[CANCER RESEARCH 49, 5581-5585, October 15, 1989]
Effect of Dietary Cellulose on Cell Proliferation and Progression of
1,2-Dimethylhydrazine-induced
Colon Carcinogenesis in Rats1
David W. Heitman, Virginia A. Ord, Keithley E. Hunter, and Ivan L. Cameron2
Department of Cellular and Structural Biology, The University of Texas Health Science Center at San Antonio, Texas 78284
ABSTRACT
The effects of different levels of dietary cellulose on colonie crypt
mitotic activity and colon carcinogenesis were studied in 190 male Sprague-Dawley rats. Rats were divided into groups and fed a basal fiber-free
diet supplemented with either 0, 5, or 15% pure cellulose (w/w), for
periods of 10 weeks (initiation stage) or 32 weeks (promotional stage).
Half of the rats in each group were given weekly s.c. injections of 9.5 mg
1,2-dimethylhydrazine (the base) (DMH) for 8 weeks. Some of the rats
were killed at 10 weeks while most were killed 22 weeks later. In some
groups the dietary cellulose level was changed to a different level at 10
weeks. Food intake and body weight data showed that the rats within
each experiment were isocalorically fed. There was a direct correlation
between crypt height and the percentage of cellulose in the diet. Addition
of 5 or 15% dietary cellulose during the initiation stage of carcinogenesis
resulted in a significant increase in crypt height. Increasing dietary
cellulose after the initiation stage (0 to 5% and 5 to 15%) or maintaining
a high dietary cellulose level throughout both the initiation and promo
tional stages (15%) resulted in a significant increase in crypt height. A
DMH-induced increase in mitotic activity that was observed during the
initiation stage was no longer evident after the 22-week promotional
stage. The significant DMH-induced increases in proliferative zone
height and crypt height that were initially observed during the initiation
stage were also observed after the 22-week promotional stage. These
data indicate that the initial DMH-induced increases observed in prolif
erative zone height and crypt height are irreversible.
Addition of 5 or 15% cellulose was found to suppress DMH-enhanced
mitotic activity in the crypts of the descending colon during the initiation
stage of carcinogenesis. This finding was correlated with a significantly
lower incidence of adenocarcinomas in rats maintained on 5 or 15%
cellulose throughout both the initiation and promotional stages.
INTRODUCTION
There are a number of published reports which support the
hypothesis that a stimulus to cell proliferation, especially during
the initiation stage of carcinogenesis, enhances tumorigenesis
(1-7). In the case of DMH3-induced colon carcinogenesis there
is already general agreement that DMH itself stimulates cell
proliferation in the colonie crypts during the initiation stage of
carcinogenesis. The corollary to this hypothesis is that inhibi
tion of cell proliferation during carcinogenesis will suppress
tumorigenesis. In an earlier study (8) we found that addition of
5 or 15% (w/w) purified cellulose to a cellulose-free AIN-76 rat
diet did cause a significant suppression of the DMH-induced
increase in mitotic activity in the descending colon of rats. A
prediction from these findings was that addition of dietary
cellulose would suppress later tumorigenesis. Although there
are at least three published reports to indicate that cellulose
when added to a fiber-free diet is protective against development
of carcinogen-induced colon cancer in rats (9-11), there are
three other reports which indicate that cellulose is ineffective
in suppression of tumorigenesis (12-14). Thus we felt it nec
essary to design carefully and to execute experiments to test
the general idea that suppression of mitotic activity in the
colonie crypts could be used to suppress DMH-induced colon
carcinogenesis. We also decided to test whether addition of
dietary cellulose during the initiation stage, the promotional
stage, or both the initiation and the promotional stages would
suppress colon tumorigenesis.
MATERIALS AND METHODS
Animals and Treatments. A total of 190 male Sprague-Dawley rats,
4 to 6 weeks old (mean weight, 104 g; HarÃ-an Sprague Dawley,
Houston, TX), were used in this study. Animals were randomly divided
into two major groups. On group was given eight weekly s.c. injections
of the base DMH at a dosage of 9.5 mg/kg body weight. DMH (Sigma
Chemical Co., St. Louis, MO) was dissolved in a vehicle solution
containing 0.18% EDTA, which was added as a stabilizing agent, and
brought to pH 6.5. Control rats (non-DMH) were given injections of
the same volume of the vehicle solution.
Dietary Protocol. The rats were randomly subdivided into groups and
were fed a basal fiber-free (AIN.76) diet formulation (see Table 1 for
the detailed formulation), supplemented with either 0, 5, or 15%
cellulose (Alphacel nonnutritive bulk), for periods of 10 or 32 weeks.
