for a final concentration of 0.6% ethanol. ADX mice
needed supplementation of NaCl or CCS to prevent the loss of essential
electrolytes into urine owing to the lack of mineralcorticoid activity.
Mice were weighed, palpated and visually inspected every week for evidence of tumors or carcinomas and all data were recorded. The week of
initiation was counted as week 0 for reporting results. Animals were continued
on the experimental diets even after 15 weeks of TPA treatment to allow the
development of papillomas and carcinomas. If at any time during the experiment the animal suffered pain from the progression of a tumor to a carcinoma
or any other serious health problem (ulceration and scratching were observed
in some groups in the present study) the mouse was killed, and visual papillomas and suspected carcinomas were collected. At week 31 all remaining mice
were killed by decapitation to eliminate hormonal changes in the blood due to
anesthesia or stress. Sham-treated mice were killed between 7:30 and 8:30 AM
and the ADX mice were killed between 6:30 and 7:30 PM; these timings were
chosen for sacrifice since the difference in corticosterone levels between AL
mice and DER mice were expected to be the maximum during these time
periods (21). All ADX mice were visually inspected for complete removal of
adrenal tissue. If there was any evidence for the presence of adrenal tissue, the
animal was removed from the statistical analysis.
Plasma analysis
The mice that survived up to the termination of the experiment were killed by
decapitation. Blood was collected from the trunk into EDTA-coated tubes. The
collected samples of blood were centrifuged at 825 g for 10 min and plasma
collected was stored at 70 C. Corticosterone concentration in the plasma was
determined by radioimmunoassay (ICN Biomedicals, Costa Mesa, CA). Insulin growth factor-1 (IGF-1) in the plasma was quantified using a radioimmunoassay kit (Nichols Institute Diagnostics, San Clemente, CA; IGF-1 IRMA 100T
Kit; catalog No. 40-2250).
Tissue handling and diagnostic criteria
Papillomas were visually diagnosed on the treated dorsal skin if a papillary
mass was projecting from the skin surface but there was no evidence of the
lesion developing into a carcinoma. Skin lesions that were suspected of being
carcinomas based on appearance, and any organ that had an abnormal appearance were collected from mice that were exsanguinated early owing to pain
caused by the carcinoma or ill health and from mice that were exsanguinated at
week 31. Tissues were fixed by immersion in 10% buffered neutral formalin
for at least 48 h. Lesions and masses from each mouse were bisected, routinely
processed, embedded in paraffin, sectioned at 6 mm thickness and stained with
hematoxylin and eosin (H&E).
The skin samples collected for histology were diagnosed as papilloma if a
keratinized papillary mass was projecting from the skin surface with a normal
basal layer. Histologically diagnosed papillomas were added to the much
larger count of visually identified papillomas for statistical analysis. In situ
carcinomas were diagnosed if there was an area of atypia or dysplasia, aberrant
keratinization in the basal layers, mitotic activity was excessive or suprabasilar, there was an aberrant growth of the basal layer or if the basal layer
lacked reasonable polarity in a significant area of the papilloma. Invasive
carcinomas were diagnosed if there was a marked pleomorphism of proliferative keratinocytes, an invasion of the dermis with neoplastic cells or significant breaching of basement membrane. The visual identification of carcinomas
was 89% accurate (208 carcinomas verified histologically out of 234 lesions
visually suspected of being carcinomas) in comparison with histological verification as carcinoma and there was no significant difference in accuracy
amongst the DMBA/TPA treated groups. All carcinomas were histologically
verified and included in the statistical analysis.
Statistical analysis
The weekly food consumption levels of all the AL mice were analyzed by
ANOVA using the MIXED procedure in SAS with multiple comparisons by
least squares means for the first 3 weeks prior to treatment with the initiator
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
dietary-restriction-inhibition of TPA promotion of skin papillomas (11,12). In these studies dietary restriction increased the
circulating corticosterone (CCS) but there was no difference in
the levels of dehydroepiandrosterone (DHEA) or aldosterone
in cases where these hormones were measured (11,12,14). In
our earlier studies of DER, with reduction of fat and carbohydrate calories, we observed elevated levels of corticosterone
in DER mice but found no evidence of activated glucocorticoid receptors in the skin of DER mice (15).
Several lines of evidence suggest that glucocorticoid hormone may be involved in the prevention of skin carcinogenesis
in DER animals. First, it is known that energy restriction
elevated circulating glucocorticoid hormones as noted above
(15). Second, glucocorticoid hormones were demonstrated in a
number of laboratories to be potent inhibitors of mouse skin
tumorigenesis (16,17). Finally, the importance of the glucocorticoid receptor in interfering with a key transcription factor,
activator protein 1 (AP1), in mouse skin carcinogenesis suggests potential mechanisms for DER elevated corticosterone in
preventing mouse skin tumorigenesis (18,19). Studies focused
on DER prevention of mammary carcinogenesis suggested
that changes in CCS metabolism contributed to, but did not
explain, cancer prevention in this model (20).
Our objectives in this research were to determine if the DER
(restriction of energy from fat and carbohydrate) inhibition of
skin tumorigenesis in intact mice, would be lost in ADX/DER
mice, whether supplementation of the ADX mice with corticosterone would restore the inhibition of skin tumorigenesis
and whether the corticosterone would inhibit skin carcinogenesis when administered to ADX mice fed the control diet.
