Carcinogenesis vol.20 no.3 pp.425–430, 1999
Prevention by aspirin and its combination with
α-difluoromethylornithine of azoxymethane-induced tumors,
aberrant crypt foci and prostaglandin E2 levels in rat colon
Hong Li, Herman A.J.Schut, Philip Conran,
Paula M.Kramer, Ronald A.Lubet1, Vernon E.Steele1,
Ernest E.Hawk1, Gary J.Kelloff1 and Michael A.Pereira2
Medical College of Ohio, Department of Pathology, 3000 Arlington Avenue,
Toledo, OH 43614-5806 and 1Chemoprevention Branch, Division of Cancer
Prevention and Control, National Cancer Institute, Rockville, MD 20892,
USA
2To
whom correspondence should be addressed
Email: [email protected]
The dose–response relationship in male F344 rats was
determined for the ability of aspirin administered in the
diet to prevent azoxymethane (AOM)-induced colon cancer
and aberrant crypt foci (ACF) and to reduce prostaglandin
E2 (PGE2) levels. Starting at either 7 or 22 weeks of age,
the rats received aspirin. All rats received two doses of
AOM (15 mg/kg each on days 7 and 14) and were killed
on day 36. The lowest concentrations of aspirin to prevent
ACF or reduce PGE2 levels were 600 and 400 mg/kg,
respectively. To evaluate the prevention of tumors, rats
received either 0 or 400 mg/kg aspirin for a total of 39
weeks with AOM (30 mg/kg) administered 7 days after the
start of treatment. Aspirin had no effect on the yield of
colon tumors. In a second experiment, rats started to
receive 0, 200, 600 or 1800 mg/kg aspirin or 1000 mg/kg
α-difluoromethylornithine (DFMO) F/– aspirin. Eight and
15 days later, all the rats received 15 mg/kg AOM. Eleven
weeks later, animals that were receiving the control diet
started to receive 0, 200, 600 or 1800 mg/kg aspirin; 1000
or 3000 mg/kg DFMO; or 1000 mg/kg DFMO F 200 or
600 mg/kg aspirin. The animals were killed 32 weeks later.
DFMO effectively reduced the yield of colon tumors when
administered starting either before or after AOM while
aspirin was much weaker. The combination of aspirin F
DFMO administered after AOM was synergistic. Both
aspirin and DFMO decreased the Mitotic Index, while
apoptosis was increased only by DFMO. Our results demonstrated that aspirin and DFMO could prevent colon cancer
when administered after AOM. Furthermore, aspirin
reduced ACF, PGE2 levels and mitosis at concentrations
that did not prevent cancer. In contrast, the ability to
enhance apoptosis did correlate with the prevention of
cancer.
Introduction
Colorectal cancer is the fourth most common type of cancer
in the world (1). One strategy to decrease its incidence is the
development of chemopreventive drugs. Aspirin and other
non-steroidal anti-inflammatory drugs (NSAIDs), including
Abbreviations: ACF, aberrant crypt foci; AOM, azoxymethane; COX, cyclooxygenase; DFMO, α-difluoromethylornithine; EHRT, Environmental Health
Research and Testing, Inc.; NSAID, non-steroidal anti-inflammatory drug;
ODC, ornithine decarboxylase; PGE2, prostaglandin E2.
© Oxford University Press
piroxicam and sulindac appear to be promising chemopreventive drugs for colon cancer (2). Although several epidemiological studies have suggested an inverse association between
the chronic use of aspirin and the risk of colorectal cancer (3–
6), one relatively large prospective study failed to observe an
association (7). Aspirin has been shown to inhibit azoxymethane (AOM)-induced colon cancer in rats when provided in
the diet before administering AOM and continuing until the
end of the study (8). Wargovich et al. have reported that
aspirin prevented aberrant crypt foci (ACF) when administered
starting 4 weeks after AOM (9). However, similar concentrations of aspirin (200 and 400 mg/kg diet) administered starting
prior to AOM did not prevent ACF (10). In the 1,2-dimethylhydrazine model of colon cancer, aspirin was effective in preventing colon tumors when administered during the initiation
phase; however, it was not effective when administered 4
weeks after the carcinogen (11). Aspirin was also ineffective
in preventing cholic acid promotion of AOM-initiated colon
cancer (12). Thus, except for the prevention of AOM-induced
ACF, aspirin has not been effective when administered only
during the promotion/progression phase of carcinogenesis. In
contrast two other NSAIDs, piroxicam and sulindac, have been
very effective in preventing colon cancer when administered
up to 14 weeks after AOM (13–17).
NSAIDs inhibit cyclooxygenase (COX) and thus reduce the
biosynthesis of prostaglandins. Aspirin preferentially acetylates
COX-1 resulting in irreversible inhibition (2,18,19). Human
colon tumors contain high levels of prostaglandins, in particular
those of the E series (20–23). High levels of prostaglandin E2
(PGE2) have also been reported in AOM-induced colon tumors
(15). Therefore it is reasonable that inhibition of prostaglandin
synthesis by NSAIDs might be associated with their prevention
of colon cancer.
Ornithine decarboxylase (ODC) inhibitors are another class
of chemicals that have shown to have efficacy in preventing
cancer. ODC is the first enzyme in polyamine biosynthesis
and is closely correlated with cell proliferation, differentiation
and carcinogenesis (24–26). α-Difluoromethylornithine
(DFMO), an irreversible inhibitor of ODC, has been shown to
inhibit the development of cancer in several animal models (27)
including prevention of AOM-induced colon cancer (16,17,28–
32). The inhibition by DFMO was dose-dependent and its
combination with piroxicam was reported to be synergistic
(16). DFMO was also effective in preventing AOM-induced
ACF in rat colon (33).
