- Email: [email protected]
The effect of chenodeoxycholic acid on the development of aberrant crypt foci in the rat colon
CANCER LETTERS Cancer
Letters 76 (1994) lOl- ,107
The effect of chenodeoxycholic acid on the development of aberrant crypt foci in the rat colon Laurie A.M. Sutherland,
Ranjana
P. Bird*
Department of Foods and Nutrition, University of Manitoba, Winnipeg,
(Received
12 September
1993; revision
received
4 November
Manitoba
1993: accepted
R3T 2N2.
5 November
Canada
1993)
Abstract Bile acids are reported to enhance experimentally-induced colonic tumorigenesis. Previously we have reported that cholic acid, a known tumor promoter, actually reduced the number of aberrant crypt foci (ACF). purported preneoplastic lesions (B.A. Magnuson and R.P. Bird, Cancer Lett., 68 (1993), 15-23). This observation was unexpected and has prompted us to explore the effect of other bile acids on the development of ACF. The primary objective of this investigation was to evaluate the effect of feeding varying dosages of chenodeoxycholic acid (CDC) on the induction and growth of ACF and on the proliferative indices of the colonic epithelium. Sprague-Dawley male rats were injected with azoxymethane (+AOM, 20 mgikg) or saline (-AOM). One week later they were randomly allocated to five groups and were fed diets containing CDC at varying levels (0.0. 0.025, 0.05, 0. I and 0.2% by weight) for 2 weeks. After completion of the feeding period the number and crypt multiplicity of ACF were quantified, and three different proliferative indices, inciuding mitotic index, BUDR labelling index (percentage S-phase cells) and proliferating cell nuclear antigen labelling index (percentage cycling cells) were determined. CDC at all dosages increased the number of ACF having the maximum effect at the 0.1% CDC level. A significant dose-related increase in crypt height was noted in CDC-fed + AOM groups when compared with the +AOM control groups. The mitotic indices of colonic crypts were higher (P 5 0.05) only in the 0.025% CDC -AOM group when compared with the 0% CDC -AOM group (5.97 f 0.63 vs. 3.92 + 0.79). The BUDR labelling indices were not altered by CDC feeding (P 2 0.05). PCNA labelling indices increased consistently among the CDC-fed groups. Among the -AOM group the 0.05% CDC group had the maximum value, which was significantly different from the control value (19.21 + 1.92 vs. 10.93 f 0.56, respectively). Among the +AOM groups the PCNA labelling indices increased with increasing levels of CDC. It was concluded that CDC stimulated the development of ACF and altered cell cycle associated events in colonic crypts undergoing neoplastic changes. Key fjords: Aberrant
* Corresponding
crypt: Colon;
Bile acid;
Cell
proliferation;
author.
1.
Abbreviations: ACF, aberrant
crypt foci; AOM, azoxymethane: BUDR, bromodeoxyuridine; CDC. chenodeoxycholic acid; CH, crypt height; LI. labelling index: MI. mitotic index; PCNA, proliferating cell nuclear antigen: PZ, proliferative zone. 0304-3835/94/$06.00 0 1994 Elsevier SSDI 0304-3835(93)03228-W
Preneoplastic
Scientific
Publishers
Ireland
lesions
Introduction Previously
we have reported
that when cholic
acid, a primary bile acid, was fed to SpragueDawley rats the number of aberrant crypt foci, Ltd. All rights reserved
102
proposed pre-neoplastic lesions, was reduced in rat colons [ 141. This finding was unexpected, since it has been demonstrated that feeding cholic acid induces a marked proliferative response in the colonic epithelium [4,6] and enhances the incidence of chemically induced colon cancer [6,13]. These observations raise questions regarding the value of the number and growth of ACF as an early predictor of the disease outcome. These observations also raise questions regarding the mechanism(s) by which cholic acid as well as other bile acids modulate colon cancer development. Chenodeoxycholic acid (CDC), another primary bile acid, is also known to promote colon tumorigenesis [18,22]. The tumor-enhancing effect of CDC has been documented when CDC was administered intrarectally [18] or fed in the diet to animals [22]. In the present study it was of interest to investigate the effect of feeding CDC at varying levels (0, 0.025, 0.05, 0.10 and 0.20% by weight of diet) on the induction and growth of ACF and the proliferative indices in the rat colonic epithelium during a 2-week feeding trial. 2. Materials and methods 2. I. Animals Sixty male weanling Sprague-Dawley rats (Central Breeding Facility, University of Manitoba, Winnipeg) were used. The animals were housed in wire cages with sawdust bedding with a 12-h lightdark cycle and had free access to both food and water. All animals were cared for according to the guidelines of the Canadian Council on Animal Care. 2.2. Chemicals Bromodeoxyuridine (BUDR), colchicine, methylene blue, phosphate-buffered saline (PBS) and chenodeoxycholic acid (CDC) were purchased from Sigma Chemical Co., St. Louis, MO. The Signet immunoperoxidase kit was purchased from Signet Laboratories Inc., Dedham, MA. The PCNA monoclonal antibody was purchased from Dimensions Laboratories Inc., Missisauga, Ont. The anti-BUDR monoclonal antibody was obtained from Becton-Dickson Immunocytometry Systems, San Jose, CA.
