![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
3
*
Department of Medicine,
Research Institute and
**
Department of Pathology, The Mary Imogene Bassett Hospital, Cooperstown, NY 13326 and
Center for Cell Biology and Cancer Research, The Albany Medical College, Albany, NY 12208
3To whom correspondence and reprint requests should be addressed at 287 Long Point Road, Harpswell, ME 04079.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
KEY WORDS: • human colon cancer • acarbose • short-chain fatty acids • colonic crypt cell proliferation • butyric acid
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) and induces
differentiation in both the rumen mucosa (Sander et al. 1959
) and colon tumor cells (Tsao et al. 1982
,
Whitehead et al. 1986
). Although butyrate inhibits
growth of transformed cells in vitro (Kim et al. 1980
,
Kruh 1982
, Morita et al. 1982
,
Whitehead et al. 1986
), in vivo animal studies
(Lupton and Kurtz 1993
, Sakata 1987
,
Velázquez et al. 1996b
) and in vitro human studies
(Bartram et al. 1993
, Scheppach et al. 1992
) suggest that butyrate stimulates normal colonic mucosal
proliferation. Numerous findings suggest that the short-chain fatty
acids (SCFA), particularly butyrate, play a role in colon cancer
prevention [for reviews see Hague et al. (1997)
,
Hassig et al. (1997)
, Jass (1985)
,
Kruh (1982)
, Roediger (1988)
,
Scheppach et al. (1995)
and Velázquez et al. (1996a)
].
Acarbose, an 
-glucosidase inhibitor, slows digestion of
disaccharides and starch and is used to treat diabetes (Chiasson et al. 1994
). Acarbose expands the proportion of butyric acid
in the SCFA of feces of subjects consuming high starch diets
(Scheppach et al. 1988
) and unrestricted diets
(Weaver et al. 1997
). Augmentation of butyrate may occur
because microbial fermentation of starch produces more microbial
butyrate production than fermentation of other carbohydrates such as
pectin or arabinogalactan (Weaver et al. 1992
). Thus,
acarbose provides a means to test the effects of augmented starch
fermentation and butyrate production on the colonic mucosa.
Testing the anticancer efficacy of potentially preventive agents and
diets takes years, and preliminary studies using surrogate markers have
been suggested (Kelloff et al. 1994
). Increased
proliferation or an upward shift in the crypt distribution of
proliferating cells has been associated with risk for colonic neoplasia
(Deschner et al. 1963
, Deschner and Lipkin 1975
, Lipkin et al. 1987
and 1983
,
Paganelli et al. 1991
, Paspatis et al. 1998
, Risio et al. 1991
, Terpstra et al. 1987
) and has been used to test preventive strategies
(Alberts et al. 1990
, Lipkin and Newmark 1985
). Although 3H-thymidine labeling was
used initially to assess proliferation, comparable results between
3H-thymidine and bromodeoxyuridine (BrdU)
labeling of S-phase cells have been demonstrated (Lin and Allison 1993
, Qin and Willems 1993
). Ki-67
(Sasaki et al. 1987
), a nuclear antigen, is a
potentially complementary marker to BrdU. BrdU labels only cells that
actively synthesize DNA. Ki-67 is expressed in proliferating cells in
late G1, S, G2 and M phases
but not in early G1 or in nonproliferating cells
(G0) (Gerdes et al. 1984
). p52 is
an intracellular cytoskeletal-protein of normal colonic epithelium
that regulates cell shape or cell-to-substrate adhesion (Higgins et al. 1991
) and occurs in the lumenal and upper crypt or
maturational zones (Higgins and Tanaka 1991
) and in the
more highly differentiated epithelial cells of other tissues
(Higgins and Tanaka 1991
). Consequently, it would be
expected to have a negative association with proliferation markers.
Apoptosis can be estimated by measuring Lewis-Y antigen, an
oligosaccharide that is commonly expressed in gastrointestinal tumors
(Hiraishi et al. 1993
). None of the above markers has
been measured simultaneously in the same study with human subjects with
a history of colonic neoplasia and subjects without neoplasia.
The first aim of this study was to determine whether the distribution of proliferative and differentiated cells detected by these biomarkers differed in the two patient populations. To this end, we compared BrdU uptake and the presence of the antigens, Ki-67, p52 and Lewis-Y, in subjects with and without a history of colonic neoplasia in a colonoscopy survey. The related aim was to compare mucosal differences from the colonoscopy survey with changes caused by augmentation of colonic fermentation and butyrate production in a second population with a history of colonic neoplasia.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subjects were patients undergoing colonoscopy. All patients received an oral electrolyte bowel preparation, Colyte (Schwarz Pharmaceuticals, Mequon, WI) or Golytely (Braintree Laboratories, Braintree, MA), containing polyethylene glycol 3350, NaCl, KCl, NaHCO3 and Na2SO4 the evening before colonoscopy.