While aspirin and DFMO are promising chemopreventive
agents for colon cancer, there is limited understanding of the
dose–response relationship of aspirin and of their efficacy when
administered in combination or only during the promotional/
progression phase of carcinogenesis. Therefore, we evaluated
the dose–response relationship for the prevention by aspirin
of AOM-induced ACF and colon tumors and for its reduction
of PGE2 level in the colon. The ability to prevent colon tumors
was also determined for aspirin and DFMO administered
425
H.Li et al.
starting 11 weeks after AOM and for combinations containing
them. Furthermore, the ability of aspirin, DFMO and the
combination of them to alter the level of apoptosis and cell
proliferation in colon adenomas is reported.
Materials and methods
Animals, chemicals and diet
Male F344 rats were obtained from Charles River Laboratories (Raleigh, NC).
They were housed at the Medical College of Ohio (Toledo, OH), except for
the first study for the prevention of colon tumor (experiment 3) that was
housed at Environmental Health Research and Testing, Inc. (EHRT; Lexington,
KY). Both Laboratory Animal Facilities are AAALAC accredited. The care
and maintenance of the animals at both facilities were in accordance to the
US Public Health Service ‘Guide for the Care and Use of Laboratory Animals’.
Solid-bottom polycarbonate cages with stainless steel wire-bar lids and Bedo-Cob bedding (Andersons, Toledo, OH) were used to house 2 rats/cage. The
light cycle consisted of 12 h each of light and dark. The animal rooms were
maintained at 64–76°F and 55 6 15% relative humidity. Drinking water and
AIN 76A diet were supplied to the animals ad libitum.
AOM was purchased from Sigma (St Louis, MO) and aspirin was from
Acros Organics (Pittsburgh, PA), except in the study performed at EHRT
where it was purchased from ICN Biochemicals (Irvine, CA). DFMO was
obtained from the DCPC Repository of the NCI (c/o McKesson BioServices,
Rockville, MD). AIN-76 diet was purchased from Dyets (Bethlehem, PA) and
used within 3 months of formulation.
Assays for dose selection, prevention of ACF and reduction of PGE2 levels
Toxicity, prevention of AOM-induced ACF and reduction of PGE2 levels were
evaluated in two experiments in which aspirin was administered in the diet
of rats starting either at 7 (experiment 1) or 22 (experiment 2) weeks of age.
These two ages corresponded to when the rats started to receive aspirin in
the second study for prevention of colon cancer. In experiment 1, rats at 7
weeks of age were assigned to one of six treatment groups consisting of 10
animals each and received either 0, 200, 400, 600, 1000 or 2000 mg aspirin/
kg diet. In experiment 2, the rats were 22 weeks of age when assigned to
eight treatment groups of 11 animals each and started to receive either 0, 200,
400, 600, 1000, 2000, 3000 or 4000 mg/kg aspirin in their diet. In both
experiments at 7 and 14 days after the rats started to receive the aspirin, they
were administered AOM (15 mg/kg) by i.p. injection. The animals received
aspirin for a total of 5 weeks (3 weeks after the second dose of AOM) and
were then killed by carbon dioxide asphyxiation. Body weight was monitored
at assignment to the treatment groups so that there was no difference among
the groups and weekly during exposure to aspirin.
At necropsy, the colons were removed, cut along the longitudinal axis and
flushed with cold saline. For determination of PGE2, a sample of the mucosa
was obtained by scraping the 0–4 cm segment of colon from the anus. Using
liquid nitrogen, it was rapidly frozen in vials coated with 10 µg/ml indomethacin
and stored at –70°C. The rest of the colon was then stained and evaluated for
ACF as described by Bird (34) and previously used by us (10,33). Briefly,
they were stained for 10 min in a solution of 0.2% methylene blue (Sigma)
dissolved in 70% alcohol. After staining, the colons (mucosal side up) were
placed on a microscope slide and with the aid of a light microscope
(magnifications of 40 and 1003) evaluated for ACF. The criteria used to
identify ACF included the following: (i) increased size of the crypts; (ii)
increased thickness of the epithelial cell lining; (iii) elongation of the luminal
opening; and (iv) increased pericryptal zone separating the crypts of the ACF
from the surrounding normal appearing crypts. The location of the ACF in
the colon and the number of crypts/focus were recorded.
Determination of the concentration of PGE2
A sample of the mucosal scraping was sonicated for 10 s in 1.0 ml 0.1 M
Tris–HCl buffer, pH 7.4, containing 0.02 M EDTA and 10 µg/ml indomethacin.
A 0.2 ml aliquot was removed for determination of protein content using the
Bio-Rad DC Protein Assay kit (Richmond, CA). To the remainder of the
sample, 0.8 ml of methanol was added and vigorously mixed. After acidifying
the sample with six drops of 1% formic acid, it was extracted with 1.5 ml
chloroform. Following centrifugation at 2000 r.p.m. for 10 min, the organic
phase was removed and dried using a Speed-Vac. The dried extract was
resuspended in assay buffer and assayed for PGE2 content using Dupont125
I-PGE2 RIA kit (Boston, MA). PGE2 concentration was calculated as pg
PGE2/mg protein.
Assays for prevention of colon tumors
In experiment 3, six-week-old male F344 rats were divided into the following
treatment groups with no significant difference in body weight among them.