L.A.M.
Sutherland, R.P. Bird/Cancer
Letr. 76 (1994) 101-107
2.3. Diets The basal diet was semi-synthetic AINdiet 0, 25, 50, 100 or 200 mg [1,2] which contained CDC1100 g of diet (0, 0.025, 0.05, 0.10 and 0.20% CDC). 2.4. Study design Sixty rats were randomly divided into two groups (30 rats/group). One group of rats was injected with a single dose of AOM (20 mg/kg) (+AOM), and the other thirty served as a vehicle control and were injected with saline (-AOM) One week after the injection the rats were randomly allocated to one of five experimental groups. These groups were fed one of the diets containing 0.00. 0.025,0.05,0.10 or 0.20% CDC. Twenty-four hour food intake records were recorded for three separate periods. Body weights were recorded weekly. Twenty-one days after initial injection time the animals were killed by CO2 asphyxiation. Twoand-a-half hours prior to termination all animals received an injection of colchicine (1 mgikg), and 1 h prior to termination all animals received an injection of bromodeoxyuridine (BUDR) (50 mgkg). After termination the colons were removed, flushed with phosphate-buffered saline (PBS), slit open from the caecum to the anus and fixed flat between filter paper in 70% ethanol. 2.5. Quantijkation of aberrant crypt foci After a minimum of 24 h in 70% ethanol, the colons were placed in a 0.2% methylene blue solution dissolved in PBS for approximately 5- 10 min, then placed mucosal side up on a microscope slide and viewed with a light microscope. Criteria used to define and quantify ACF have been described previously [3]. 2.6. Assessment of proliferative indices A 2-cm2 piece of tissue from each animal was embedded in paraffin and processed for histology. Haematoxylin and eosin sections were evaluated for the presence of metaphase cells in complete longitudinal sections of colonic crypts as described previously [4]. Assessment of the BUDR and PCNA-labelled cells was carried out by immunohistochemical techniques [ 191 employing a Signet immunoperoxidase staining kit utilized according
L.A.M.
Sutherland, R. P. Bird/ Cancer Lett. 76 (1994)
iOl-107
to the manufacturer’s instructions. The antiBUDR monoclonal antibody and the PCNA monoclonal antibody were diluted 1:40. 2.7. Statistical analysis Statistical analysis of the data was performed by analysis of variance and by the new Duncan’s multiple range test. A P value 5 0.05 was considered significant. 3. Results LEVEL
Throughout the study, no significant differences were found in body weights or food consumption of either the AOM-injected groups or the noninjected groups fed varying levels of CDC. The mean aberrant crypt foci (ACF) in the AOM-injected groups are shown in Fig. 1. There was a significant increase in the number of ACF in the 0.05,O. 1 and 0.2% CDC groups compared with the control group (0% CDC). In terms of crypt multiplicity, ACF with one and two crypts per focus were significantly higher in the 0.1 and 0.2% CDC groups when compared with the control group (Table 1). For all of the groups the majority of ACF were found in the 3-l l-cm regions of the fixed colon (Fig. 2). Among the groups receiving higher levels of CDC (0.05, 0.10 and 0.20%) the majority of ACF occurred closer to the rectal end (3-7 cm).
Table I Distribution
of aberrant
crypt Number
foci with varying
crypt
OF CDC
(X)
Fig. I. Effect of CDC feeding in mean total number of ACF/colon/group. An asterisk symbolizes those values that are significantly different from the control value (P 5 0.05).