Some of the rats were killed at 10 weeks while some were killed 22
weeks later. In some groups the dietary cellulose level was changed to
a different level at 10 weeks. The details of dietary cellulose levels fed
during the earlier part of the experiment (first 10 weeks) and the latter
part of the experiment (next 22 weeks) are described in Table 2. The
first 10 weeks were defined as the initiation stage and the 22 weeks
after the initiation stage were defined as the promotional stage. Water
was available ad libitum. Animals were individually housed in metabolic
cages, throughout the duration of the study, at 25°C(14 h light, 10 h
dark). Caloric intake and body weight changes were monitored for each
rat on the different diets.
Tissue Preparation and Analysis. Three h prior to being killed, each
rat received an i.p. injection of colchicine at a dosage of 1 mg/kg body
weight. The rationale and details of this technique, which arrests cells
at metaphase and therefore collects all cells which enter metaphase
over a 3-h period, has been described by Cameron (15). Animals were
killed by an ether overdose. This was done between 9 and 11 a.m. to
avoid diurnal variation. All scoring for tumors at autopsy and all
histológica!analyses were done in a double-blind fashion. The descend
ing colon was removed, opened longitudinally, pinned onto corkboard
serosal side down, and scored for tumor incidence using a magnification
lens. Removal of the colon, pinning down of the colon, and analysis of
the colonie crypts were all done in a double-blind fashion and in a
uniform and consistent manner. Because of the longitudinal folds in
the colon, it is possible to pin down the colon without changing the
morphology of the individual crypts. For light microscopy, the entire
descending colon and obvious and suspected tumors in the ascending
Received 12/12/88; revised 5/25/89; accepted 7/21/89.
and descending colon were fixed in 10% formalin. A 5-mm transverse
The costs of publication of this article were defrayed in part by the payment
segment of the distal colon taken 2 cm above the level of the pelvic rim
of page charges. This article must therefore be hereby marked advertisement in
was embedded in paraffin, and 4-Mm-thick sections were prepared.
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1Supported by National Cancer Institute Grant CA36372 and American
Slides prepared for histológica!analysis were stained with hematoxylin
Cancer Society Grant BC-641.
and eosin.
2To whom requests for reprints should be addressed.
For light microscopic analysis of the crypts in the descending colon,
'The abbreviations used are: DMH, 1,2-dimethylhydrazine; MC, mean num
10 complete midaxial (longitudinal) crypts were used to obtain a mean
ber of metaphase figures per midaxial crypt section; PZH, proliferative zone
value for the epithelial parameters measured in each rat. A midaxial
height; CH, crypt height.
5581
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1989 American Association for Cancer Research.
CELLULOSE,
MITOSIS, AND COLON CARCINOGENESIS
Table 1 Composition of diet by percentage
The diet was the American Institute of Nutrition AIN-76 semipurified diet for
rats. The recommended formula normally contains 95 parts of the listed ingre
dients and 5 parts of cellulose on a weight basis. The exact formulation for this
diet, recommended by the American Institute of Nutrition, is intended for growth
and maintenance during the first year of life.
%CaseinDL-MethionineCorn
Ingredient"
starchSucroseCorn
oilAIN
mix*AIN
mineral
mix*Choline
vitamin
bitartrate21.050.3215.7952.635.263.681.050.21100.00
" To this formula we added 0, 5, or 15% cellulose by weight. According to the
supplier, the cellulose is Alphacel nonnutritive bulk with 5-7% water, 0% protein,
0% starch, 0.2% fat, and 0.2% ash. The three formulated diets were made and
supplied by ICN Nutritional Biochemicals (Cleveland, OH).
4 For a complete description of AIN mineral mix and AIN vitamin mix, see
Ref. 21.
Table 2 Effect of dietary cellulose level and DMH administration on mean daily
caloric intake during the initiation (I) and the promotional (P) stages of
carcinogenesis
Experiment A lasted for a period of 10 weeks (initiation) while Experiment B
lasted for a period of 32 weeks, with a 10-week initiation and 22-week promotion.
•¿
Cellulose
IExperiment
A0515Experiment
B005551515P050515515No
Mean caloric intake
(kcal/day/100g
body weight)
DMH24.9
*23.9
±
o.r
0.324.6
±
0.21 ±
0.224.5
±
0.224.7
±
0.31 ±
0.2e-''13.0
2.9 ±
±0.213.7
±0.212.8
0.213.4
+
±0.113.7
±0.213.8
±0.2DMH25.0
±0.413.1
2.2
±0.313.1
±0.312.6
0.313.3
+
±0.313.1
±0.213.2
+ 0.3
" Mean ±SE for 7 or 8 rats/group.