Groups 1 and 2 contained 30 and 10 animals, respectively, and received
426
400 mg aspirin/kg diet. Groups 3 and 4 contained 30 and 10 animals,
respectively, and continued to receive the control AIN 76A diet. Seven days
later the animals in treatment groups 1 and 3 received an s.c. injection of
AOM (30 mg/kg body wt) and those in groups 2 and 4 received 4.0 ml/kg of
the saline vehicle. The animals continued to receive the assigned diets until
killed by carbon dioxide asphyxiation 38 weeks after receiving the AOM.
Body weight was monitored weekly for the first 4 weeks and every 2–3
months thereafter.
Experiment 4: in the above two dose-selection studies, aspirin concentrations
of 2000 (experiment 1) and 4000 (experiment 2) mg/kg diet did not appear
to affect the health and body weight of the animals. Furthermore, the reduction
by aspirin of the yield of ACF and of the level of PGE2 reached maximums
by 600–2000 mg/kg diet as discussed later in Results. Therefore, in this
experiment 1800 mg/kg diet was chosen as the high concentration of aspirin.
Male F344 rats (7 weeks old) were divided into 14 treatment groups containing
34 animals each so that there was no significant difference in the body weights
among the groups. Then the animals in treatment groups 1–6 started to receive
in their AIN 76A diet either 0 (control diet), 200, 600 or 1800 mg/kg aspirin
or 1000 mg/kg DFMO 1 0, 200 or 600 mg/kg aspirin. Eight and 15 days
later, all the rats in the 14 treatment groups were administered AOM (15 mg/
kg body wt) by i.p. injection. Treatment groups 7–14 continued to receive the
AIN-76A diet until 11 weeks after the second dose of AOM. At that time,
the animals in groups 7–13 received in their diet either 200, 600 or 1800 mg/
kg aspirin; 1000 or 3000 mg/kg DFMO; or 1000 mg/kg DFMO 1 200 or
600 mg/kg aspirin. Treatment group 14 continued to receive the AIN-76A
diet. The animals were fed their respective diets until they were killed 43
weeks after the second dose of AOM. Body weight was monitored weekly
for the first 6 weeks and monthly thereafter throughout the experiment.
At necropsy in experiments 3 and 4, the colons were removed from cecum
to anus, slit open longitudinally, flushed with saline and examined for tumors.
The location and size of the tumors were recorded prior to harvesting. The
small intestine was palpated for the presence of tumors that were also
harvested. The tumors were fixed in 10% buffered formalin overnight and
then stored in 70% alcohol until embedded in paraffin. Sections (5 µm) were
stained with hematoxylin and eosin for histopathologic evaluation.
Determination of apoptosis and cell proliferation in adenomas
Adenomas from treatment groups in experiment 4 that received 0, 600 or
1800 mg/kg aspirin; 1000 mg/kg DFMO; or 1000 mg/kg DFMO 1 600 mg/
kg aspirin were evaluated for apoptosis and mitosis. Only adenomas were
evaluated because of the limited number of carcinomas. Apoptotic and mitotic
cells were counted using H&E stained sections. Apoptotic cells were identified
by cell shrinkage with a halo separating them from surrounding cells,
nuclear condensation, formation of apoptotic bodies (nuclear fragments), and
eosinophilic and condensed cytoplasm (35). At least 1000 epithelial cells were
evaluated and the number of apoptotic and mitotic cells recorded. The
Apoptotic and Mitotic Indexes were determined by dividing the number of
affected cells by the total number of cells evaluated multiplied by 100.
Results
Experiments 1 and 2: dose selection, prevention of ACF and
reduction of PGE2
In experiments 1 and 2, aspirin administered for 5 weeks at
concentrations as high as 2000 and 4000 mg/kg diet did not
affect the body weight of the animals or their food consumption.
There was also no indication of toxicity observed during the
daily evaluation of the animals. Therefore, the maximum
concentrations of aspirin evaluated in experiments 1 and 2,
i.e. 2000 and 4000 mg/kg diet, would appear not to be toxic.
The effect of aspirin administered to rats starting at 7 or 22
weeks of age on AOM-induced ACF and on PGE2 level in
the colon is presented in Figure 1. Aspirin caused a dosedependent reduction in the yield of ACF and in the level of
PGE2. The lowest concentration of aspirin in both experiments
to significantly prevent ACF was 600 mg/kg and to reduce
PGE2 was 400 mg/kg. In the two experiments, maximum
reduction of ACF was 40–45% (600–2000 mg/kg), while for
PGE2 levels was 65–85% (1000–2000 mg/kg). Hence, although
both prevention of ACF and reduction of PGE2 reached
maximums at similar concentrations of aspirin, the reduction
of PGE2 was greater than the prevention of ACF. The yield
Prevention by aspirin of colon cancer
Fig. 2. Effect of aspirin and DFMO administered starting before AOM on
the multiplicity of neoplastic lesions in the colon. Results are means 6 SE.
The data were statistically analyzed by a one way ANOVA followed by the
Tukey test. The * indicates P , 0.05 when compared with the AOM 1
control diet group; a, P , 0.05 when the combinations were compared with
the treatment group that received only DFMO.
Fig. 1. Inhibition by aspirin of ACF and PGE2 in rats of 7 (A) and 22
(B) weeks of age at the start of the experiments. Results are the percentage
inhibition when compared with the AOM 1 control diet group and are
means 6 SE for treatment groups consisting of 10 (A) and 11 (B) animals
each. The yield of ACF and the level of PGE2 were 98.0 6 10.9 ACF/
animal and 93.2 6 8.7 pg PGE2/mg protein and 159.0 6 10.0 ACF/animal
and 285.2 6 54.8 pg PGE2/mg protein in the AOM 1 control diet group
for 7- and 22-week-old rats, respectively. *Significant difference from the
AOM 1 control diet group by one way ANOVA followed by the Tukey
test, P , 0.05.
of ACF of all sizes were reduced, i.e. foci with 1, 2 or 31
crypts/focus.