The percentage of ACF consisting of 1 to 5 crypt(s)/foci was similar in all groups. Generally, CDC at all levels increased the mitotic indices in the -AOM group, with the exception of the 0.2% CDC level, but only the 0.025% CDC level significantly increased the mitotic index when compared with the control (Table 2). Among the +AOM groups, CDC at all levels did not increase the mitotic indices but tended to increase the size of the proliferative zones in an inconsistent manner. For crypt height (CH) there was a dose-related increase in CH, with a significant increase occurring in the +AOM group
multiplicity
of crypts/focus
‘X CDC
ACF-I
ACF-2
ACF-3
ACF-4
ACF-5
0.00
r&It *Il.7 81.7Tf ~8.2 lO8.5t* f 14.8 114.0* f 20.0 I l2.3* f 13.7
36.8t *4.9 50.2 t* l 9.l 62.2t* f 11.6 68.8* iz5.6 64x* zt9.1
6.3 *0.8 IO.2 l I.7 12.0 zt2.6 I I.2 zt2.5 12.3 f I.8
I.0 *0.5 1.8 l 0.7 2.5 l I.1 I.8 +0.6 0.7 *0.5
0.0 *o.o 0.3 l 0.3 0.2 ZtO.2 0.0 *o.o 0.3 *0.2
0.025 0.05 0.10 0.20
ACF-1 to ACF-5 represent with different superscripts
ACF with one to five crypts. Values are mean f S.E.M.. n = 6 animals/group. are significantly different (P 5 0.05).
Means within each column
L.A.M.
2
30
2
25
2
20
2
15
-101
’
’
’
’
’
’
’
’
’
’
0
2
4
6
6
10
12
14
16
18
20
csntimeter
of colon
CDC on crypt
heightl,
Treatment
group
size of proliferative
zoneb and mitotic
-AOM CDC Level (%)
0.00 0.025 0.05 0.10 0.20
Crypt
R.P. Bird/Cancer
Letr. 76 (1994)
101-107
only at the 0.1 and 0.2% CDC levels. There was no consistent effect of the level of CDC on crypt height in the -AOM groups. This dose-related increase in crypt height in the injected groups and the inconsistent increase in CH in the non-injected group were found to be similar when these parameters were determined while quantifying BUDR and PCNA-labelled cells. There was no significant effect of feeding CDC on the BUDR labelling indices or size of the proliferative zone in the +AOM or -AOM groups (Table 3). The size of the proliferative zone containing PCNA-labelled cells was significantly higher for the 0.05 and 0.1% groups in both the -AOM and +AOM groups when compared with the corresponding control groups (Table 4). The CDC feeding increased the PCNA labelling indices in the -AOM as well as in the +AOM groups. However, this effect was more pronounced among the +AOM group, which had significantly higher PCNA labelling indices at all levels of CDC, with a maximum value at the 0.05% CDC level when compared with the 0% CDC control group.
Fig. 2. Distribution of ACF along the length of the colon of rats fed varying levels of CDC; (0) control, (0) 0.025%, (V) 0.05%, (r) 0.10% and (0) 0.20%.
Table 2 Effect of feeding
Sutherland,
activityC of rat colonic
cryptsd.
+AOM Prolif. zone
Mitotic
Prolif. zone
Mitotic
(CW
height
(PZ)
(MI)
index
(CW
Crypt
height
(PZ)
(MI)
29.l* *I.1 25.4t ho.8 26.7t* l 0.9 29.2* Zto.9 28.3* Zto.7
8.7t* f 1.6 10.4* Zto.9 9.5*t kO.6 10.4’ Zko.7 6.6t Zt1.0
3.9V ~0.8 6.0’ ~0.6 4.97; +0.6 5.3t* l 0.5 2.8$ Zto.4
25.9t *I.3 27.6*t *I.2 27.8t* *0.4 29.6* Zto.5 29.5* ho.6
7.1t f 1.2 9.2t* l 0.9 10.6t* ~0.8 7.3t f 1.3 12.1* l 1.2
3.4 Zto.5 3.8 *0.4 4.0 *0.5 3.2 Yko.5 4.4 *0.5
index
The effect of CDC on each parameter for -AOM and +AOM groups was assessed separately. Means within each column with different superscripts are significantly different (P 5 0.05). ‘Number of cells per mid-axial crypt. bProliferative zone is assessed by enumerating the highest location of the metaphase cells in each crypt, then averaging those values to determine the size of the proliferative zone/group. ‘Number of metaphase cells/l00 cells/group. dValues are mean + S.E.M., n = 6 animals/group.
L.A.M.
Sutherland,
R. P. Bird/ Cancer Lett.
Table 3 Effect of feeding chenodeoxycholic
acid on size of proliferative
Treatment
105
76 (‘1994) 101-107
zone” and BUDR
labelling
indicesb
of rat colonic
crypts’.
group
-AOM
+AOM
CDC level (%)
Prolif.