Effect of ADX and CCS on DER inhibited skin tumorigenesis
Initiator/
Promoter
DMBA-TPA
Diet
AD LIB
Surgery
Sham
ACETONE
DER
ADX
Sham
AD LIB
ADX
Sham
DER
ADX
Sham
Drinking soln* H2O Saline CCS H2O Saline CCS H2O Saline CCS
Number of mice
(wk 0)
31
30
39
32
37
40
8
10
ADX
H2O Saline CCS
10
8
10
11
* All contained 0.6 % Ethanol
Table II. Food consumption of ad libitum (AL) fed mice
Weeks
DMBA/TPA
N
3 to 1
6 to 10
16 to 20
26 to 30
29
29
29
17
Acetone
N
AL/sham
3.8
4.3
4.9
4.7
a
0.1
0.1a
0.1a
0.2a
31
31
29
17
AL/ADX/saline
3.8
4.3
4.9
4.6
a
0.1
0.1a
0.1a
0.2a
N
12
12
12
10
AL/ADX/CCS
3.6
4.0
4.2
4.1
a
0.1
0.2a,b
0.2b
0.2b
N
8
8
8
8
N
AL/sham
3.9
3.7
3.9
4.0
a
0.1
0.2b
0.2b
0.2b
9
9
9
9
AL/ADX/saline
3.8
3.8
3.7
3.8
a
0.1
0.2b
0.2b
0.2b
N
AL/ADX/CCS
5
5
5
5
3.7
3.9
3.9
4.4
0.1a
0.3a,b
0.3b
0.3a
Values are shown as mean g/day SEM. Food consumption was statistically analyzed from 3 to 30 weeks of the experiment using 5-week averages. Data
for weeks 0--5, 11--15 and 21--25 are similar but not shown. The DMBA/TPA effect on food consumption was significant, P 5 0.0069, 0.0001 and 0.0236
for weeks 6--10, 16--20 and 26--30, respectively. Individual means across rows were analyzed by ANOVA--Mixed and then compared with multiple comparisons
by least squares means. Row means with different superscript letters are significantly different at the P 5 0.05 level and values with more than one letter are
not significantly different from means sharing either of the letters. Analysis by Anova--Mixed revealed that there was a significant interaction between
DMBA/TPA and corticosterone treatment at weeks 26--30 for DMBA/TPA and acetone treated AL/ADX/CCS mice.
and 5-week blocks during the rest of the experiment. All the animals on the AL
diet were included in the data set until they were terminated from the experiment owing to ill health or the completion of the experimental period. Logrank and Wilcoxon tests were used to compare Kaplan--Meier estimates of
survivor curves for the six groups of DMBA/TPA treated mice (24). In this
analysis, survival times for animals that completed the entire 31 weeks of the
experiment were treated as censored at 31 weeks.
Body weights were analyzed with repeated ANOVA measures using the
MIXED procedure in SAS with multiple comparisons by least squares means
at each time point. The variable of surgery had three possibilities: sham/tap
water þ 0.6% ethanol, ADX/0.9% NaCl in 0.6% ethanol or ADX/60 mg
cortiosterone in 0.6% ethanol. Since carcinoma multiplicity at the time of death
exhibited Poisson distributions (variance ¼ mean), log-linear model analysis
using the GENMOD procedure in SAS was used to analyze treatment effects on
carcinoma multiplicity with multiple treatment comparisons made with least
squares means. Effective mice for the carcinoma analysis were defined as the
mice treated with carcinogen that survived at least 23 weeks. Data for mice that
survived 523 weeks were excluded from the carcinoma analysis. Papilloma
multiplicity at specific time points were also analyzed with the GENMOD
procedure, using all the mice that survived to the specified time point, but greater
variability necessitated the use of the negative binomial distribution (variance >
mean). Fisher’s exact test was used as an overall test for treatment effects on final
carcinoma incidence or papilloma incidence at each time point. If the overall test
showed significant differences at the 0.05 level, Fisher’s exact test was used to
compare each pair of treatments. Plasma corticosterone and IGF were analyzed in
the surviving mice at week 31 by ANOVA, using the GLM procedure in SAS
with multiple comparisons by least squares means.
Results
Food consumption
Consumption of diet by AL-fed mice is shown in Table II.
Within the DMBA/TPA treated mice the AL/sham and
AL/ADX/saline groups consumed significantly more control
diet from 16 to 30 weeks than the AL/ADX/CCS mice.
In comparison with acetone treated mice, DMBA/TPA treated
AL/sham and AL/ADX/saline groups consumed significantly more control diet from 6 to 30 weeks. In contrast,
within the AL/ADX/CCS group, the acetone treated mice
consumed more control diet after week 26 than the DMBA/
TPA treated mice.
Body weight
The mice on DER showed a significant loss in body weight
during the first 4 weeks following the switch to this diet but
maintained the reduced body weight for the rest of the experimental period (Figure 2). The AL-fed mice showed increase in
body weight as the experiment progressed (Figure 2). At week
31 the DER-fed mice weighed significantly less than the ALfed mice (P 5 0.0001) as shown in Figure 2. Animals treated
with acetone exhibited the same significant difference in body
weights between the AL-fed mice versus the DER-fed mice
(data not shown).
The changes in the body weight at the beginning of the
experiment and weeks 4, 8 or 31 are provided in Table III.
It is to be noted that the maximum body weight gains were in
the AL/ADX/CCS treated groups in both the carcinogen
(DMBA/TPA) and vehicle (acetone) administered groups. In
contrast, among the DER treated groups, body weight changes
did not differ between the sham, ADX/saline or ADX/CCS
treated groups.