* One-way analysis of variance of data from Experiment A showed no signifi
cant differences (P < 0.05 or better) means.
c One-way analysis of variance of data from Experiment B showed no signifi
cant differences (P < 0.05 or better) between means.
äOne-way analysis of variance of data from both Experiments A and B showed
both groups, no DMH and with DMH, in Experiment A to be significantly higher
in their mean caloric intake (P < 0.05) than both in Experiment B.
crypt section in the same 10 crypts was also recorded. All obvious and
suspected colon tumors were scored at autopsy and were fixed for
histological preparation. Serial sections of all tumors were examined
for validation and for classification of tumor type. Overt adenocarcinomas (invasive through the muscularis mucosa) and adenomas were
included in the scoring of tumor incidence. The data on each crypt
epithelial parameter as well as tumor incidence were subjected to twoway analysis of variance and, where significant differences were de
tected, a Student-Neuman-KeuFs multiple range test was run to deter
mine which means differed significantly (17).
Rats used for tumor analyses were killed 24 weeks after the end of
an 8-week DMH injection period. The first 10 weeks after the start of
DMH treatment were defined as the initiation stage and the 22 weeks
after the initiation stage were defined as the promotional stage.
RESULTS
Analysis of daily caloric intake of kcal, expressed as kcal/
day/100 g body weight is summarized in Table 2. The data in
the upper portion of the table (Experiment A) are from a
previous study in which animals remained on the same per
centage of dietary cellulose during the initiation stage of carci
nogenesis. These rats were killed after the 10-week initiation
stage. The data in the lower portion of the table (Experiment
B) are from the present study in which rats were killed after 32
weeks (10-week initiation and 22-week promotional stages of
carcinogenesis). The daily caloric intake in Experiment A was
calculated for the entire 10 weeks, and the daily caloric intake
in Experiment B was calculated for the last 2 weeks of the 32week experiment. Analysis of variance showed the rats within
Experiment A (younger rats) to be isocalorically fed and the
rats within Experiment B (older rats) to be isocalorically fed.
Analysis of variance between Experiment A and Experiment B
showed the younger rats to have a significantly (P < 0.05)
higher daily caloric intake compared to the older rats.
The caloric intake calculations assumed that the cellulose did
not make a contribution to the rat. That the calculated caloric
intake is not different between groups within each experiment
indicates that the rats compensate for dilution of their food
with either 5 or 15% cellulose simply by eating more food.
Thus all rats within each experiment got essentially the same
intake of all nutrients but different levels of dietary cellulose.
This study was performed in three parts. In the first part the
percentage of dietary cellulose was kept constant during both
the initiation and the promotional stages. In the second part
the dietary cellulose was kept constant during the initiation
stage (5%) but was varied during the promotional stage (0, 5,
or 15%). In the third part dietary cellulose was varied during
the initiation stage (0, 5, or 15%) but was kept constant during
the promotional stage (5%). The group which received 5%
dietary cellulose during both the initiation and the promotional
stages was common to all three parts of this study; therefore,
these data were pooled for statistical analyses.
As seen in the upper portion of Table 3, the addition of 5 or
15% cellulose to a basal fiber-free diet resulted in suppression
of DMH-enhanced mitotic activity in the crypts of the descend
ing colon during the initiation stage of carcinogenesis. This
finding was correlated with a significantly lower incidence of
adenocarcinomas in rats maintained on 5 or 15% cellulose (P
= 0.0023 and 0.0025, respectively) during both the initiation
and promotional stages, as compared to rats on a fiber-free diet
(Fig. 1). There was an 88% incidence of adenocarcinomas in
rats which remained on a fiber-free diet, as compared to a 61
and 38% incidence in rats which remained on 5 and 15%
cellulose, respectively. Therefore, by maintaining rats on 5 or
crypt was analyzed if its lumen was exposed from the mouth to the
base of the crypt. The sections selected for examination were spaced so
that each crypt section was encountered only once. For scoring and
analysis, each midaxial crypt section was divided at the base into two
crypt columns. Starting at the base of the crypt column, nuclei were
numbered up to the mouth of the crypt and the number and position
of each metaphase figure were scored. Crypt column height was meas
ured in number of cells from the base to the mouth of the crypt. To
define the upper limit of the proliferative zone height in number of
cells, we first scored the cell position of each metaphase figure from
the base of midaxial crypt sections from 10 crypts of each of seven or
eight rats in each treatment group. A frequency distribution from each
rat was made using this information and was analyzed for normality
using the Kolmogorov-Smirnov goodness of fit test (16, 17). The
untransformed data showed much greater and in almost every rat
significant deviation from normality but the square root-transformed
distributions showed normality.