The patients were divided into two diagnostic groups, i.e., patients with a history of colonic neoplasia (n = 26, 3 with carcinoma) and control subjects (n = 24), those patients without a history of colonic neoplasia. Exclusion criteria were as follows: incomplete colonoscopy; history of more than a limited segmental colon resection; resection of the ileocecal valve; inflammatory bowel disease; use of >750 mg calcium/d; use of >625 µg vitamin D/wk; daily use of bile salts; or antibiotic use within 1 mo. The ages of the 30 men and 20 women enrolled in the survey ranged from 35 to 78 y with a mean of 60.6 y. The mean ± SEM age for the neoplasia subjects (n = 26) and controls (n = 24) was 65.7 ± 1.97 y and 55.2 ± 2.52 y, respectively. Patients defined as having neoplasia required the presence or history of at least one of the following: 0.5 cm or greater adenoma; at least two smaller adenomas; or colon cancer. Controls had no abnormality other than diverticulosis or hyperplastic polyps. Four biopsies of normal mucosa were taken 10 cm from the anal verge during colonoscopy using standard biopsy forceps.
Acarbose trial.
The study was a randomized double-blind placebo-controlled
crossover trial. Subjects were randomized to receive either acarbose
(100 mg three times per day) (Bayer, West Haven, CT) or placebo three
times per day. After 4 mo, the subjects took no study agent for 3–4 mo
and then took the opposite study agent (acarbose or placebo) for 4 mo
(Fig. 1
). Biopsies of normal mucosa were taken as in the colonoscopy survey but
without prior bowel preparation.
|
Subjects (n = 50) were enrolled and randomly assigned
to receive either placebo or acarbose first. Randomization, data
collection points and subject withdrawals are shown in Figure 1
.
Forty-four subjects (15 women and 29 men) completed all study
visits (age range: 44.3–74.5 y with a mean of 61.9 y). For
technical reasons, some analyses could not be performed. For most
analyses, 43 or 44 paired subject values were available.
Placebo tablets were prepared by Bayer; they consisted primarily of starch and were identical in color, size and appearance to acarbose tablets. As subjects were recruited, they were assigned to a previously determined, computer-generated randomization schedule. Each sequence of 10 study numbers contained five assigned initially to acarbose and five assigned initially to placebo.
Compliance was monitored by counting the tablets returned at each visit. The average daily tablet intake at the completion of each treatment was 2.73 ± 0.07 for acarbose and 2.68 ± 0.07 tablets for placebo.
Histologic methods (colonoscopy survey and acarbose trial).
Biopsies were placed in Hank’s balanced salt solution (GIBCOBRL Life
Technologies, Grand Island, NY) for transport. The incubation solution
consisted of 9 mL of RPMI-1640 medium with 0.025 mmol/L Hepes buffer
and L-glutamine (GIBCOBRL Life Technologies), 1 mL of fetal
calf serum and 0.05 g/L of BrdU (Sigma Chemical, St. Louis, MO). A
mixture of 5% CO2/95%O2 gas was bubbled for
30 s into the flasks containing the incubation solution and
mucosal biopsies before sealing and incubation at 37°C for 1 h
in a shaker bath. After 24 h in Carnoy’s fixative, the tissues
were embedded in paraffin. Tissue sections (4 µm) were
taken at 40-µm intervals. Slides for BrdU, Ki-67 and
Lewis-Y were deparaffinized, placed in buffer (citric acid
monohydrate and sodium citrate dihydrate, pH 5.5–5.7, Biotek
Solutions, Santa Barbara, CA) and microwaved. The slides were then
processed through a 2-d protocol on the TechMate 1000 stainer (Biotek
Solutions), using a series of Biotek reagents including intermediate
buffer washes, and sequentially treated with Biotek enzyme solution
containing pepsin at pH 7.0 (only BrdU) and blocking solution. Slides
were incubated overnight in the primary antibody, a 1:1000 dilution of
monoclonal mouse anti-BrdU (DAKO, Carpinteria, CA); a 1:100
dilution of monoclonal mouse anti-Ki-67 (Immunotech, Westbrook, ME); or
a 1:75 dilution of monoclonal mouse anti-Lewis-Y antibody (DAKO). The
slides were then placed in biotinylated polyvalent (rabbit, mouse, rat)
secondary antibody followed by hydrogen peroxide, avidin-biotin
complex, diaminobenzidine chromogen and hematoxylin counterstain
(Bioteck Solutions). Control slides known to stain positively were
processed with each batch of investigational slides and examined for
appropriate and inappropriate staining. Slides for p52 were
deparaffinized with xylene, rehydrated in a graded ethanol series and
immunocytochemistry carried out basically as described by Holt et al. (1995)
.
Labeled hemicrypts scored per subject in the colonoscopy survey were (mean ± SEM) 66 ± 2.1 for BrdU, 64.7 ± 3 for Ki-67, 18.3 ± 1 for p52 and 34.2 ± 1.9 for Lewis-Y. Labeled hemicrypts scored per subject in the acarbose trial after acarbose use and placebo use, respectively, were (mean ± SEM) 40.4 ± 3.6 and 47.5 ± 4.1 for BrdU, 53.9 ± 3.1 and 59.1 ± 3.2 for Ki-67, 38.4 ± 4.4 and 39.7 ± 4 for Lewis-Y and 17.7 ± 0.6 and 17.7 ± 0.7 for p52. Fewer hemicrypts were sought for Lewis-Y and p52 because of the continuous distribution of these antigens. Each crypt was divided longitudinally into two hemicrypts and each cell in a column of hemicrypt cells was scored as 1 (positive staining) or 0 (no staining) with the score electronically placed in a spread sheet column and transferred to SAS. p52 slides were scored with manual recording of positive (1) or no staining (0) of each hemicrypt cell. SAS programs divided each spread sheet column into five equal compartments when the number of cells was divisible by 5. Otherwise, the lower four compartments were assigned equal numbers of cells and the remainder assigned to compartment 5. The total labeling percentage was defined as the percentage of cells labeled in a hemicrypt. The distributions of labeled crypt cells were analyzed as the percentage of compartment cells labeled in each of the five approximately equal hemicrypt compartments (number of labeled cells in a compartment divided by the number of cells in that compartment as a percentage).