Experiments 3 and 4: prevention of colon tumors
In experiment 3, the prevention of colon cancer was examined
for 400 mg/kg of aspirin administered in the diet from 1 week
before AOM until killed 38 weeks later. Aspirin did not affect
the body weight of the animals during the experiment including
the terminal killing presented in Table I. Table I also contains
the yield of colon adenomas, adenocarcinomas and tumors
(adenomas 1 adenocarcinomas). Aspirin did not affect the
yield or incidence of animals with these neoplastic lesions.
No tumors were found in animals that did not receive AOM.
In experiment 4, aspirin, DFMO and combinations containing them were administered starting either before or 11
weeks after AOM (Table II). Aspirin, DFMO or their combinations did not affect the body weight of the animals during the
study. Table II contains the body weight of the animals at
killing and the incidence of animals with colon adenomas,
adenocarcinomas or tumors (adenomas 1 adenocarcinomas).
The multiplicity of adenomas, adenocarcinomas and tumors/
animal are presented in Figures 2 and 3. When administered
starting before AOM, concentrations of aspirin up to 1800
mg/kg diet did not significantly alter the incidence (Table II)
or multiplicity (Figure 2) of adenomas, adenocarcinomas or
tumors. When administered starting 11 weeks after AOM, the
high concentration of aspirin (1800 mg/kg diet) but not its
two lower concentrations, significantly reduced the multiplicity
of adenomas and tumors (Figure 3). However, aspirin did not
effect the incidence of animals with adenomas, adenocarcinomas or tumors (Table II).
DFMO administered starting before AOM reduced the
multiplicity of adenomas and tumors (Figure 2) and the
incidence of animals with adenomas although the reduction in
the incidence was not statistically significant (P 5 0.069)
(Table II). When DFMO was administered in combination
with either 200 or 600 mg/kg aspirin, the incidence and
multiplicity of adenomas, adenocarcinomas and tumors were
reduced further with a dose-dependent trend. When administered after AOM, DFMO (1000 and 3000 mg/kg diet) reduced
both the incidence and multiplicity of adenomas and tumors
(Table II, Figure 3). The combination of either 200 or 600
mg/kg aspirin with DFMO (1000 mg/kg diet) administered
after AOM did not reduce further the incidence or multiplicity
of adenomas, adenocarcinomas or tumors.
Effect of aspirin and DFMO on apoptosis and mitosis in
colon tumors
The effect of 600 and 1800 mg/kg aspirin, 1000 mg/kg DFMO
and the combination of 1000 mg/kg DFMO 1 600 mg/kg
aspirin on the Apoptotic and Mitotic Indexes in adenomas is
presented in Table III. Aspirin administered at either 600 or
1800 mg/kg, DFMO and the combination of DFMO 1 aspirin
reduced the Mitotic Index in adenomas by 49–63%. DFMO
with and without 600 mg/kg aspirin increased the Apoptotic
Index 2.79- and 3.70-fold, respectively. However, aspirin (600
and 1800 mg/kg diet) administered by itself did not affect the
Apoptotic Index.
Discussion
Epidemiological and experimental studies have suggested a
low to moderate efficacy of aspirin to prevent colon cancer
427
H.Li et al.
Table I. Effect of aspirin on AOM-induced colon tumors
Treatment groups
AOM
na
Tumor yieldb
Adenomas
1.
2.
3.
4.
400 mg/kg aspirin
400 mg/kg aspirin
Control diet
Control diet
1
–
1
–
27
10
27
10
0.37
0.0
0.41
0.0
6
6
6
6
0.11 (33.3)
0.0 (0)
0.11 (37.0)
0.0 (0)
Carcinoma
Total
6
6
6
6
0.37
0.0
0.59
0.0
0.0
0.0
0.19
0.0
0.0 (0)
0.0 (0)
0.11 (18.5)
0.0 (0)
6
6
6
6
0.11 (33.3)
0.0 (0)
0.11 (55.6)
0.0 (0)
an, number of
bMeans 6 SE
animals at killing. At the start of the study there were 28 animals in treatment groups 1 and 3, and 10 animals in the other two groups.
for the number of lesions per animal with the number in parentheses being the percentage of rats with lesion. There were no statistically
significant differences between the group treated with aspirin and the corresponding control diet group.
Table II. Effect of aspirin, DFMO and their combination on AOM-induced colon tumors
Treatment groups
na
Body weightb (g)
Incidencec
Adenoma
Carcinoma
Total
Treatment started before AOM
1. 200 mg/kg aspirin
2. 600 mg/kg aspirin
3. 1800 mg/kg aspirin
4. 1000 mg/kg DFMO
5. 1000 mg/kg DFMO 1 200 mg/kg aspirin
6. 1000 mg/kg DFMO 1 600 mg/kg aspirin
34
33
29
34
30
30
457.3
439.5
461.2
447.8
465.1
458.4
6
6
6
6
6
6
7.7
8.3
6.1
6.2
5.8
7.5
61.8
54.5
48.3
38.2
26.7 d
6.7 e
23.5
30.3
13.8
29.4
6.7
10.0
70.6
63.6
51.7
61.8
30.0 e
13.3 e
Treatment started 11 weeks after AOM
7. 200 mg/kg aspirin
8. 600 mg/kg aspirin
9. 1800 mg/kg aspirin
10. 1000 mg/kg DFMO
11. 3000 mg/kg DFMO
12. 1000 mg/kg DFMO 1 200 mg/kg aspirin
13. 1000 mg/kg DFMO 1 600 mg/kg aspirin
14. Control diet
28
27
30
28
31
32
32
31
454.1
449.5
448.5
436.1
441.6
487.9
463.6
451.4
6
6
6
6
6
6
6
6
6.0
5.7
9.7
8.1
8.3
7.8
7.6
7.4
67.9
70.4
50.0
32.1d
25.8d
53.1
28.1d
64.5
25.0
18.5
10.0
10.7
16.1
9.4
12.5
16.1
78.6
74.1
53.3
39.3d
29.0d
59.4
31.2d
71.0
an, number of
bBody weight
animals at killing. Each treatment group started with 34 animals.