Because the transformed data showed good fits to a normal distri
bution, we could use all of the frequency distribution data from each
rat to help determine the upper limit of the proliferative zone. We
chose to define the upper limit by the 95% confidence limit of the
distribution and this value was then untransformed for use in further
statistical analysis. The mean number of metaphase figures per midaxial
5582
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1989 American Association for Cancer Research.
CELLULOSE.
MITOSIS, AND COLON CARCINOGENESIS
Table 3 Statistical analysis of dietary cellulose as a modulator of the number of
metaphase figures per mid-axial crypt in the descending colon of DM H-treated
and non-DMH-treated rats
Table 4 Statistical analysis of dietary cellulose as a modulator of the
proliferative zone height in number of cells from the base of colon crypts in the
descending colon of DMH-treated and non-DMH-treated rats
The first set of data in this table is from rats that were autopsied 2 weeks after
the end of an 8-week course of injections of either DMH or the vehicle. The first
10 weeks after the start of injections was defined as the initiation (I) stage and
the next 22 weeks after the initiation stage was defined as the promotional (P)
stage.
The first set of data in this table is from rats killed 2 weeks after the end of an
8-week course of injections of either DMH or the vehicle and the second set of
data is from rats killed 22 weeks thereafter.
%Cellulose,
I Stage0
DMH(8)"
means(8)
5.46 ±0.48
3.45 ±
(8)" 3.20 ±0.52
(8) 3.56 ±0.52
(7) 3.63 ±0.53
(7) 3.174.23'No
2.71
±3.140.34*
0.63DMHRow
515Column
% Cellulose,I
Stage0515Columnmeans(8)°(8)(7)No
DMH16.75
±18.38±18.08±17.721.14*0.750.69(8)(8)(7)DMH22.44
±20.09
±23.04
±21.80f0.881.701.66Rowmeans19.5919.2320.56%
4.46r
3.38
CelluloseI00555IS15ColumnP050515515meansNo
means%ICelluloseP0
DMH(8)"(8)(7)(23)(8)(7)(8)11.24±16.23
means68
DMH(8)
0
5
5
5
15
15Column
15
0
5
0
5
15
5
±
(8) .91 ±0.19
(7) ±.52
.90 0.44
(23)
±0.12
(8) .65 ±0.35
(7) .44 ±0.25
(8)2.60
.24
±1.700.24
0.22(8)
±0.69
(7) 1260 ±0.23
(8) 12 ±0.36
(23) 1 80 ±0.14
(7) 1 50 ±0.34
(7) 2 30 ±0.32
71
±0.41SI*.64''
(8)DMH1
11Row
±12.38
±13.16±12.81
.77
!.02
.66
.58
.87
.98
±16.51
±13.89
±13.98±15.05±13.99±15.07±14.42''0.8
±13.19
±12.56±13.110.470.350.610.470.511.080.91(8)(7)(8)(23)(7)(7)(8)DM
meansNo
" Numbers in parentheses, number of rats scored in each group.
* Mean ±SE for each group of rats.
c A Student-Newman-Keul's multiple range test showed the row mean of rats
on 0% cellulose to be significantly higher than the row mean of rats with 5 or
15% cellulose in their diet. The column mean of DMH-treated rats was signifi
cantly higher than the group of rats not given DMH.
d No significant differences between row means or column means.
" Number in parentheses, number of rats scored in each group.
* Mean ±SE for each group of rats.
' Two-way analysis of variance showed that the column mean of the DMHtreated rats was significantly higher than that of the rats not given DMH (P =
0.002).
A Student-Newman-Keul's multiple range test showed that the row mean of
rats on 0% fiber during the initiation stage and 5% fiber during the promotional
stage were significantly higher (P < 0.05) than all other row means, whereas the
column mean of the DMH-treated rats was significantly higher than the rats not
given DMH (P = 0.002).
Table 5 Statistical analysis of dietary cellulose as a modulator of the crypt height
in number of cells from the base to the mouth in the descending colon of DMHtreated and non-DMH-treated rats
The first set of data in this table is from rats autopsied 2 weeks after the end
of an 8-week course of injections of either DMH or the vehicle and the second
set of data is from rats killed 22 weeks thereafter.