Fecal samples and SCFA analyses (acarbose trial).
Subjects voided fecal samples into polypropylene biohazard bags (Fisher
Scientific, Pittsburgh, PA), closed the bags and placed the bags in
styrofoam coolers on ice. Samples were collected the evening before or
morning of study visits. When subjects were unable to collect a sample
in the 24 h before a study visit, samples were collected as soon
as possible after the study visit. Dry feces (Weaver et al. 1986
) and SCFA (Weaver et al. 1997
) were
determined on fecal suspensions.
Dietary history.
Acarbose trial subjects recorded what they ate for the 4 d before each study visit. These diet records were analyzed using the Nutritional Data System (Nutrition Coordinating Center, University of Minnesota, Minneapolis, MN).
Statistical methods (colonoscopy survey and acarbose trial).
The sample size estimate for the colonoscopy survey was based on the
results of Risio et al. (1991)
, suggesting that a sample
of 25 in each diagnostic group would provide a power of 0.83 to detect
a difference in labeling index of 2.0 between the groups with an of
0.05 (two-tailed).
Sample size estimates for the acarbose trial were based initially on
the results of a wheat bran trial (Alberts et al. 1990
).
Six of eight high labeling patients significantly lessened labeling
with wheat bran treatment. Using these proportions, a sample size of 40
with an of 0.05 (two-tailed) and a proportional difference of 0.3
would have a power of 0.83. A requirement for 40 subjects was also
deduced from the results of our colonoscopy survey described in this
report using BrdU labeling of crypt compartment 3. Fifty subjects were
enrolled, with the anticipation that 10 would drop out of the study.
Statistical analyses were carried out with the Statistical Analysis System (SAS Institute, Cary, NC) using a MicroVAX 3190 computer. Differences between diagnostic groups for the colonoscopy survey were tested using mixed-model repeated-measures ANOVA with crypt compartment as a repeated measure. Bonferroni t tests were used to isolate crypt compartment differences.
The primary outcomes for the acarbose trial, i.e., changes in histologic labeling and SCFA between acarbose and placebo treatments, were assessed using two-way repeated-measures ANOVA to examine treatment (acarbose or placebo) and duration of treatment for SCFA (1, 2 or 4 mo of treatment) or crypt compartments (1–5) for the histologic methods. Type III sum-of-squares error terms were used for analyses except as noted.
The secondary outcomes, correlations between labeling methods and SCFA, used Pearson’s correlation coefficients. Correlation coefficients were compared for differences between diagnostic groups or treatment-related changes using Fisher’s z test for difference between two correlation coefficients.
Intestinal gas symptoms were rated on a scale of 1 to 10 with 5 being the normal condition, 1 indicated fewer and 10 more symptoms. These scores were compared by repeated-measures ANOVA. Dietary variables were compared using repeated-measures ANOVA as described above for the SCFA. All data values are expressed as the mean ± SEM.
Study approval.
Both studies were reviewed and approved by the Institutional Review Board of Bassett Hospital. Signed informed consent was obtained from each subject after a detailed discussion of the study plan and potential risks.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The BrdU and Ki-67 total labeling percentages and compartment labeling
percentages are shown in Figure 2
. Percentages of both markers lessened as the surface was approached, as
expected for markers of proliferation. The control group had a higher
percentage of BrdU-labeled cells. Bonferroni t tests for
BrdU labeling of individual crypt compartments did not reach the
P < 0.01 required for significance
(P-values for the differences between crypts 3, 4 and 5 of
the two groups were 0.012, 0.026 and 0.029, respectively). Crypt
compartment scores for BrdU labeling were used to predict the
diagnostic group using a stepwise logistic regression procedure. The
best prediction of diagnostic group was achieved when the data for
compartment 3 alone were used. Predicted probabilities were concordant
69.3% and discordant 30.3%. The 
2 score for
the third BrdU compartment was 6.444 (P = 0.011).
|
).
|
). When the correlation coefficients of individual compartments were
compared, the correlations between the two labeling methods were
significantly different between diagnostic groups for compartment 4 and
approached significance (P = 0.06) in
compartment 3.
|
). There was no significant difference in the percentage of labeling
between groups for either p52 or Lewis-Y antigens. There was a
significant difference in the pattern of labeling for p52, consistent
with earlier expression of p52 antigen in the neoplasia group.
|
Scores rating rectal gas and intestinal gas discomfort were both higher during acarbose use (P = 0.0001). The scores for gas output at the completion of each treatment were 8.1 ± 0.2 for acarbose and 5.6 ± 0.3 for placebo and were consistent with augmentation of colonic fermentation during acarbose use.