Label. index (LI)
Prolif. zone (PZ)
Label. index (LI)
0.00
5.1 ho.8
3.6 *0.8
0.025
4.7 f 1.0 2.9 *0.6
3.1 ho.6 2.8 *0.5
4.1 f 1.3 4.0 f 1.2
3.5 *I..5 2.2 *0.7
8.9* l 1.4 4.1t + 1.8 10.6* *0.9 4.8t +0.8 10.7* f I.1
4.6 *to.3 3.3 f 1.2 3.9 *0.2 2.6 *0.4 4.5 *to.8
0.05 0.10 0.20
zone (PZ)
The effect of CDC on each parameter for -AOM and +AOM groups was assessed separately. ferent superscripts are signiticantly different (P 5 0.05). “The proliferative zone is assessed by enumerating the highest location of the BUDR-labelled those values to determine the size of the proliferative zone/group. bNumber of BUDR-labelled cells/l00 cells/group. ‘Values are mean f S.E.M., n = 6 animals/group.
Table 4 Effect of feeding chenodeoxycholic Treatment
acid on size of proliferative
zone a and PCNA
labelling
Means within each column cells in each crypt,
indices’
of rat colonic
with dif-
then averaging
crypts’
group
+AOM
-AOM
CDC level (‘X))
Prolif. zone (PZ)
Label. index (LI)
Prolif. zone (PZ)
Label. index (LI)
0.00
10.6t
I0.9T kO.6 16.8t* *2.0 19.2* f 1.9 i5.8?* l 3.1 13.5t* +0.6
13.8t
Il.5$ l 1.0 l7.91_ f 1.4 26.6* l 2.5 23.2t* *2.2 24.1t* *2.7
0.025 0.05 0.10 0.20
10.6 13.5*t f 1.3 16.1* f 1.5 16.2* f 1.6 13.0*t kO.8
*0.7 17.3’ hO.9 16.6* l 0.7 I7.4f *0.7 17.3* * 1.4
The effect of CDC on each parameter for -AOM and +AOM groups was assessed separately. ferent superscripts are significantly different (P 5 0.05). “The proliferative zone is assessed by enumerating the highest location of the PCNA-labelled those values to determine the size of the proliferative zone/colon. bNumber of PCNA-labelled cells/l00 cells/group. ‘Values are mean + S.E.M.. n = 6 animals/group.
Means within each column ceils in each crypt.
with dif-
then averaging
106
4. Discussion In the present study, dietary feeding of CDC resulted in a dose-related increase in the number of ACF in the rat colon. CDC feeding did not increase mitotic and BUDR labelling indices in carcinogen-treated animals; however, an increase in mitotic indices was observed in the normal colonic epithelium. Analysis of the number and distribution of cells exhibiting PCNA revealed that significant increases occurred in CH, PZ and LI, with higher levels of CDC (0.05, 0.1 and 0.2) in both the injected (+AOM) and non-injected (-AOM) colons when compared with the 0% CDC group. This effect was more pronounced in the +AOM group. Aberrant crypt foci have been identified in rodent [3] and human colons [16,21] and are proposed to be pre-neoplastic lesions [3]. The ACF system has been used extensively by several other investigators to screen and identify potential modulators of colon carcinogenesis [ 15,241. It has been demonstrated that the genotypic and phenotypic features of ACF are those to be expected in preneoplastic lesions [12]. The ability of CDC to increase the number of ACF in the rat colon is in agreement with the tumor-enhancing effect of CDC [7,8,22]. CDC is one of the primary bile acids derived from cholesterol metabolism and is currently used clinically to dissolve gallstones [5]. Chenodeoxycholic acid is thought to exert its effect through its metabolism into lithocholic acid by colonic microflora. Lithocholic acid has been reported to be a colonic tumor promoter and also augments the genotoxicity of known carcinogens 1171. The exact mechanism(s) by which CDC or lithocholic acid may enhance the number of ACF or tumor incidence is unknown. Measurements of the proliferative indices of colonic epithelium have been used frequently in animal studies to assess risk for cancer development [ 10,191. Several approaches have been taken in this regard, including the quantification of mitotic and/or BUDR labelling indices (as a measure of Sphase cells) and the quantification of cells exhibiting proliferating cell nuclear antigen (PCNA) [ 191. It has been reported that there is general correspondence between BUDR and PCNA labelling indices in normal-appearing crypts, the latter In the present study being slightly higher [ 191.