at killing (mean 6 SE).
cPercentage of animals with the neoplastic lesion.
dSignificantly different from the control diet group by the Kruskal-Wallis test, P , 0.05.
eSignificantly different from the control diet group and from the 1000 mg/kg DFMO group by the Kruskal-Wallis test, P , 0.05.
Table III. Effect of aspirin, DFMO and their combination on the Mitotic
and Apoptotic Indexes in adenomas
Treatment
na
Mitotic Indexb Apoptotic Indexb
1. Control diet
2. 600 mg/kg aspirin
3. 1800 mg/kg aspirin
4. 1000 mg/kg DFMO
5. 1000 mg/kg DFMO 1
600 mg/kg aspirin
7
6
5
5
4
2.94 6 0.48
1.45 6 0.25c
1.11 6 0.12c
1.50 6 0.14c
1.42 6 0.22c
an,
1.09
1.78
1.36
3.04
4.03
6
6
6
6
6
0.20
0.15
0.25
0.76c
0.65c
number of tumors analyzed; each tumor was from a different animal.
bMean 6 SE.
cSignificant difference
from the control diet group by one way ANOVA
followed by the Tukey test, P , 0.05.
Fig. 3. Effect of aspirin and DFMO administered after AOM on the
multiplicity of neoplastic lesions in the colon. Results are means 6 SE. The
data were statistically analyzed by a one way ANOVA followed by the
Tukey test. *P , 0.05 when compared with the AOM 1 control diet group.
None of the combinations was statistically different from the treatment
group that received only DFMO.
(2–9,11,12,36). Reddy et al. reported that 200 and 400 mg/kg
aspirin in the diet administered starting before AOM and
continuing for 52 weeks significantly reduced the incidence
428
and multiplicity of colon tumors in male F344 rats (8).
However, we were unsuccessful in confirming prevention by
400 mg/kg aspirin. Others have found that aspirin at greater
concentrations in the diet including 4000 mg/kg were not toxic
to rats (37–39). Therefore, we first confirmed that aspirin
concentrations of 2000 and 4000 mg/kg diet were not toxic
and then determined the effect of non-toxic concentration
.400 mg/kg (600 and 1800 mg/kg) on the yield of AOM-
Prevention by aspirin of colon cancer
induced colon cancer. When administered starting before and/
or after AOM, aspirin at 200, 600 and 1800 mg/kg diet did
not affect the yield of colon tumors except for a reduction
resulting from the highest concentration of aspirin administered
after AOM. Aspirin administered to rats at 1800 mg/kg diet
is equivalent to ~240 mg/kg body wt, 16.8 g/70 kg person or
3.36 tablets (5 g tablets)/day. The low activity of this relatively
high daily dose of aspirin could account for the conflicting
results of several epidemiological studies that found an inverse
association between chronic use and the risk of colorectal
cancer (3–6), while one relatively large prospective study
failed to observe an association (7).
Compared with other NSAIDs, the prevention of colon
cancer by aspirin is relatively weak. Piroxicam and sulindac
have been reported to prevent colon cancer at concentration
as low as 25 and 160 mg/kg diet (13,15). Aspirin is primary
a COX-1 inhibitor, while these two NSIADs inhibit both COX1 and 2 (2,18,49). The level of COX-2 in the colon has been
associated with the development of tumors. Hence, the greater
ability of the other two NSAIDs to inhibit COX-2 could
explain their greater efficacy in preventing AOM-induced
colon cancer (13–17).
DFMO has previously been reported to prevent colon cancer
(17,30) and ACF (33) when administered starting before AOM.
The present study demonstrated that DFMO was also effective
in preventing colon tumors when administered starting 11
weeks after AOM (Table II, Figure 3). There was no significant
difference between the efficacy of DFMO administered before
and/or after AOM. Aspirin only significantly prevented colon
tumors when administered after AOM. Thus much of the
ability of aspirin and DFMO to prevent colon cancer would
appear to occur during the promotion/progression phase.
The combination of DFMO with aspirin appeared to be
more effective in preventing colon tumors than DFMO.
Although, aspirin at 200 and 600 mg/kg did not prevent colon
cancer, when administered with DFMO starting before AOM
they caused an apparent dose-related reduction of colon tumors
(Figure 3). The combination of DFMO with piroxicam has
been reported to be synergistic in preventing colon cancer
when administered before AOM (16,17). When administered
after AOM, we did not observe an enhancement of the efficacy
of DFMO by the addition of aspirin (200 and 600 mg/kg).
The inability to detect an additive or synergistic effect of
aspirin was probably due to the already low yield of colon
tumors in the presence of DFMO (,0.5 tumors/rat). Hence,
evaluation of lower concentrations of DFMO that cause less
of a reduction in tumor yield appears to be required to
demonstrate synergism with aspirin.
In the two studies to examine the toxicity of aspirin, we
also determined its effect on the yield of AOM-induced ACF
and on PGE2 levels in the colon. Aspirin concentrations of as
low as 600 and 400 mg/kg diet produced significant reduction
in the yield of ACF and the level of PGE2, respectively.