No important dietary changes were noted between treatment periods. Nutritional Data System variables, total carbohydrate, percentage of carbohydrate, total fiber, soluble fiber, insoluble fiber, pectin, starch, sucrose, galactose, glucose, fructose, lactose and vegetable protein were examined by ANOVA. The only changes noted were higher sucrose consumption at the 1-mo placebo visit and an average of 2 g/d less lactose consumption during acarbose use.
Butyric acid concentrations and percentages of total SCFA were
augmented with acarbose treatment (Fig. 5
). The ANOVA interaction term for butyric acid for Treatment and Time
was not significant, suggesting that steady states of butyrate
production were present. Acetate and propionate percentages of total
SCFA and propionate concentrations lessened with acarbose treatment.
|
were more concentrated with acarbose
use than placebo. The interaction term for Treatment and Crypt
Compartment was not significant. Crypt length measured in cells during
Ki-67 assessment lengthened with acarbose use (60.9 ± 1) compared
with placebo use (55.4 ± 0.8) by paired t test
(P = 0.0001), consistent with growth stimulation.
Correlation coefficients between BrdU and Ki-67 labeling improved with
acarbose treatment (Table 1)
.
The p52 antigen (Paganelli et al. 1993
) was most
concentrated in the lumenal compartments but did not change with the
two treatments (Fig. 4)
. Lewis-Y antigen universally stained the
mucosal surface and the percentage of crypt cells in compartments 2 and
3 was greater with acarbose treatment (Fig. 4)
.
Associations of fermentation products and proliferation markers were
sought using correlation coefficients and stepwise linear regression.
The most consistent and significant relationship was between Ki-67
labeling and fecal butyrate concentration and percentage of total SCFA
(Table 1)
. Ki-67 labeling and butyrate were positively correlated with
acarbose use, whereas some negative correlations were present with
placebo use. No significant correlation patterns were present between
BrdU labeling and SCFA concentrations or percentages.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Expansion of mucosal cell proliferation has been suggested as both a
marker of risk for colon cancer and an important factor in
carcinogenesis. Although some crypt labeling studies (Deschner and Lipkin 1975
, Lipkin et al. 1987
,
Paganelli et al. 1991
, Paspatis et al. 1998
, Risio et al. 1991
, Terpstra et al. 1987
) support these concepts, our colonoscopy survey results
did not show expansion of proliferation in subjects with a history of
colonic neoplasia. Other crypt labeling studies (Jass et al. 1997
, Kashtan et al. 1992
, Kohn et al. 1997
, Marra et al. 1994
, Rozen et al. 1998
, Wong et al. 1995
) and flow cytometric
studies (Nakamura et al. 1995
, Nsien et al. 1991
) do not support expansion of proliferation as a marker of
risk. Furthermore, patchy expansion of the proliferative zone into the
upper crypt (Lipkin 1974
) as initially described may be
explained by areas of microadenoma (Jass et al. 1997
,
Nakamura et al. 1995
). Confounding variables could
potentially explain conflicting results; however, we found no influence
on our results of the potential confounders of age, family history of
neoplasia or a history of hyperplastic polyps. Because of the variable
results of studies using single proliferation markers (S-phase labeling
or Ki-67), single markers do not appear to be reliable indicators of
risk for colon neoplasia.
Although we found no difference in Ki-67 labeling percentage
between the diagnostic groups of the colonoscopy survey, BrdU and Ki-67
labeling percentages of individual control subjects were better
correlated than those of subjects with a history of colonic neoplasia
(Table 1)
. This suggests greater synchrony between
G1 duration and S-phase entry in the
proliferating cells of the control subjects. Kohn et al. (1997)
did not find Ki-67 proliferation differences between
colon cancer patients and controls but did find better correlation
between retinoblastoma protein labeling and Ki-67 labeling in controls
than in colon cancer patients. Because retinoblastoma protein is
important for S-phase entry, their results are consistent with the
BrdU and Ki-67 correlation differences found in our study. Differences
in synchronous labeling in our colonoscopy survey subjects and in other
studies (Kohn et al. 1997
, Shmakov et al. 1995
) suggest that studies designed to detect disruption of
cell cycle synchrony might help in understanding the disordered
cellular physiology leading to colonic neoplasia or in detecting
populations prone to colonic neoplasia.
The p52 labeling distributions in colonoscopy survey neoplasia subjects
suggest a more basilar distribution than in the control subjects.
Although no correlation was found between subject age and p52
expression in crypt compartments 2 and 3 in which the greatest
difference between diagnostic groups occurred, p52 distribution does
change in aging rats (Holt et al. 1995
). Because p52 is
not present in immature crypt base cells, cell differentiation and p52
formation may have occurred earlier in the neoplasia group. As with
p52, greater expression of Lewis-Y antigen occurred in surface
crypt compartments. However, crypt compartment distributions were
similar for the two diagnostic groups, suggesting similar levels of
apoptosis in both groups.