1079
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
Fig. 1. Experimental design.
J.W.Stewart et al.
Table III. Percentage of body weight change
Percentage change compared with week 0
DMBA/TPA
AL/sham
AL/ADX/saline
AL/ADX/CCS
DER/sham
DER/ADX/saline
DER/ADX/CCS
Acetone
AL/sham
AL/ADX/saline
AL/ADX/CCS
DER/sham
DER/ADX/saline
DER/ADX/CCS
P5
N
0--4 Weeks
N
0--8 Weeks
N
0--31 Weeks
31
25
37
32
21
36
3 11a
2 2a
9 1ab
17 2cd
12 2cd
17 2cd
31
23
34
32
18
36
8 2a
6 2a
18 2b
27 2de
20 3d
27 2de
18
10
8
30
10
13
19 3a
19 4a
41 4b
20 3d
15 4d
16 4d
8
10
9
8
9
10
2 3a
0 3a
9 3ab
15 3cd
9 3c
13 3cd
0.0004
8
10
10
8
9
10
6 4a
1 3ac
22 3b
22 4d
14 4d
16 4d
0.0001
7
9
5
8
5
7
17 5a
7 4ac
33 6b
19 5d
11 5d
16 5d
0.0002
The changes in body weight were compared with week 0. Values represent
mean SEM. P 5 0.0001. Individual means within the column were
analyzed by ANOVA--Mixed and then compared with multiple comparisons
by least squares means. Column means with different superscript letters are
significantly different at the P 5 0.05 level and values with more than one
letter are not significantly different from means sharing either of the letters.
Surgery became a significant factor for 0--8 and 0--31 weeks (P ¼ 0.0004
and 30.0009, respectively). For the DMBA/TPA treated mice, the interaction
between diet and surgery effects was significant for 0--4, 0--8, 0--31 weeks,
mostly owing to a larger difference in mean body weight gains between the
AL and DER diets for the ADX/CCS surgery group than for the sham or
ADX/saline surgery group.
Survival
Mice in the DER/sham group exhibited the best survival with
30/32 (94%) of the DMBA/TPA treated mice surviving up to
31 weeks. Intermediate survival was observed in the AL/sham
group with 18/31 (58%) of the DMBA/TPA treated mice
surviving the experiment. All ADX mice experienced poorer
survival with an average of 41/139 (29%) of the DMBA/TPA
treated mice alive at week 31 (P 5 0.0004). In general, mice in
the ADX groups died gradually throughout the experiment.
However, the DER/ADX group had 19/37 (51%) mice surviving
1080
Fig. 3. Papilloma incidence (A) and multiplicity; mean count/mouse (B).
Data shown are for mice treated with DMBA/TPA. Values represent mean SEM, N ¼ 8--38. (A) Incidence analyzed by Fisher’s exact test. (B) Analyzed
by GENMOD with multiple comparisons by least squares means.
P 5 0.0001 for differences by diet at each time point 8 weeks. Means at
each time point with different superscript letters are significantly different at
the P 5 0.05 level and values with more than one letter are not significantly
different from means sharing either of the letters. Only one papilloma was
seen in one acetone treated mice fed the AL/ADX/saline diet while 75%
(40/53) of all of the acetone treated mice survived up to the end of the
experiment.
at week 6 whereas the other three ADX groups had 91/101
(90%) mice surviving at that time in the experiment.
Papilloma incidence
Mice in the AL/sham and AL/ADX/saline groups exhibited a
significantly higher papilloma incidence at 8 weeks (P 5
0.0001) compared with other mice (Figure 3A). By week 10
the DER/ADX/saline treated mice were also exhibiting a significantly increased papilloma incidence compared with the
mice fed the other DER diets and the AL/ADX/CCS. The mice
receiving the DER/ADX/CCS protocol exhibited the most
pronounced inhibition of papilloma incidence consistently
throughout the study (Figure 3A). At week 31 DER significantly reduced papilloma incidence and CCS reduced papilloma incidence in both AL/ADX and DER/ADX groups
(P 5 0.0001).
Papilloma multiplicity: mean count per animal
Papilloma multiplicity was defined as the number of papillomas per animal. Mice in the AL/sham and AL/ADX/saline
groups exhibited a significant increase in papilloma multiplicity by 9 weeks of treatment (P 5 0.03) and by week 11 DER/
ADX/saline treated mice exhibited a significant increase in
papilloma multiplicity compared with the other DER and
AL/ADX/CCS treated mice (P 5 0.003) (Figure 3B). At
week 31 the number of papillomas on DER/ADX/saline mice
was not significantly different from the number on the groups
that received AL diets or the DER/sham mice (Figure 3B).
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
Fig. 2. Body weights. Data shown are for mice treated with DMBA/TPA.
Values represent mean SEM, N ¼ 8--38. P 5 0.0001 for differences by
diet at each time point 0 weeks (ANOVA--MIXED). Means at each time
point with different superscript letters are significantly different at the P 5
0.05 level (multiple comparisons by least squares means) and values with
more than one letter are not significantly different from means sharing either
of the letters. Body weights of mice treated with acetone were similar to the
body weights of the mice treated with DMBA/TPA in all parallel groups
except the acetone/DER/ADX/saline group that maintained a constant body
weight throughout the experiment (data not shown).