5 15
15 5
% Cellulose.
I Stage05
15 Initiation
15 Promotion
Fig. 1. Effect of dietary cellulose on the incidence of DMH-induced colonie
adenocarcinomas. a. the first 10 weeks after the start of DMH treatment was
defined as the initiation stage and the 22 weeks after the initiation stage was
defined as the promotional stage, b, rats maintained on 5 or 15% cellulose during
both the initiation and promotional stages had a significantly lower incidence of
colonie adenocarcinomas, compared to the rats maintained on 0% fiber during
both the initiation and promotional stages, as determined by the normal approx
imation of the binomial distribution (P = 0.0023 and 0.0025, respectively).
15% dietary cellulose there was a 31% and 57% reduction in
incidence of adenocarcinomas, as compared to rats which re
ceived no fiber in their diet.
Tables 3-5 summarize the results of crypt morphometric
measurements and the numbers of metaphase figures per
midaxial crypt section in the descending colon of all rats, as
well as similar measures from a previous study (8). The results
from the previous study were obtained from DMH-treated and
non-DMH-treated rats that were given 0, 5, or 15% cellulose
during the initiation stage of carcinogenesis and were autopsied
at 10 weeks into the experiment (at a time before any tumors
became apparent). The inclusion of these results in the present
report allows one to determine (a) if any of the effects due to
DMH injections that are noted at 10 weeks persist under the
influence of the different dietary cellulose intake levels and (b)
if there are age-dependent changes in the crypts of the descend
ing colon that are attributable to different dietary cellulose
intake levels.
The top set of data in Table 3 showed DMH to significantly
increase the MC and showed that addition of dietary cellulose
significantly decreased the MC in rats killed 2 weeks after the
last DMH or vehicle injection. Comparison of the overall mean
DMH(8)°
34.84 ±0.91*
(8) 36.26 ±0.64
(7) ±0.6438.14No
44.06
15Column
means35.86
1.11
(8) 42.82 ±1.41
1.3143.40DMH(8)
(7)51.47+
39.5647.76Row
means%
CelluloseI
P0
DMH(8)
00
5
5
5
15
15Column
15
5
0
5
15
5
29.02 ±1.09
30.93 ±0.76
(8) 3 1.49 ±0.67
(7) 36.04 ±0.94
(7)29.29 ±1.61
(8) 30.23 ±0.62
(23) 29.23 ±0.50
(23) 31.35 ±0.64
(8) 32.96 + 0.96
(7)34.80+ 1.50
(7) 32.49 ±1.37
(7) 31.36 ±0.81
(8)33.63+
1.1532.28"Row
(8)32.11
1.3530.57DMH(8)36.89+
±
means29.98'
33.6 Ie
29.79'
30.29'
33.82C
31.93
32.87C
meansNo
" Numbers in parentheses, number of rats scored in each group.
* Mean ±SE for each group of rats. Results of two-way analysis of variance
showed a significant interaction between cellulose and DMH treatment (P =
0.03). Appropriate Student-Neuman-Keul's multiple range testing showed the
group of rats given 15% cellulose and DMH to have a significantly higher crypt
height than all other groups. The crypt height of rats on 15% cellulose and no
DMH is significantly higher than the groups on 0 and 5% with no DMH. The
crypt height of rats on 5% cellulose and given DMH is significantly higher than
the crypt height of rats on 0% cellulose and given DMH and is significantly
higher than the crypt height of rats on 5% cellulose but not given DMH.
' The row means marked c are significantly higher than the row means
marked e.
d The column mean of DMH-treated rats was significantly higher than that of
the rats not given DMH (P = 0.0001).
MC value for these rats versus the rats killed 22 weeks later
showed the overall mean MC value to decrease from 3.69 to
1.76, a 52% decrease. This reveals a striking age-dependent
decrease in MC. The significant suppressive effect due to ad-
5583
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1989 American Association for Cancer Research.
CELLULOSE,
MITOSIS, AND COLON CARCINOGENESIS
dition of dietary cellulose seen in the younger rats was not seen
in the older rats, nor was the stimulatory effect of DMH which
was observed in the younger rats seen in the older rats.
The results of two-way analysis of variance for effects of
cellulose level and DMH injections on the PZH are summarized
in Table 4. The top set of data in Table 4 showed DMH to
cause a significant increase in PZH. Comparison of the overall
mean of the younger rats scored in this experiment with the
overall mean of the older rats scored 22 weeks later shows a
decrease in PZH from 19.76 to 13.77, a 30% decrease. Thus
an age-dependent decrease was observed in PZH. The signifi
cant stimulatory effect of DMH on PZH seen in the younger
rats was still present in the older rats. Thus a significant increase
in PZH due to DMH was still present in the rats 24 weeks after
the last DMH injection.