Because both BrdU and p52 labeling showed differences between diagnostic groups, their crypt compartment scores were used in a stepwise logistic regression procedure by SAS to predict the diagnostic group. The best prediction was achieved using only compartment 3 for BrdU. Considering this and the correlation patterns of BrdU and Ki-67, the most important labeling difference between the diagnostic groups of the colonoscopy survey was for the correlations of BrdU and Ki-67 labeling. Therefore it would be of interest to have further studies that simultaneously measure other cell cycle markers that assess differing portions of the cell cycle.
Acarbose trial.
Acarbose use caused significant changes in mucosal proliferation and SCFA. The positive correlation of butyrate with Ki-67 during acarbose use suggests a causal relationship between butyrate production and mucosal proliferation. Although acarbose simultaneously reduced propionate production there were no significant correlations between propionate and Ki-67 or BrdU labeling during acarbose use, suggesting that propionate reduction did not enhance proliferation.
The responses to acarbose occurred without dietary restrictions or
starch supplementation throughout the 4-mo treatment period. Comparison
of dietary records between the study periods showed only minor dietary
carbohydrate differences between study periods so that acarbose use was
the primary dietary intervention. The enhancement of butyrate values
was comparable to that found in previous short-term studies
(Scheppach et al. 1988
, Weaver et al. 1997
). A study using higher doses of acarbose (300 mg three
times per day) (Holt et al. 1996
) showed lesser fecal
butyrate differences, possibly because poorly absorbed acarbose may
inhibit microbial (Weaver et al. 1997
) as well as host
amylases.
Positive butyrate and Ki-67 correlations with acarbose use
contrast with the negative butyrate and Ki-67 correlations with placebo
use (Table 1)
. The positive correlations might be explained by the more
rapid or frequent transition of cells from early
G1-phase to later G1-phase.
Ki-67 does not label early G1 cells. The lack of
butyrate and BrdU (S-phase labeling) correlation also suggests that
butyrate’s effect occurs early in the cell cycle.
There were significantly more cells expressing Lewis-Y antigen in crypt compartments 2 and 3 with acarbose treatment than with placebo treatment. Earlier expression of Lewis-Y antigen suggests that the mucosal growth caused by acarbose may have been compensated in part by an enhanced commitment to apoptosis. p52 protein had similar distributions between the treatment periods. No clear associations between SCFA and Lewis-Y and p52 antigen distributions were found, suggesting that fermentation products were not closely associated with their expression.
Comparison of the colonoscopy survey and the acarbose trial.
Acarbose-treated subjects and control subjects of the colonoscopy
survey showed relatively higher proliferation (assessed by BrdU
labeling). The correlation of BrdU and Ki-67 labeling significantly
improved with acarbose (Table 1)
, paralleling a better correlation of
BrdU and Ki-67 labeling in control subjects of the colonoscopy survey
(Table 1)
. These parallels suggest that augmentation of butyrate from
acarbose created a more benign histologic milieu. These results and
higher butyrate values in normal subjects compared with those with
colonic neoplasia (Clausen et al. 1991
, Kashtan et al. 1992
, Weaver et al. 1988
) offer a
circumstantial linkage of butyrate and histologic change that may
reflect reduced neoplasia risk. Improved cell cycle synchrony
associated with butyrate suggests that butyrate optimizes cell cycle
timing for the least probability of replication error.
We have made relative comparisons between the colonoscopy survey and
the acarbose trial rather than direct comparisons of labeling
percentages because biopsies from the colonoscopy survey were taken
from an empty colon, whereas acarbose trial biopsies were from an
unprepared colon. Although the agent used for bowel preparation was not
thought to affect proliferation (Fireman et al. 1989
),
the effects of colonic fermentation seen in the acarbose trial make it
reasonable to infer that biopsies taken from a colon empty of feces for
a period of time would differ from those taken with colonic contents
present. This colonic nutritional difference between studies may have
been responsible for the poor correlation of BrdU and Ki-67 with
placebo treatment in the acarbose trial compared with the neoplasia
subjects of the colonoscopy survey. This implies that removal of
nutrients for a limited time has a cell cycle synchronizing effect.
It is not surprising that butyrate, as a key colonocyte nutrient
(Roediger 1980
), stimulates mucosal proliferation.
Higher rates of colonic mucosal proliferation occur in the proximal
colon (Potten et al. 1992
) where the highest
concentrations of butyrate are found (Cummings et al. 1987
). The more concentrated colonic mucosal proliferation with
acarbose use has benign characteristics compared with our colonoscopy
survey results and agrees with the in vitro stimulatory effects of
butyrate on normal colonic mucosa (Bartram et al. 1993
,
Scheppach et al. 1992
) and with in vivo animal studies
(Lupton and Kurtz 1993
, Sakata 1987
,
Velázquez et al. 1996b
). These findings, together
with improved cell cycle synchrony associated with butyrate and
epidemiologic data relating the negative correlation of colon cancer
incidence by country with dietary starch intake by country
(Cassidy et al. 1994
), support a role for butyrate in
prevention of colon cancer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported by National Institutes of Health, National Cancer Institute grant CA56432, the Irving A. Hansen Memorial Foundation and Bayer Corporation.
Manuscript received May 30, 2000. Initial review completed June 21, 2000. Revision accepted August 9, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1.