Effect of ADX and CCS on DER inhibited skin tumorigenesis
Table IV. Epidermal carcinoma and lymphosarcoma
N
AL/sham
AL/ADX/saline
AL/ADX/CCS
DER/sham
DER/ADX/saline
DER/ADX/CCS
P5
30
17
13
32
14
19
Incidence (%)
Multiplicity
Total
In situ
a
a,b
83
77a,b
59b,c
69a,b
50b,c
21c
0.0003
67
71a,b
54a,b
59a,b
50b,c
16c
0.0001
Invasive
a
73
12b
23b
22b
29b
5b
0.0001
Incidence
In situ
Total
3.13 2.77 1.85 1.56 2.29 0.21 0.0001
a
0.59
0.70a,b
0.58a,b
0.32b
0.66a,b
0.11c
1.97 2.53 1.31 1.34 1.86 0.16 0.0002
Invasive
a
1.17 0.24 0.54 0.22 0.43 0.05 0.0001
0.44
0.73a
0.48a
0.31a
0.62a
0.10b
Lymphosarcoma (%)
a
0.23
0.12b,c
0.22a,b
0.09b,c
0.19b,c
0.05c
10a,b
35a
8a,b
3b
7a,b
5b
0.0350
N ¼ Effective mice treated with DMBA/TPA that survived at least 23 weeks. Carcinoma multiplicity is the number of carcinomas per carcinoma bearing
animal that survived at least 23 weeks or longer.
Values represent mean SEM. (P 5 0.0003 x2 with multiple comparisons by Fisher’s exact test for incidence rates and P 5 0.0001 GENMOD with
multiple comparisons by least square means for multiplicity rates.) Column means with different superscript letters are significantly different at the
P 5 0.05 level and values with more than one letter are not significantly different from means sharing either of the letters.
Carcinoma incidence rates and multiplicity
Total carcinoma incidence rates include in situ and invasive
carcinomas. The response to diet and CCS was similar for the
total carcinoma incidence rates and the papilloma incidence
rates (Table IV). There were significant differences among
total carcinoma incidence rates (P50.0003) and also among
in situ (P50.0001) and invasive (P50.0001) carcinoma
incidence rates. Total carcinoma incidence rates were lowest
in the DER/ADX/CCS group and highest in the AL/sham
group. In situ carcinoma incidence rates were lowest in the
DER/ADX/CCS group and highest in the AL/ADX/saline
group. Invasive carcinoma incidence rate was significantly
increased in the AL/sham diet compared with the other five
groups (P50.0001) (Table IV). Carcinoma multiplicity was
defined as the number of carcinomas per carcinoma bearing
animal. Total carcinoma multiplicity was significantly lower
for the DER/sham than the AL/sham group (P50.05) and it
was lowest in the mice in the DER/ADX/CCS group
(P50.0001). In situ carcinoma multiplicity was lowest in the
DER/ADX/CCS group compared with the other five groups
(P50.0002). Invasive carcinoma multiplicity also varied significantly across treatment groups (P50.0001) and was lowest
in the DER/ADX/CCS group and highest in the AL/sham mice
(Table IV).
Other pathology
There was a significant increase in the appearance of lymphosarcoma in the spleens of the AL/ADX/saline mice compared
with the DER/sham and DER/ADX/CCS groups (P 5 0.0350)
(Table IV).
Gross observations
Skin ulceration in DMBA/TPA treated mice that survived at
least 23 weeks differed significantly between groups (P 5
0.00005). This lesion was observed in 27% (8/30) of the
mice in the AL/sham, 47% (8/17) of the mice in the AL/
ADX/saline and 6% (2/32) of the mice in the DER/sham
groups. Skin ulcerations were not observed in other groups
(0/56 mice). Casual observation of intense scratching directly
correlated with the observations on skin ulcerations.
Table V. Plasma corticosterone concentration (ng/ml) in mice surviving
to 31 weeks
AL/sham
AL/ADX/saline
AL/ADX/CCS
DER/sham
DER/ADX/saline
DER/ADX/CCS
P5
N
Sham
33
31
15
36
22
23
143 20a
151 19a
NS
ADX
107 20a
75 30a
101 25a
282 24b
0.0001
There was no significant difference between animals treated with DMBA/
TPA or acetone for any dietary--surgery treatment. So DMBA/TPA and
acetone treatment groups were combined to obtain treatment group mean
concentrations. Mice were killed when maximal plasma corticosterone levels
were expected. Thus sham-operated mice were killed in the morning while
ADX mice were killed in the evening. Therefore, these groups are presented
in separate columns and analyzed separately. Values represent mean SEM.
Individual means within the column were analyzed by ANOVA--GLM and
then compared with multiple comparisons by least squares means. Means
within a column with different superscript letters were significantly different
at the P 5 0.05 level. DER/ADX/CCS corticosterone concentration was
significantly higher than the other adrenalized treatment groups, P 5 0.0001
(ANOVA--GLM).
Plasma corticosterone
There was no statistical difference in plasma corticosterone
concentrations between mice treated with DMBA--TPA or
acetone for any dietary--surgery treatment. So these groups
were pooled and are presented as dietary--surgery treatment
group in Table V. Since the sham-operated and ADX mice
were killed in the morning and evening, respectively, these
groups are presented in separate columns in Table V and
analyzed separately. Mice in the DER/ADX/CCS group exhibited significantly higher plasma corticosterone concentrations
than any of the other ADX groups (P 5 0.0001).