Wargovich et al. (9) have reported prevention of ACF by 200
and 400 mg/kg aspirin administered 4 weeks after AOM. We
achieved maximum reduction of the yield of ACF and of the
level of PGE2 with 600–2000 and 1000–2000 mg/kg diet,
respectively. Hence, concentrations of aspirin that resulted in
maximum prevention of ACF and reduction of PGE2 levels
did not prevent colon cancer. Since, aspirin is primary a COX1 inhibitor, we propose that it is more active in preventing
ACF and reducing PGE2 levels because this is the primary
isoform in the mucosa, while tumors because of their COX-2
activity are less sensitive.
Both increased apoptosis and/or decreased cell proliferation
have been proposed as mechanisms for the chemopreventive
activity of NSAIDs and other agents (40–45). However, the
enhancement of apoptosis rather than the reduction of cell
proliferation has been more consistently associated with the
ability of NSAIDs to inhibit human cancer cells in culture
(40,41,43,44) and of sulindac to induce the regression of
polyps in FAP patients (46,47). Samaha et al. reported that other
chemopreventive agents including two NSAIDs (curcumin
and sulindac) and phenylethyl-3-methylcaffeate enhanced the
Apoptotic Index in AOM-induced colon tumors (48). In our
study, DFMO and the combination of DFMO 1 aspirin
decreased the Mitotic Index and enhanced apoptosis in adenomas and prevented colon cancer. However, both 600 and
1800 mg/kg aspirin were very effective in decreasing the
Mitotic Index while not affecting apoptosis and very weakly
preventing colon cancer by only the high concentration.
Similarly, we have found that the prevention of AOM-induced
colon cancer by piroxicam (unpublished data) and retinoids
was more closely associated with enhancement of apoptosis
than with the reduction of cell proliferation and the prevention
of ACF (49). However, the prevention by retinoids of AOMinduced ACF was highly associated with decreased cell proliferation (50). Hence, the enhancement of apoptosis appears to
be associated with prevention of colon cancer while decreased
cell proliferation is more associated with prevention of ACF.
This would suggest that the screening of agents for prevention
of colon cancer should include prevention of ACF followed
by an evaluation for the ability to enhance apoptosis in order
to distinguish agents that prevent ACF but do not prevent
colon cancer.
In summary, we report that aspirin and DFMO were effective
in preventing colon cancer when administered 11 weeks after
AOM during the promotional/progression phase of carcinogenesis. However, aspirin had only very weak activity at the
relatively high concentration of 1800 mg/kg diet. Nevertheless
aspirin at much lower concentrations did demonstrate doserelated prevention of ACF and reduction of PGE2 levels in
the colon. The combination of aspirin with DFMO when
administered before AOM was synergistic in preventing colon
cancer. The strong chemopreventive activity of DFMO and
the weak activity of aspirin were more closely associated with
enhancement of apoptosis than with decreased cell proliferation.
Acknowledgement
This work was supported in part by the National Cancer Institute, contracts
NO1-CN-25295–01, NO1-CN-55174–01 and NO1-CN-55175–01.
References
1. World Health Organization (1997) The World Health Report. WHO, Geneva,
Switzerland.
2. Marnett,L.J. (1992) Aspirin and the potential role of prostaglandins in colon
cancer. Cancer Res., 52, 5575–5589.
3. Thun,M.J., Namboodiri,M.M. and Heath,C.W. (1975) Aspirin use and
reduced risk of fatal colon cancer. New Engl. J. Med., 325, 1593–1596.
4. Thun,J.M., Namboodiri,M.M., Calle,E.E., Flanders,W.D. and Heath,C.W.
(1993) Aspirin use and risk of fatal cancer. Cancer Res., 53, 1322–1327.
5. Rosenberg,L., Palmer,J.R., Zauber,A.G., Warshaner,M.E., Stolley,P.D. and
Shapire,S. (1991) A hypothesis: nonsteroidal anti-inflammatory drugs reduce
the incidence of the large-bowel cancer. J. Natl Cancer Inst., 83, 355–358.
6. Martinez,M.E., Mcpherson,R.S., Levin,B. and Annegers,J.F. (1995) Aspirin
429
H.Li et al.
and other nonsteroidal anti-inflammatory drugs and risk of colorectal
adenomatous polyps among endoscoped individuals. Cancer Epidemiol.
Biomarkers Prev., 4, 703–707.
7. Paganini-Hill,A., Hsu,G., Ross,R.K. and Henderson,B.E. (1991) Aspirin use
and incidence of large-bowel cancer in a California retirement community.
J. Natl Cancer Inst., 83, 1182–1183.
8. Reddy,B.S., Rao,C.V., Rivenson,A. and Kelloff,G. (1993) Inhibitory effect
of aspirin on azoxymethane-induced colon carcinogenesis in F344 rats.
Carcinogenesis, 14, 1493–1497.
9. Wargovich,M.J., Chen,C.D., Harris,C., Yang,E. and Velasco,M. (1995)
Inhibition of aberrant crypt growth by non-steroidal anti-inflammatory agents
and differentiation agents in the rat colon. Int. J. Cancer, 60, 515–519.
10. Pereira,M.A., Barnes,L.H., Rassman,V.L., Kelloff,G.V. and Steele,V.E.
(1994) Use of azoxymethane-induced foci of aberrant crypts in rat colon
to identify potential cancer chemopreventive agents. Carcinogenesis, 15,
1049–1054.
11. Craven,P.A. and DeRubertis,F.R. (1992) Effects of aspirin on 1,2dimethylhydrazine-induced colonic carcinogenesis. Carcinogenesis, 13,
541–546.