Alberts D. S., Einspahr J., Rees-McGee S., Ramanujam P., Buller M. K., Clark L., Ritenbaugh C., Atwood J., Pethigal P., Earnest D., Villar H., Phelps J., Lipkin M., Wargovich M., Meyskens F. L., Jr Effects of dietary wheat bran fiber on rectal epithelial cell proliferation in patients with resection for colorectal cancers. J. Natl. Cancer Inst. 1990;82:1280-1285
2.
Bartram H. P., Scheppach W., Schmid H., Hofmann A., Dusel G., Richter F., Richter A., Kasper H. Proliferation of human colonic mucosa as an intermediate biomarker of carcinogenesis: effects of butyrate, deoxycholate, calcium, ammonia, and pH. Cancer Res 1993;53:3283-3288
3. Cassidy A., Bingham S. A., Cummings J. H. Starch intake and colorectal cancer risk: an international comparison. Br. J. Cancer 1994;69:937-942[Medline]
4.
Chiasson J. L., Josse R. G., Hunt J. A., Palmason C., Rodger N. W., Ross S. A., Ryan E. A., Tan M. H., Wolever T. M. The efficacy of acarbose in the treatment of patients with non-insulin-dependent diabetes mellitus. A multicenter controlled clinical trial. Ann. Intern. Med. 1994;121:928-935
5.
Clausen M. R., Bonnen H., Mortensen P. B. Colonic fermentation of dietary fibre to short chain fatty acids in patients with adenomatous polyps and colonic cancer. Gut 1991;32:923-928
6.
Cummings J. H., Pomare E. W., Branch W. J., Naylor C. P., MacFarlane G. T. Short chain fatty acids in human large intestine, portal hepatic and venous blood. Gut 1987;28:1221-1227
7. Deschner E., Lewis C. M., Lipkin M. In vitro study of human rectal epithelial cells. I. Atypical zone of H3 thymidine incorporation in mucosa of multiple polyposis. J. Clin. Investig. 1963;42:1922-1928
8. Deschner E. E., Lipkin M. Proliferative patterns in colonic mucosa in familial polyposis. Cancer 1975;35:413-418[Medline]
9. Fireman Z., Rozen P., Fine N., Chetrit A. Reproducibility studies and effects of bowel preparations on measurements of rectal epithelial proliferation. Cancer Lett 1989;45:59-64[Medline]
10. Gerdes J., Lemke H., Wacker H. H., Schwab J., Stein H. Cell cycle analysis of a cell proliferation associated human nuclear antigen defined by the monoclonal antibody Ki67. J. Immunol. 1984;133:1710-1715[Abstract]
11. Hague A., Singh B., Paraskeva C. Butyrate acts as a survival factor for colonic epithelial cells: further fuel for the in vivo versus in vitro debate. Gastroenterology 1997;112:1036-1040[Medline]
12. Hassig C. A., Tong J. K., Schreiber S. L. Fiber-derived butyrate and the prevention of colon cancer. Chem. Biol. 1997;4:783-789[Medline]
13. Higgins P. J., Chaudhari P., Ryan M. P. Cell-shape regulation and matrix protein p52 content in phenotypic variants of ras-transformed rat kidney fibroblasts. Functional analysis and biochemical comparison of p52 with proteins implicated in cell-shape determination. Biochem. J. 1991;273:651-658
14. Higgins P. J., Tanaka Y. Cytoarchitectural response and expression of c-fos/p52 genes during enhancement of butyrate-initiated differentiation of human colon carcinoma cells by 1,25-dihydroxyvitamin D3 and its analogs. Lipkin M. Newmark H. I. Kelloff G. eds. Calcium, vitamin D, and Prevention of Colon Cancer 1991:305-326 CRC Press Boca Raton, FL.
15.
Hiraishi K., Suzuki K., Hakomori S., Adachi M. LeY antigen expression is correlated with apoptosis (programmed cell death). Glycobiology 1993;3:381-390
16. Holt P. R., Atillasoy E., Lindenbaum J., Ho S. B., Lupton J. R., McMahon D., Moss S. F. Effects of acarbose on fecal nutrients, colonic pH, and short-chain fatty acids and rectal proliferation indices. Metabolism 1996;45:1179-1187[Medline]
17. Holt P. R., Higgins P. J., Atillasoy E., Davis P. J., Lipkin M. Abnormal cell proliferation and p52/p35-CSK expression in the colons of aging rats. Exp. Gerontol. 1995;30:495-503[Medline]
18. Jass J. R. Diet, butyric acid and differentiation of gastrointestinal tract tumors. Med. Hypotheses 1985;18:113-118[Medline]
19. Jass J. R., Ajioka Y., Radojkovic M., Allison L. J., Lane M. R. Failure to detect colonic mucosal hyperproliferation in mutation positive members of a family with hereditary non-polyposis colorectal cancer. Histopathology 1997;30:201-207[Medline]
20.
Kashtan H., Stern H. S., Jenkins D.J.A., Jenkins A. L., Thompson L. U., Hay K., Marcon N., Minkin S., Bruce W. R. Colonic fermentation and markers of colorectal-cancer risk. Am. J. Clin. Nutr. 1992;55:723-728
21.