Plasma insulin growth factor-1 (IGF-1)
In the ADX mice there was no significant difference in plasma
IGF-1 concentration between the animals receiving corticosterone or saline in the drinking water. Therefore these groups
were combined in the analysis of the dietary effects. Since the
sham-operated and ADX mice were killed in the morning and
evening, respectively, these groups are presented in separate
columns in Table VI and analyzed separately. Plasma IGF
concentrations in the DER diet mice were significantly
1081
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
DER/ADX/CCS significantly reduced papilloma multiplicity
compared with all other groups (P 5 0.0001).
J.W.Stewart et al.
Table VI. Plasma insulin growth factor (IGF-1) concentration (ng/ml) in
mice surviving to 31 weeks
N
AL/sham
AL/sham
AL/ADX1
AL/ADX
DER/sham
DER/sham
DER/ADX
DER/ADX
P5
4
2
7
2
9
2
8
2
Carcinogen/vehicle
DMBA/TPA
Acetone
DMBA/TPA
Acetone
DMBA/TPA
Acetone
DMBA/TPA
Acetone
Sham
ADX
a
169 31
255 44a
56 21b
69 44b
0.0042
201 31c
323 58c
42 29d
88 58d
0.0015
1
reduced to approximately one-third and one-forth of the values
in mice that received the AL diet in the sham-operated groups
(P 5 0.0042) and the ADX-operated groups (P50.0015).
Discussion
A striking inhibition of DMBA initiated and TPA promoted
papilloma development was observed in sham-operated DER
mice, particularly with respect to papilloma number. However,
this inhibition was partially blocked in ADX-operated DER
mice. In comparison, ADX did not appreciably alter papilloma
development in control mice. These observations demonstrate
the dependence of DER, in the prevention of mouse skin
papillomas, on an intact adrenal gland. Furthermore, daily
supplementation with corticosterone in the drinking water to
achieve peak corticosterone concentrations in the range of the
DER mice (15) resulted in cancer prevention in both AL/ADX
and DER/ADX mice. This suggests that the loss of corticosterone may have contributed to the higher tumor rate in the
ADX operated DER mice.
The impact of DER and ADX on mouse skin carcinoma
rates largely paralleled the effects observed on papillomas.
Total carcinoma multiplicity was higher in the DER/ADX
mice than in the DER/sham mice but this difference was
not statistically significant. However, corticosterone was a
potent inhibitor of both the total number of carcinomas and
the invasive carcinomas. In contrast to the general absence of
impact of ADX in the control-fed mice on papilloma and total
carcinoma rates, invasive carcinomas were inhibited by ADX
in control-fed mice. The reasons for the different patterns of
papillomas, total carcinomas and invasive carcinomas are not
clear, but these contrasts suggest that CCS in the absence of
the adrenal gland may play a very different role in the conversion to invasive carcinomas than it plays in papilloma
development.
It is interesting that administering corticosterone to the
ADX/DER mice resulted in papilloma and carcinoma rates
1082
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
The effect of corticosterone supplementation in the ADX mice was not
significant; therefore, the IGF data for ADX/saline or ADX/CCS drinking
water were combined in the analysis of differences between DMBA/TPA and
acetone. Mice were killed when maximal plasma corticosterone levels were
expected. Thus sham-operated mice were killed in the morning while ADX
mice were killed in the evening. These groups are presented in separate
columns and analyzed separately. Values represent mean SEM. Individual
means within the column were analyzed by ANOVA--GLM and then
compared with multiple comparisons by least squares means. Means within a
column with different superscript letters were significantly different at the
P 5 0.05 level. DER/sham was significantly lower than AL/sham and
DER/ADX was significantly lower than AL/ADX, P 5 0.005
(ANOVA--GLM).
that were lower than those observed in the sham-operated
DER mice. It is possible that the corticosterone administered
in the drinking water resulted in a circulating level that was
higher than was observed in the sham/DER mice. Currently, it
is not possible to perfectly mimic the CCS of intact mice in
ADX animals. Indeed, there are extra-adrenal sites near the
thoracic and abdominal sympathetic ganglia that may produce
glucocorticoid hormone (25) and these sites were not removed
in our ADX mice. The data presented in this paper in Table V
on plasma corticosterone in the mice surviving until the final
sacrifice show that the DER/ADX/CCS group had the highest
CCS of all treatment groups. Sham and ADX mice were killed
at different times to allow for comparison between AL and
DER mice at the times when the greatest differences in CCS
would be expected. Thus, we did not compare CCS concentration in sham and ADX mice in this experiment. While our
observations on CCS at the termination of the experiment are
intriguing, it is important to note that data from surviving
animals may be biased by uncontrolled differential survival
factors. Indeed, in mice that were fed these diets and treated
only with a single treatment of TPA or acetone 1 or 4 h before
killing, corticosterone was generally equally elevated in both
the sham-operated DER mice and in the corticosterone supplemented control-fed, and the corticosterone-supplemented
DER-fed groups as described below (17,26). An alternative
interpretation for the low rate of papilloma and carcinoma in
the DER/ADX/CCS group is that the adrenal gland might
provide both cancer-preventative (presumably elevated corticosterone in DER mice) and cancer-enhancing impacts.
However, we are not aware of any data in support of this
alternative hypothesis.