12. Pence,B.C., Dunn,D.M., Zhao,C., Landers,M. and Wargovich,M.J. (1995)
Chemopreventive effects of calcium but not aspirin supplementation in
cholic acid-promoted colon carcinogenesis: correlation with intermediate
endpoints. Carcinogenesis, 16, 757–765.
13. Reddy,B.S., Maruyama,H. and Kelloff,G. (1987) Dose-related inhibition of
colon carcinogenesis by dietary piroxicam, a non-steroidal antiinflammatory drug during different stages of rat colon tumor development.
Cancer Res., 47, 5340–5346.
14. Pereira,M.A., Barnes,L.H., Steele,V.E., Kelloff,G.V. and Lubet,R.A. (1996)
Piroxicam-induced regression of azoxymethane-induced aberrant crypt foci
and prevention of colon cancer in rats. Carcinogenesis, 17, 373–376.
15. Rao,C.V., Rivenson,A., Simi,B., Zang,E., Kelloff,G.V., Steele,V.E. and
Reddy,B.S. (1995) Chemoprevention of colon carcinogenesis by sulindac, a
nonsteroidal anti-inflammatory agent. Cancer Res., 55, 1464–1472.
16. Reddy,B.S., Nayini,J., Tokumo,K., Rigotty,J., Zang,E. and Kelloff,G.V.
(1990) Chemoprevention of colon carcinogenesis by concurrent
administration of piroxicam, a nonsteroidal anti-inflammatory drug with
D,L-α-difluoromethylornithine, an ornithine decarboxylase inhibitor in diet.
Cancer Res., 50, 2562–2568.
17. Rao,C.V., Tokumo,K., Rigotty,J., Zang,E., Kelloff,G.V. and Reddy,B.S.
(1991) Chemoprevention of colon carcinogenesis by dietary administration
of piroxicam, α-difluoromethylornithine, 16α-fluoro-5-androsten-17-one,
and ellagic acid individually and in combination. Cancer Res., 51, 4528–
4534.
18. Lands,W.E.M. and Hanel,A..M. (1983) Inhibitors and activators of
prostaglandins biosynthesis. In Pace-Asciak,C. and Granstorm,E. (eds)
Prostaglandins and Related Substances. North Holland Biomedical Press,
Amsterdam, Elsevier, pp. 203–223.
19. Roth,G.J. and Majerus,P.W. (1975) The mechanism of the effect of aspirin
on human platelets 1. Acetylation of a particulate fraction protein. J. Clin.
Invest., 56, 624–632.
20. Bennett,A. and del Tacca,M. (1975) Prostaglandins in human colonic
carcinoma. Gut, 16, 409.
21. Jaffe,B.M. (1974) Prostaglandins and cancer: an update. Prostaglandins, 6,
453–461.
22. Karmali,R. (1985) Lipid nutrition, prostaglandins and cancer. In
Lands,W.E.M. (ed.) Biochemistry of Arachidonic Acid Metabolism. Martinus
Nijhoff Press, Boston, MA, pp. 203–211.
23. Narisawa,T., Kusaki,H. and Yamasaki,Y. (1990) Relationship between blood
plasma prostaglandin E2 and liver and lung metastases in colorectal cancer.
Dis. Colon Rectum, 33, 840–845.
24. Pegg,A.E. (1988) Polyamine metabolism and its importance in neoplastic
growth and a target for chemotherapy. Cancer Res., 48, 759–774.
25. Pegg,A.E. (1986) Recent advances in the biochemistry of polyamines in
eukaryotes. Biochem. J., 234, 249–262.
26. McCann,P.P., Bitoni,A.J. and Pegg,A.E. (1992) Inhibition of polyamine
metabolism and the consequent effects on cell proliferation., In
Wattenberg,L., Lipkin,M., Boone,C.W. and Kelloff,G.J. (eds) Cancer
Chemoprevention. CRC Press, Boca Raton, FL, pp. 531–539.
27. Verma,A.K. (1989) The enzyme-activated irreversible inhibitor of ornithine
decarboxylase, D,L-α-difluoromethylornithine: a chemopreventive agent.
Prev. Med., 18, 646–652.
28. Nigro,N.D., Bull,A.W., Wilson,P.S., Soullier,B.K. and Alousi,M.A. (1982)
Combined inhibitions of carcinogenesis: effect on azoxymethane-induced
intestinal cancer in rats. J. Natl Cancer Inst., 69, 103–107.
29. Luk,G.D., Hamilton,S.R., Yong,P., Smith,J.A., O’Ceallaigh,D.,
McAvinchey,D. and Hyland,J. (1986) Kinetic changes in mucosal ornithine
430
decarboxylase
activity
during
azoxymethane-induced
colonic
carcinogenesis in the rat. Cancer Res., 46, 4449–4452.
30. Luk,G.D., Zang,S.Z. and Hamilton,S.R. (1989) Effects of timing of
administration and dose of difluoromethylornithine on rat colonic
carcinogenesis. J. Natl Cancer Inst., 81, 421–427.
31. Sing,J, Kelloff,G. and Reddy,B.S. (1992) Effect of chemopreventive agents
on intermediate biomarkers during different stages of azoxymethane-induced
colon carcinogenesis. Cancer Epidemiol. Biomarkers Prev., 1, 405–411.
32. Kullkarni,N., Zang,E., Kelloff,G.J. and Reddy,B.S. (1992) Effect of the
chemopreventive agents piroxicam and D,L-α-difluoromethylornithine on
intermediate biomarkers of colon carcinogenesis. Carcinogenesis, 13, 995–
1000.