Kelloff G. J., Boone C. W., Steele V. E., Crowell J. A., Lubet R., Sigman C. C. Progress in cancer chemoprevention: perspectives on agent selection and short-term clinical intervention trials. Cancer Res 1994;54:2015s-2024s
22. Kim Y. S., Taso D., Siddiqui B., Whitehead J. S., Arnstein P., Bennett J., Hicks J. Effects of sodium butyrate and dimethylsulfoxide on biochemical properties of human colon cancer cells. Cancer 1980;45:1185-1192[Medline]
23. Kohn G. J., Schwartz H. J., Ruebner B. H., Wong A. J., Lawson M. J. Colonic retinoblastoma protein and proliferation in cancer and non-cancer patients. J. Gastroenterol. Hepatol. 1997;12:198-203[Medline]
24. Kruh J. Effects of sodium butyrate, a new pharmacological agent, on cells in culture. Mol. Cell. Biochem. 1982;42:65-82[Medline]
25. Lin P., Allison D. C. Measurement of DNA content and of tritiated thymidine and bromodeoxyuridine incorporation by the same cells. J. Histochem. Cytochem. 1993;41:1435-1439[Abstract]
26. Lipkin M. Phase 1 and Phase 2 proliferative lesions of colonic epithelial cells in diseases leading to colonic cancer. Cancer 1974;34:878-888
27.
Lipkin M., Blattner W. A., Fraumeni J.F.J., Lynch H. T., Deschner E., Winawer S. Tritiated thymidine (
ph) labeling distribution as a marker for hereditary predisposition to colon cancer. Cancer Res 1983;43:1899-1904
28. Lipkin M., Enker W. E., Winawer S. J. Tritiated-thymidine labeling of rectal epithelial cells in "non-prep" biopsies of individuals at increased risk for colonic neoplasia. Cancer Lett 1987;37:153-162[Medline]
29. Lipkin M., Newmark H. Effect of added dietary calcium on colonic epithelial cell proliferation in subjects at high risk for familial colonic cancer. N. Engl. J. Med. 1985;313:1381-1384[Abstract]
30. Lupton J. R., Kurtz P. P. Relationship of colonic luminal short-chain fatty acids and pH to in vivo cell proliferation in rats. J. Nutr. 1993;123:1522-1530
31. Marra G., Anti M., Percesepe A., Armelao F., Ficarelli R., Coco C., Rinelli A., Vecchio F. M., D’Arcangelo E. Circadian variations of epithelial cell proliferation in human rectal crypts. Gastroenterology 1994;106:982-987[Medline]
32.
Morita A., Tsao D., Kim Y. S. Effect of sodium butyrate on alkaline phosphatase in HRT-18, a human rectal cancer cell line. Cancer Res 1982;42:4540-4545
33.
Nakamura S., Goto J., Kitayama Y., Sheffield J. P., Talbot I. C. Flow cytometric analysis of DNA synthetic phase fraction of the normal appearing colonic mucosa in patients with colorectal neoplasms. Gut 1995;37:398-401
34. Nsien E., Steinberg W. M., Wilkinson D. S., Rhame J. G., Henry J. P. Single-parameter DNA flow cytometric analysis of normal-appearing colonic mucosa does not predict the presence of colonic neoplasia. Am. J. Gastroenterol. 1991;86:1477-1481[Medline]
35. Paganelli G. M., Biasco G., Santucci R., Brandi G., Lalli A. A., Miglioli M., Barbara L. Rectal cell proliferation and colorectal cancer risk level in patients with nonfamilial adenomatous polyps of the large bowel. Cancer 1991;68:2451-2454[Medline]
36. Paganelli G. M., Higgins P. J., Biasco G., Lipkin M., Brandi G., Santucci R., Miglioli M., Luigi B. Abnormal rectal cell proliferation and p52/p35 protein expression in patients with ulcerative colitis. Cancer Lett 1993;73:23-28[Medline]
37. Paspatis G. A., Zizi A., Chlouverakis G. J., Giannikaki E. S., Vasilakaki T., Elemenoglou I., Karamanolis D. G. Proliferative patterns of rectal mucosa as predictors of advanced colonic neoplasms in routinely processed rectal biopsies. Am. J. Gastroenterol. 1998;93:1472-1477[Medline]
38.
Potten C. S., Kellett M., Rew D. A., Roberts S. A. Proliferation in human gastrointestinal epithelium using bromodeoxyuridine in vivo: data for different sites, proximity to a tumour, and polyposis coli. Gut 1992;33:524-529
39. Qin Y., Willems G. Comparison of the classical autoradiographic and the immunohistochemical methods with BrdU for measuring proliferation parameters in colon cancer. Anticancer Res 1993;13:731-735[Medline]
40.
Risio M., Lipkin M., Candelaresi G. L., Bertion A., Coverlizza S., Rossini F. P. Correlations between rectal mucosa cell proliferation and the clinical and pathological features of nonfamilial neoplasia of the large intestine. Cancer Res 1991;51:1917-1921
41.