CH data analyses are summarized in Table 5. The younger
group of rats (top of Table 5) showed significant and marked
increases in CH due both to DMH and to dietary cellulose
level. Comparison of the CH in the younger and older rats
revealed an age-dependent decrease in CH from a mean of
40.77 cells to 31.43 cells, a 23% decrease. Increasing the dietary
cellulose after the initiation stage (0 to 5% and 5 to 15%) or
maintaining a high dietary cellulose level (15%) throughout
both initiation and promotion resulted in a significant increase
inCH.
DISCUSSION
There are several published reports which support the idea
that a stimulus to cell proliferation, especially during the initi
ation stage of carcinogenesis, enhances tumorigenesis (1-7).
Barthold and Jonas (18) have shown that intestinal infections
by Citrobacter freundii induced increased cell proliferation and
decreased the latent period for tumor development in DMHtreated mice. Cohen and Mosbach (19) showed that elevation
of cell proliferation by feeding cholic acid increased colon tumor
incidence in carcinogen-treated rats. Williamson and Rainey
(4) have shown that compensatory proliferative responses to
surgical procedures are correlated with the increased incidence
of carcinogen-induced colon tumors in animals. Likewise, as
summarized by Deschner (20), dietary addition of wheat bran
and carrageenan stimulated cell proliferation and increased the
incidence of carcinogen-induced colon tumors. Conversely, the
dietary addition of 0-sitosterol, butylated hydroxyanisole, and
ascorbic acid all lowered cell proliferation in the colonie crypts
and reduced colon tumorigenesis. If stimulation of mitotic
activity during the initiation stage of carcinogenesis enhances
the carcinogenic process, then it follows that suppression of
mitotic activity during the initiation stage of DMH-induced
carcinogenesis might eventually reduce the percentage of rats
which develop colon cancer.
Our study was originally designed to test the hypothesis that
suppression of mitotic activity in the colonie crypts by addition
of dietary cellulose will suppress the carcinogenic process. In a
previous study (8) we demonstrated that addition of dietary
cellulose to a fiber-free diet suppressed a DMH-induced in
crease in the number of metaphase figures per crypt during the
initiation stage of carcinogenesis. The results of the present
study showed that addition of 5 or 15% dietary cellulose during
both the initiation and promotional stages of carcinogenesis
decreased the incidence of adenocarcinomas by 31 to 57%,
respectively. This degree of protection was shown to be highly
significant.
A summary of the literature on the effects of cellulose as the
sole dietary fiber type on DMH-induced colon tumors in rats
is compiled in Table 6. The diets containing cellulose showed
a decrease in the percentage of rats with tumors in five of six
possible comparisons. A \~ test showed the probability of this
occurring by chance to be <0.05. Although individual past
studies have shown a significant inhibition in the number of
tumors per rat due to cellulose added to the fiber-free diet (9,
10), no single past study has, to our knowledge, been able to
show a significant decrease in the percentage of rats which
developed tumors due to addition of cellulose to a fiber-free
diet. The analysis of all data, as shown in Table 6, indicates
that dietary cellulose does indeed have an overall significant
protective effect on the incidence of DMH-induced tumors in
rats. This is in agreement with the present study, in which
addition of 5 or 15% dietary cellulose during both initiation
and promotion significantly decreased the percentage of rats
which developed colonie adenocarcinomas. The type of cellu
lose used in the studies summarized in Table 6 was Avicel
microcrystalline cellulose type PH-105, which is >97% cellu
lose, whereas the type of cellulose used in the present study was
Alphacel nonnutritive bulk, which is >90% cellulose with a
hem ice11u lose content of less than 10%. This means that in the
present study there was a hemicellulose content of less than
0.5% in the 5% cellulose and less than 1.5% in the 15%
cellulose. Of these possible percentages of hemicellulose(s), only
a portion of the molecule (short side chains of sugars) would
be digestible. The long linear backbone of hemicellulose is
composed of /31—>4-linkedsugar units and therefore, is essen
tially nondigestible. This would further decrease the actual
digestible matter present in the diet supplemented with Alpha
cel nonnutritive bulk fiber. The contribution of this small
percentage of digestible matter to the caloric intake, colonie
Table 6 Summary of the effects of purified dietary cellulose on 1,2dimethylhydrazine-induced colon cancers expressed as a percentage of rats with
tumors and as the number of tumors per rat
The data were compiled from the literature. Only studies using s.c. injections
of 1,2-dimethylhydrazine were included in this table.