In previous short term studies we assessed CCS in the diet
and hormonal protocols used in the experiment reported in the
present paper. In both studies after 10--12 weeks of treatment,
the plasma corticosterone concentration in 17- to 22-week-old
mice was elevated (3- to 5-fold) in sham/DER mice compared
with sham/AL mice (17,26). ADX mice had lower corticosterone and did not differ by diet. An elevation in corticosterone in hormone supplemented groups to the level in the sham/
DER mice was observed in blood collected late in the day
(26). When the blood was collected in the morning, the supplemented groups had only a minor increase in corticosterone
(17). However, we would not anticipate observing an elevation in CCS in the corticosterone-supplemented groups in the
morning because mice consumed more drinking water later in
the day when they had food available. Administration of
corticosterone in the drinking water was done in the current
experiment instead of implanting corticosterone pellets
because drinking water administration mimicked the CCS
levels of DER mice in earlier studies (21) more closely.
ADX and corticosterone supplementation did not change the
glucocorticoid receptor activity in our earlier work (15).
Another hormone that merits being studied for contributing
to the DER prevention of skin carcinogenesis is IGF-1. IGF-1
is a cell survival factor that inhibits apoptosis and increases
cell proliferation (27). Diet restriction was shown to reduce
circulating IGF-1 in our laboratory (26) and by others (28). In
addition, IGF-binding proteins, which appear to control the
release of IGF-1, were also modulated by diet restriction in a
manner that would be expected to reduce IGF-1 activation
(29). The most compelling study to suggest that reduction in
circulating IGF-1 plays a key role in DER prevention of cancer
was conducted by Dunn et al. (28). Urinary bladder cancer was
Effect of ADX and CCS on DER inhibited skin tumorigenesis
weight gain with prolonged mineralocorticoid treatment (25).
Intracellular potassium is replaced with sodium ions and water
is retained with the sodium ions (25). Interestingly, the
DMBA/TPA initiated, AL/ADX/CCS mice generally did not
show elevated food intake in comparison with all of the acetone initiated groups (Table II).
An intriguing observation was the elevated lymphosarcoma
incidence in the AL/ADX mice in comparison with the DER/
sham and DER/ADX/CCS groups. This lesion appears to have
been reduced by both DER and CCS treatments, and elevated
by ADX in mice not supplemented with CCS. There is no
evidence that this lesion interfered with skin tumor development in the current experiment since the results did not correlate clearly with any of the skin lesions. The other lesion
observed was the high rate of skin ulceration in the AL/sham
and AL/ADX/saline groups and a low rate in the DER/sham
group. Skin ulceration was reduced either by CCS treatment or
by DER. These lesions may have reduced survival in the two
AL groups and thus may have blunted the papilloma and
carcinoma rates in these groups.
In separate studies we are assessing the mechanism for DER
and corticosterone prevention of mouse skin carcinogenesis.
These studies suggest that DER and corticosteroid hormone
prevention of mouse carcinogenesis occurs through inhibiting
signaling down the extracellular signal regulated kinase
(ERK)1,2, reducing activator protein 1 (AP1): DNA binding
and inhibiting transcriptional regulation through AP1 (26,31).
This inhibition appeared to be downstream from the inhibition
of specific isoforms of protein kinase C—PKC a and PKC z
(32,33). These isoforms were also reduced by corticosterone
supplementation in ADX mice in short-term studies (21). The
inhibition of ERK1,2 (17) and the reduction in AP1:DNA (26)
binding by DER were eliminated in ADX mice demonstrating
the dependence of these DER effects on an intact adrenal
gland. Supplementation with CCS in the ADX animals
restored the inhibition of TPA-induced ERK1,2 and
AP1:DNA binding (17,26). Studies of AP1-regulated genes
indicate that DER inhibits TPA-induced transcriptional activation of an AP1-luciferase reporter gene in a transgenic mouse
model (34). The known importance of TPA-induced ERK1,2
signaling and AP1 transcriptional regulation in mouse skin
carcinogenesis suggests that DER blockage of these events
causally contributes to DER prevention of skin carcinogenesis.
The data presented in this paper show that skin tumor development in mice in protocols with DER, ADX and CCS treatment can be explained, at least in part, by these molecular
effects of DER and CCS.
In summary, DER significantly inhibited tumorigenesis in
intact mice as was previously observed. This inhibition was
dependent upon an intact adrenal gland and corticosterone
supplementation to DER/ADX mice restored the tumorigenesis inhibition. Corticosterone supplementation itself inhibited
tumor progression when given to ADX mice fed the control
diet and had an additive inhibitory impact when given to ADX
mice fed the DER diet.
Acknowledgements
This research was supported by National Institute of Health/National Cancer
Institute grant CA77451, the Iowa Agriculture and Home Economics Experiment Station and the Center for Designing Foods to Improve Nutrition (Iowa
Nutrition Special Grant from the USDA).
1083
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
induced by p-cressidine and after pre-neoplasia was confirmed, mice were fed 20% DER with or without recombinant
IGF-1 administered with mini-pumps. Mice fed ad libitum
were included in a parallel group. The circulating IGF-1 was
reduced by 24% after 5 weeks of feeding DER diet and the
levels were restored to the AL values in the DER plus IGF-1
group. The progression of tumors was inhibited in the DER
group but it was not reduced in the DER plus IGF-1 group (28).
Studies by Zhu et al. (20) used dietary corticosterone supplements to assess the role of CCS in DER prevention of mammary cancer. In the Sprague--Dawley rat model used for these
studies they observed elevated CCS and decreased IGF-1 in
response to the dietary corticosterone supplement.