Roediger W.E.W. Role of anaerobic bacteria in the metabolic welfare of the colonic mucosa in man. Gut 1980;21:793-798
42. Roediger W.E.W. Bacterial short-chain fatty acids and mucosal disease of the colon. Br. J. Surg. 1988;75:346-348[Medline]
43. Rozen P., Lubin F., Papo N., Zajicek G. Rectal epithelial proliferation in persons with or without a history of adenoma and its association with diet and lifestyle habits. Cancer 1998;83:1319-1327[Medline]
44. Sakata T. Stimulatory effect of short-chain fatty acids on epithelial cell proliferation in the rat intestine: a possible explanation for trophic effects of fermentable fibre, gut microbes and luminal trophic factors. Br. J. Nutr. 1987;58:95-103[Medline]
45.
Sander E. G., Warner R. G., Harrison H. N., Lossli J. K. The stimulatory effect of sodium butyrate and sodium propionate on the development of rumen mucosa in the young calf. J. Dairy Sci. 1959;42:1600-1605
46. Sasaki K., Murakami T., Kawasaki M., Takahashi M. The cell cycle associated change of the Ki-67 reactive nuclear antigen expression. J. Cell. Physiol. 1987;133:579-584[Medline]
47. Scheppach W., Bartram H. P., Richter F. Role of short-chain fatty acids in the prevention of colorectal cancer. Eur. J. Cancer 1995;31A:1077-1080
48.
Scheppach W., Bartram P., Richter A., Richter F., Liepold H., Dusel G., Hofstetter G., Ruthlein J., Kasper H. Effect of short-chain fatty acids on the human colonic mucosa in vitro. J. Parent. Enteral Nutr. 1992;16:43-48
49. Scheppach W., Fabian M., Sach M., Kasper H. J. The effect of starch malabsorption on fecal short chain acid excretion in man. Scand. J. Gastroenterol. 1988;23:755-759[Medline]
50. Shmakov A. N., Morey A. L., Ferguson D.J.P., Fleming K. A., O’Brien J. A., Savidge T. C. Conventional patterns of human intestinal proliferation in a severe-combined immunodeficient xenograft model. Differentiation 1995;59:321-330[Medline]
51. Terpstra O. T., van Blankenstein M., Dees J., Eilers G.A.M. Abnormal pattern of cell proliferation in the entire colonic mucosa of patients with colon adenoma or cancer. Gastroenterology 1987;92:704-708[Medline]
52.
Tsao D., Morita A., Bella A., Jr, Luu P., Kim Y. S. Differential effects of sodium butyrate, dimethyl sulfoxide, and retinoic acid on membrane-associated antigen, enzymes, and glycoproteins of human rectal adenocarcinoma cells. Cancer Res 1982;42:1052-1057
53. Velázquez O. C., Lederer H. M., Rombeau J. L. Butyrate and the colonocyte, implications for neoplasia. Dig. Dis. Sci. 1996a;41:727-739[Medline]
54.
Velázquez O. C., Zhou D., Seto R. W., Jabbar A., Choi J., Lederer H. M., Rombeau J. L. In vivo crypt surface hyperproliferation is decreased by butyrate and increased by deoxycholate in normal rat colon: associated in vivo effects on c-Fos and cJun expression. J. Parent. Enteral Nutr. 1996b;20:243-250
55.
Weaver G. A., Krause J. A., Miller T. L., Wolin M. J. Incidence of methanogenic bacteria in a sigmoidoscopy population: an association of methanogenic bacteria and diverticulosis. Gut 1986;27:698-704
56.
Weaver G. A., Krause J. A., Miller T. L., Wolin M. J. Short chain fatty acid distributions of enema samples from a sigmoidoscopy population: an association of high acetate and low butyrate ratios with adenomatous polyps and colon cancer. Gut 1988;29:1539-1543
57.
Weaver G. A., Krause J. A., Miller T. L., Wolin M. J. Corn starch fermentation by the colonic microbial community yields more butyrate than cabbage fiber fermentation: corn starch fermentation rates correlate negatively with methanogenesis. Am. J. Clin. Nutr. 1992;55:70-77
58.
Weaver G. A., Tangel C. T., Krause J. A., Parfitt M. M., Jenkins P. L., Rader J. M., Lewis B. A., Miller T. L., Wolin M. J. Acarbose enhances human colonic butyrate production. J. Nutr. 1997;127:717-723
59.
Whitehead R. H., Young G. P., Bhathal P. S. Effects of short chain fatty acids on a new human colon carcinoma cell line (LIM 1215). Gut 1986;27:1457-1463
60. Wong A. J., Kohn G. J., Schwartz H. J., Ruebner B. H., Lawson M. J. Colorectal cancer and noncancer patients have similar labeling indices by microscopy and computed image analysis. Hum. Pathol. 1995;26:1329-1332[Medline]
This article has been cited by other articles:
|
|
 |
|
  N. J. Emenaker, G. M. Calaf, D. Cox, M. D. Basson, and N. Qureshi Short-Chain Fatty Acids Inhibit Invasive Human Colon Cancer by Modulating uPA, TIMP-1, TIMP-2, Mutant p53, Bcl-2, Bax, p21 and PCNA Protein Expression in an In Vitro Cell Culture Model J. Nutr., November 1, 2001; 131(11): 3041S - 3046. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