33. Pereira,M.A. and Khoury,M.D. (1991) Prevention by chemopreventive
agents of azoxymethane-induced foci of aberrant crypts in rat colon. Cancer
Lett., 61, 27–33.
34. Bird,R.P. (1987) Observations and quantification of aberrant crypts in the
murine colon treated with a colon carcinogen: preliminary findings. Cancer
Lett., 37, 147–151.
35. Wyllie,A.H., Kerr,J.F.R. and Currie,A.R. (1980) Cell death: the significance
of apoptosis. Int. Rev. Cytol., 68, 251–306.
36. Steel,V.E., Moon,R.C., Lubet,R.A. et al. (1994) Preclinical efficacy
evaluation of potential chemopreventive agents in animal carcinogenesis
models: methods and results from the NCI chemoprevention drug
development program. J. Cell. Biochem., 20, 32–54.
37. Cohen,S.M., Zenser,T.V., Murasaki,G., Fukushima,S., Mattammal,M.B.,
Rapp,N.S. and Davis,B.B. (1981) Aspirin inhibition of N-[4-(5-nitro-2furyl)-2-thiazoly]formamide-induced lesions of the urinary bladder
correlated with inhibition of metabolism by bladder prostaglandin
endoperoxide synthetase. Cancer Res., 41, 3355–3359.
38. Sakata,T., Hasegawa,R., Johansson,S.J., Zenser,T.V. and Cohen,S.M. (1986)
Inhibition by aspirin of N-[4-(5-nitro-2-furyl)-2-thiazoly]formamide
initiation and sodium saccharin promotion of urinary bladder carcinogenesis
in male F344 rats. Cancer Res., 46, 3903–3906.
39. Sharma,S., Zhu,S., Jokinen,M.P., Steele,V.E. and Kelloff,G.J. (1995)
Dysplasia and gamma glutamyl transpeptidase as intermediate endpoint
markers for N-butyl-N-(4-hydroxybutyl)nitrosamine-induced urinary
bladder carcinogenesis in rats and the chemopreventive effect of aspirin.
Proc. Am. Assoc. Cancer Res., 35, A3735.
40. Shiff,S.J., Qiao,L., Tsai,L.L. and Rigas,B (1995) Sulindac sulfide, an aspirinlike compound, inhibits proliferation, cause cell cycle quiescence and
induces apoptosis in HT-29 colon adenocarcinoma cells. J. Clin. Invest., 96,
491–503.
41. Elder,D.J., Hague,A., Hicks,D.J. and Paraskeva,C. (1996) Differential
growth inhibition by the aspirin metabolite salicylate in human colorectal
tumor cell lines: enhanced apoptosis in carcinoma and in vitro-transformed
adenoma relative to adenoma cell lines. Cancer Res., 56, 2273–2276.
42. Hison,L., Albert,D., Krutzsch,M. et al. (1994) Antiproliferative effect of
nonsteroidal anti-inflammatory drugs (NSAIDs) against human colon cancer
cells. Cancer Epidemiol. Biomarkers Prev., 3, 433–438.
43. Piazza,G.A., Kulchak-Rahm,A.L., Krutzsch,M. et al. (1995) Antineoplastic
drugs sulindac sulfide and sulfone inhibit cell growth by inducing apoptosis.
Cancer Res., 55, 3110–3116.
44. Goldberg,Y., Pittas,K. and Reed,J.C. (1995) Sulindac and sulindac sulfide
alter the levels of bcl-2 and bax proteins in HT-29 colon cancer cells.
Gastroenterology, 108A, 475–481.
45. Bedi,A., Pasricha,P.J., Barber,J.P., Bedi,G.C., Piantadosi,S., Giardiello,F.M.,
Zehnbauer,B.A., Hamilton,S.R. and Jones,R.J. (1995) Inhibition of apoptosis
during development of colorectal cancer. Cancer Res., 55, 1811–1816.
46. Pasricha,P.J., Bedi,A., O’Connor,K., Rashid,A., Akhar,A.J., Zahurak,M.L.,
Piantadosi,S., Hamilton,S.R. and Giardiello,F.M. (1995) The effects of
sulindac on colorectal proliferation and apoptosis in familial adenomatous
polyposis. Gastroenterology, 109, 994–998.
47. Boolbol,S.K., Dannenberg,A.J., Chadburn,A. et al. (1996) Cyclooxygenase2 overexpression and tumor formation are blocked by sulindac in a murine
model of familial adenomatous polyposis. Cancer Res., 56, 2556–2560.
48. Samaha,H.S., Kelloff,G.J., Steele,V.E., Rao,C.V. and Reddy,B.S. (1995)
Modulation of apotosis by sunlindac, curcumin, phenylethyl-3methylcaffeate, and 6-phenylhexyl isothiocyanate: apoptotic index as a
biomarker in colon cancer chemoprevention and promotion. Cancer Res.,
57, 1301–1305.
49. Zheng,Y., Kramer,P.M., Lubet,R.A., Steele,V.E., Kelloff,G.J. and
Pereira,M.A. (1999) Effect of retinoids on AOM-induced colon cancer in
rats: Modulation of cell proliferation, apoptosis and aberrant crypt foci.
Carcinogenesis, 20, 255–260.
50. Zheng,Y., Kramer,P.M., Olson,G., Lubet,R.A., Steele,V.E., Kellof,G.J. and
Pereira,M.A. (1997) Prevention by retinoids of azoxymethane-induced colon
cancer and aberrant crypt foci and their modulation of cell proliferation in
rats. Carcinogenesis, 18, 2119–2125.
Received July 13, 1998; revised October 14, 1998; accepted October 30, 1998