IGF-I was measured in the mice in our tumor study is
presented in this paper and in previous 12 week, short term
studies of SENCAR mice (26). In the short term studies DER
decreased circulating IGF-I (231 14 ng/ml, mean SEM)
(P 5 0.05) in comparison with AL mice (402 18 ng/ml), but
TPA treatment, ADX and corticosterone supplementation did
not alter these values (26). In the present tumor study circulating IGF-I was reduced in the DER mice irrespective of the
other experimental treatments. These were important observations since considerable research has suggested that reduced
circulating IGF-I or elevated IGF--BPs in the DER mice contributes to cancer prevention by DER (30). Thus, since the
hormonal treatments that we used to modulate CCS did not
mediate circulating IGF-1, the effects that we attribute to
CCS were probably not owing to IGF-1 in the present
investigation.
It is unlikely that aldosterone, epinephrine or norepinephrine
were responsible for the adrenal dependence of DER inhibition
of mouse skin carcinogenesis. Earlier studies in our laboratory
indicated that there was no difference in aldosterone levels
between the AL-fed mice or the mice on DER diet (15).
Furthermore, we did not find reports on the role of epinephrine
or norepinephrine in skin carcinogenesis.
Our tumor study determined that DER inhibition of skin
tumor promotion was in large measure mediated through
increasing CCS. However, it is noteworthy that CCS did not
explain all the difference between the AL and DER mice since
supplementation with corticosterone was more effective in
preventing papillomas and carcinomas in DER/ADX mice
than in AL/ADX mice. These results support the hypothesis
that, while CCS contributes to cancer prevention by DER,
other factors are also involved. It is important to note that
changes in circulating IGF-I did not appear to contribute to
the DER effects that were attributed to CCS in our study as
noted above. However, the elevated glucocorticoid hormone
may interact with other hormonal effects, such as the lowered
IFG-1 to prevent skin tumorigenesis.
The changes in the body weight over time showed that the
DMBA/TPA initiated, AL/ADX/CCS treated mice actually
gained considerably more weight (more or less double) than
the AL/sham operated mice (Table III). This observation was
paralleled by data showing a lower food intake in these mice in
comparison with the AL/sham and AL/ADX/saline groups
from weeks 10--30 (Table II). These results suggest that the
dose of corticosterone in these mice may have caused salt and
water retention, known impacts of excessive mineralocorticoid
in humans (25). This may have been owing to changes in
potassium balance in these mice since there is evidence in
humans that excessive glucocorticoid therapy can cause potassium depletion and associated water retention, and body
J.W.Stewart et al.
References
1084
Received October 4, 2004; revised February 1, 2005;
accepted February 8, 2005
Downloaded from http://carcin.oxfordjournals.org/ at Pennsylvania State University on March 6, 2014
1. International Agency for Research On Cancer. (2002) International
Agency for Research on Cancer Handbooks of Cancer Prevention. IARC
Press, Lyon, France.
2. Lahti-Koski,M., Pietinen,P., Heli€ovaara,M. and Vartiainen,E. (2002)
Associations of body mass index and obesity with physical activity, food
choices, alcohol intake, and smoking in the 1982--1997 FINRISK Studies.
Am. J. Clin. Nutr., 75, 809--817.
3. Tannenbaum,A. (1942) The genesis and growth of tumors. II Effects of
caloric restriction per se. Cancer Res., 2, 460--467.
4. Tannenbaum,A. (1945) The dependence of tumor formation on the
composition of the calorie-restricted diet as well as on the degree on the
restriction. Cancer Res., 5, 616--625.
5. Birt,D.F., Barnett,T., Pour,P.M. and Copenhaver,J. (1996) High-fat diet
blocks the inhibition of skin carcinogenesis and reductions in protein
kinase C by moderate energy restriction. Mol. Carcinog., 16, 115--120.
6. Pollard,M., Luckert,P.H. and Pan,G.-Y. (1984) Inhibition of intestinal
tumorigenesis in methylazoxymethanol-treated rats by dietary restriction.
Cancer Treat. Rep., 68, 405--408.
7. Birt,D.F., Pelling,J.C., White,L.T., Dimitroff,K. and Barnett,T. (1991)
Influence of diet and calorie restriction on the initiation and promotion of
skin carcinogenesis in the SENCAR mouse model. Cancer Res., 51,
1851--1854.
8. Kritchevsky,D. (1999) Caloric restriction and experimental carcinogenesis. Toxicol. Sci., 52, 13--16.
9. Birt,D.F. (1990) The influence of dietary fat on carcinogenesis: lessons
from experimental models. Nutr. Rev., 48, 1--5.
10. Birt,D.F., White,L.T., Choi,B. and Pelling,J.C. (1989) Dietary fat effects
on the initiation and promotion of two-stage skin tumorigenesis in the
SENCAR mouse. Cancer Res., 49, 4170--4174.
11. Pashko,L.L. and Schwartz,A.G. (1992) Reversal of food restrictioninduced inhibition of mouse skin tumor promotion by adrenalectomy.
Carcinogenesis, 13, 1925--1928.
12. Pashko,L.L. and Schwartz,A.G. (1996) Inhibition of 7,12dimethylbenz[a]anthracene-induced lung tumorigenesis in A/J mice by
food restriction is reversed by adrenalectomy. Carcinogenesis, 17,
209--212.
13. Goodman,H.M. (1994) Basic Medical Endocrinology. Raven Press,
New York.
14. Pugh,T.D., Oberley,T.D. and Weindruch,R. (1999) Dietary intervention at
middle age: caloric restriction but not dehydroepiandrosterone sulfate
increases lifespan and lifetime cancer incidence in mice. Cancer Res., 59,
1642--1648.