Cellulose
0
10
0
4.5
% Rats with
tumors"
Mean
number of
tumors/ratfl
33(23)
49(23)
0.57
0.78
Jacobs and Lupton (13) 12 weekly
s.c. injections. 9.5 mg DMH
base/kg body weight, vehicle
contained no EDTA. Male
Sprague-Dawley rats autopsied
30 weeks after initiation of
DMH injections.
70(20)
30(20)
1.30
0.40
Freeman ef al. (9) 16 weekly s.c.
injections containing 12 mg
DMH base/kg body weight, ve
hicle contained EDTA. Male
Wistar rats autopsied in groups
of three or four starting 24
weeks after initiation of DMH
injections.
Reference and Comments
s.c.injections
et al. (IO). 18 weekly
00-4.5*4.5900-4.5*4.5967(15)67(15)33(15)33(15)76(21)52(21)62(21)57(21)1.531.600.670.463.331.861.671.52Fr
DMHbase/kg
of 16.6 mg
vehiclecontained
body weight,
Wistarrats EDTA. Male
onegroup
autopsied at two times,
at 20-22 weeks after ini
tiation of DMH and the
othergroup
at about 36 weeks.
0 The diets containing cellulose show a decrease in the percentage of rats with
tumors in five of six possible comparisons. The probability of this occurring by
chance is <0.05. The diets containing cellulose show a decrease in the mean
number of tumors per rat in five of six possible comparisons. The probability of
this occurring by chance is <0.05. The numbers in parenthesis are the numbers
of rats scored in each experiment.
6 This diet was changed to 4.5% cellulose after the last DMH injection.
5584
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1989 American Association for Cancer Research.
CELLULOSE, MITOSIS, AND COLON CARCINOGENESIS
pH, short chain fatty acids, etc., would be negligible. For this
reason the data presented in Table 6 are comparable to the
present data with respect to the type of cellulose used in the
diets.
Comparisons of the colon crypt morphometric and mitotic
activity values obtained from the younger and the older rats, as
well as their ability to respond to the different dietary levels of
cellulose, revealed several significant findings. The most strik
ing finding was the marked age-dependent decrease in MC in
the descending colon (52%), as well as the decrease in PZH
and CH. Thus there is a strong age-dependent change in these
crypt parameters. The increased proliferation rate of the
younger rats may be due, at least in part, to a significantly
higher caloric intake as compared to the older rats. These age
changes may help explain the decreased or absent significant
cryptai responses to the different dietary cellulose levels in the
older versus the younger rats. One may speculate from these
facts that (a) if suppression of MC is protective against colon
carcinogenesis then the older rats would be less susceptible to
DMH-induced carcinogenesis because of a naturally low MC
and (b) addition of cellulose to a fiber-free diet in DMH-treated
older rats would afford little if any protection against colon
carcinogenesis.
To simplify data interpretation, only one dietary fiber type,
cellulose, was included in our study. This decision allowed us
to draw some firm conclusions about cellulose. However, it
would be wrong to assume that these conclusions will hold true
for other dietary fiber types or will hold true for complex
mixtures of fibers and other nutrients in various food types.
Further experimental studies are needed to clarify which fiber
types are protective against development of tumors as well as
to determine the physicochemical and physiological effects of
the various fiber types in prevention of colon carcinogenesis. It
seems prudent to undertake such experimental studies and to
compile this type of information before large-scale clinical trials
of various fiber supplements are implemented.
ACKNOWLEDGMENTS
The authors are grateful to Lita Chambers for her careful typing of
this manuscript.
REFERENCES
1. Jacobs, L. R. Enhancement of rat colon carcinogenesis by wheat bran
consumption during the stage of 1,2-dimethylhydrazine administration. Can
cer Res., 43: 4057-4061, 1983.
2. Farber, E. Cellular biochemistry of the stepwise development of cancer with
chemicals. G. H. A. Clowes memorial lecture. Cancer Res., 44: 5463-5474,
1984.
3. Romagnoli, P., Filipponi, F., Bandettini, L., and Bregnola, D. Increase of
mitotic activity in the colonie mucosa of patients with colorectal cancer. Dis.
Colon Rectum, 27: 305-308, 1984.
4. Williamson, R. C. N., and Rainey, J. R. The relationship between intestinal
hyperplasia and carcinogenesis. Scand. J. Gastroenterol., 19 (Suppl. 104):
57-76, 1984.