15. Yaktine,A., Vaughn,R., Blackwood,D., Duysen,E. and Birt,D.F. (1998)
Dietary energy restriction in the SENCAR mouse: elevation of glucocorticoid hormone levels but no change in distribution of glucocorticoid
receptor in epidermal cells. Mol. Carcinog., 21, 62--69.
16. Slaga,T.J., Lichti,U., Hennings,H., Elgjo,K. and Yuspa,S.H. (1978)
Effects of tumor promoters and steroidal anti-inflammatory agents on
skin of newborn mice in vivo and in vitro. J. Natl. Cancer Inst., 60,
425--431.
17. Liu,Y., Wang,W., Hawley,J. and Birt,D.F. (2002) Adrenalectomy abrogates reduction of TPA-induced ERK activity in the epidermis of dietary
energy restricted SENCAR mice: implications of glucocorticoid hormone.
Cancer Epidem. Biomarkers Prev., 11, 299--304.
18. Jonat,C., Rahmsdorf,H.J., Park,K.-K., Cato,A.C.B., Gebel,S., Ponta,H.
and Herrlich,P. (1990) Antitumor promotion and antiinflammation:
down-modulation of AP-1 (Fos/Jun) activity by glucocorticoid hormone.
Cell, 62, 1189--1204.
19. Beato,M. (1989) Gene regulation by steroid hormones. Cell, 56, 335--344.
20. Zhu,Z.J., Jiang,W.Q. and Thompson,H.J. (2003) Mechanisms by which
energy restriction inhibits rat mammary carcinogenesis: in vivo effects of
corticosterone on cell cycle machinery in mammary carcinomas.
Carcinogenesis, 24, 1225--1231.
21. Birt,D.F., Duysen,E., Wang,W.Q. and Yaktine,A. (2001) Corticosterone
supplementation reduced selective protein kinase C isoform expression in
the epidermis of adrenalectomized mice. Cancer Epidem. Biomarkers
Prev., 10, 679--685.
22. Hofert,J.F., Goldstein,S. and Phillips,L.S. (1989) Glucocorticoid effects on
IGF-I/somatomedin-C and somatomedin inhibitor in streptozotocin
diabetic rats. Metab. Clin. Exp., 38, 594--600.
23. Gallo,P.V. and Weinberg,J. (1981) Corticosterone rhythmicity in the rat:
interactive effects of dietary restriction and schedule of feeding. J. Nutr.,
111, 208--218.
24. Collett,D. (2003) Modeling Survival Data in Medical Research. Chapman
and Hall, CRC, Boca Raton, Florida.
25. Ganong,W.F. (2004) The adrenal medulla and adrenal cortex. In Review
of Medical Physiology. Appleton and Lange, Norwalk, Connecticut,
pp. 323--347.
26. Przybyszewski,J., Yaktine,A.L., Duysen,E., Blackwood,D., Wang,W.Q.,
Au,A. and Birt,D.F. (2001) Inhibition of epidermal c-Jun protein, mRNA
and AP-1:DNA binding by 40% energy restriction is reversed by
adrenalectomy in SENCAR mouse. Carcinogenesis, 22, 1421--1427.
27. Yu,H. and Rohan,T. (2000) Role of the insulin-like growth factor family
in cancer development and progression. J. Natl Cancer Inst., 92,
1472--1489.
28. Dunn,S.E., Kari,F.W., French,J., Leininger,J.R., Travlos,G., Wilson,R. and
Barrett,J.C. (1997) Dietary restriction reduces insulin-like growth factor I
levels, which modulates apoptosis, cell proliferation and tumor progression in p53-deficient mice. Cancer Res., 57, 4667--4672.
29. Donovan,S.M., Atilano,L.C., Hintz,R.L., Wilson,D.M. and Rosenfeld,R.G.
(1991) Differential regulation of the insulin-like-growth factors (IGF-I
and-II) and IGF-1 binding proteins during malnutrition in the neonatal rat.
Endocrinology, 129, 146--156.
30. McCarty,M.F. (1997) Up-regulation of IGF binding protein-1 as an
anticarcinogenic strategy: relevance to caloric restriction, exercise and
insulin sensitivity. Med. Hypotheses, 48, 297--308.
31. Liu,Y., Duysen,E., Yaktine,A., Au,A., Wang,W. and Birt,D.F. (2001)
Dietary energy restriction inhibits ERK but not JNK or p38 activity in the
epidermis of SENCAR mice. Carcinogenesis, 22, 607--612.
32. Birt,D.F., Copenhaver,J., Pelling,J.C. and Anderson,J. (1994) Dietary
energy restriction and fat modulation of protein kinase C isoenzymes and
phorbol ester binding in Sencar mouse epidermis. Carcinogenesis, 15,
2727--2732.
33. Kris,E.S., Choe,M., Luthra,R., Conway,H., Barnett,T., Yaktine,A. and
Birt,D.F. (1994) Protein kinase C activity is reduced in epidermal cells
from energy-restricted SENCAR mice. J. Nutr., 124, 485--492.
34. Birt,D.F., Przybyszewski,J., Wang,W., Stewart,J.W. and Liu,Y. (2004)
Identification of molecular targets for dietary energy restriction prevention
of skin carcinogenesis: an idea cultivated by Edward Bresnick. J. Cell
Biochem., 91, 258--264.