5. Jacobs, L. R. Stimulation of rat colonie crypt cell proliferative activity by
wheat bran consumption during the stage of 1,2-dimethylhydrazine admin
istration. Cancer Res., 44: 2458-2463, 1984.
6. Heitman, D. W., Grubbs, B. G., Heitman, T. O., and Cameron, I. L. Effects
of 1,2-dimethylhydrazine treatment and feeding regimen on rat colonie
epithelial cell proliferation. Cancer Res., 43: 1153-1162, 1983.
7. Deschner, E. E. Cell kinetic approaches to defining premalignant conditions.
In: J. R. F. Ingall and J. Mastromarino (eds.), Carcinoma of the Large Bowel
and Its Precursors, pp. 187-202. New York: A. R. Liss, 1985.
8. Cameron, I. L., Ord, V. A., Hunter, K. E., Padilla, G. M., and Heitman, D.
W. Suppression of a carcinogen (DMH)-induced increase in mitotic activity
in the colonie crypts of rats by addition of dietary celluloses. Cancer Res.,
49:991-995, 1989.
9. Freeman, H. J., Spiller, G. A., and Kim, Y. S. A double-blind study on the
effects of purified cellulose dietary fiber on 1,2-dimethylhydrazine-induced
rat colonie neoplasia. Cancer Res., 38: 2912-2917, 1978.
10. Freeman, H. J., Spiller, G. A., and Kim, Y. S. A double-blind study on the
effects of differing purified cellulose and pectin fiber diets on 1,2-dimethylhydrazine-induced rat colonie neoplasia. Cancer Res., 40: 2661-2665, 1980.
11. Nigro, N. D., Bull, S. W., KJopfer, B. A., Pak, M. S., and Campbell, R. L.
Effect of dietary fiber on azoxymethane-induced intestinal carcinogenesis in
rats. J. Nati. Cancer Inst., 62: 1097-1102, 1979.
12. Kritchevsky, D. Fiber and cancer. In: G. V. Vahouny and D. Kritchevsky
(eds.), Dietary Fiber: Basic and Clinical Aspects, pp. 427-432. New York:
Plenum Press, 1984.
13. Jacobs, L. R., and Lupton, J. R. Relationship between colonie luminal pH,
cell proliferation, and colon carcinogenesis in 1,2-dimethylhydrazine-treated
rats fed high fiber diets. Cancer Res., 46: 1727-1734, 1986.
14. Ward, J. M, Yamamoto, R. S., and Weisberger, J. H. Cellulose dietary bulk
and azoxymethane-induced intestinal cancer. J. Nati. Cancer Inst., 57: 713715, 1973.
15. Cameron, I. L. Cell proliferation and renewal in the mammalian body. In: I.
L. Cameron and J. D. Thrasher (eds.), Cellular and Molecular Renewal in
the Mammalian Body, pp. 45-85. New York: Academic press, 1971.
16. Sokal, R. R., and Rohlf, F. J. Biometry, 2nd Ed., pp. 617-669. New York:
Freeman and Co., 1981.
17. Nie, N., Hull, C. H., Jenkins, J. G., Stinbrenner, K., and Brent, D. H. SPSS
Users Guide. New York: McGraw-Hill Book Co., 1983.
18. Barthold, S. W., and Jonas, A. M. Morphogenesis of early 1,2-dimethylhy
drazine-induced lesions and latent period of colon carcinogenesis in mice by
a variant of Citrobacterfreundii. Cancer Res., 37:4352-4360, 1977.
19. Cohen, B. I., and Mosbach, E. H. The role of bile acid in colon cancer. In:
G. V. Vahouny and D. Kritchevsky (eds.), Dietary Fiber, pp. 487-496. New
York: Plenum Press, 1984.
20. Deschner, E. E. Kinetics of normal, preneoplastic, and neoplastic colonie
epithelium. In: M. P. Moyer and G. Poste (eds.). Colon Cancer Cells, pp.
41-61. San Diego: Academic Press, 1989.
21. Beri, J. G., et al. Report of the American Institute of Nutrition Ad Hoc
Committee on Standards for Nutritional Studies. J. Nutr., 707: 1340-1348,
1977.
5585
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1989 American Association for Cancer Research.
Effect of Dietary Cellulose on Cell Proliferation and Progression
of 1,2-Dimethylhydrazine-induced Colon Carcinogenesis in Rats
David W. Heitman, Virginia A. Ord, Keithley E. Hunter, et al.
Cancer Res 1989;49:5581-5585.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/49/20/5581
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1989 American Association for Cancer Research.