L.A.M.
Sutherland, R.P. Bird/Cancer
Leu. 76 (1994)
101-107
the proliferative indices of colonic epithelia of animals treated with a carcinogen (+AOM) or saline (-AOM) and fed CDC were measured by all three different methods. Feeding CDC resulted in an increase in the crypt height, suggesting that this bile acid did exert a hyperplastic response. Interestingly, this observation only occurred in the +AOM group, suggesting that the carcinogentreated colons responded differently to CDC feeding. Our observation that 0.2% CDC seemed to decrease the mitotic index in the colonic crypts supports the observation by Simanowski and coworkers [23], who also demonstrated that a 1% CDC treatment reduced the proliferative response in the rat colon. These effects are incongruous to the proposed role for cell proliferation in mediating the biological effects of bile acid [9]. The mechanisms by which CDC feeding increased the number of PCNA-labelled cells and the significance of these changes in the development of ACF and colon cancer remain to be evaluated. Proliferating cell nuclear antigens are nuclear proteins functioning as auxiliary proteins of DNA polymerase [20]. These proteins may be regulated differently in normal and neoplastic tissue [20]. Furthermore, increased expression of PCNA has been associated with DNA excision repair synthesis [ll]. It is noteworthy that the effects of CDC on the induction of ACF and proliferative indices of the colon are quite different from those reported for cholic acid [ 141. It has been reported previously that cholic acid is highly mitogenic and reduces the number of ACF, while it is capable of enhancing tumor incidence. In the present study CDC was able to increase the number of ACF without exerting a distinct mitogenic response. These findings allude to the fact that the ability of various bile acids to enhance tumor growth may not necessarily be mediated via increased cell proliferation. It is proposed that bile acids may serve as an important group of compounds with structural and biological heterogenicity which may reveal cellular and biochemical events important in the initiation and post-initiation stages of colon carcinogenesis. 5. Acknowledgments This study was supported by the Natural Sciences and Engineering Research Council of
L.A.M.
Sutherland, R.P. Bird/Cancer
Lett. 76 (1994)
101-107
Canada (NSERC) and the National Cancer Institute of Canada. L. Sutherland is the recipient of the undergraduate NSERC award. 6. References 1
2
3
4
5
6
7
8
American Institute of Nutrition (1977) Report of the AIN ad hoc committee on standards for nutrition studies. J. Nutr., 107, 1340-1348. American Institute of Nutrition (1980) Second report of the ad hoc committee on standards for nutrition studies. J. Nutr., 110, 1726. Bird, R.P. (1987) Observation and quantification of aberrant crypts in the murine colon treated with a colon carcinogen: preliminary findings. Cancer Lett., 37, 147-151. Bird, R.P., Schneider, R., Stamp, D. and Bruce, W.R. (1986) Effect of dietary calcium and cholic acid on the proliferative indices of murine colonic epithelium. Carcinogenesis, 7, 1657-1661. Cohen, B.1. and Raicht, R.F. (1981) Effects of bile acids on colon carcinogenesis in rats treated with a carcinogen. Cancer Res., 41, 3759-3760. Cohen, B.I., Raicht, R.F., Deschner. E.E. Takahashi, M., Sarwal, A.N. and Fazzini, E. (1980) Effect of cholic acid feeding in N-methyl-N-nitrosourea induced colon tumors and cell kinetics in rats. J. Natl. Cancer Inst., 64, 573-578. Cyzan, P., Seitz, H.K., Waldherr, R., Stiehl, A., Raedsch, R. and Kommerell, B. (1982) Chenodeoxycholic acid but not ursodeoxycholic acid enhances colon carcinogenesis in the rat. In: Bile Acids and Cholesterol in Health and Disease, pp. 343-45. Editors: G. Paumgartner, A. Stiehl, G. Gerok. MTP Press, Lancaster, PA. Danzinger, R.G., Hofmann, A.F., Schoenfteld, L.J. and Thistle, J.L. (1972) Dissolution of cholesterol gallstones by chenodeoxycholic acid. N. Engl. J. Med., 286, l-8. Deschner, E.E. and Raicht, R.F. (1979) Influence of bile in kinetic behavior of colonic epithelial cells of the rat. Digestion, 19, 322-327. Deschner, E.E., Long, F.C., Hakissian, M. and Herrmann, S.L. (1983) Differential susceptibility of AKR, C57BL/6 and CFI mice to 1,2-dimethylhydrazine-induced colonic tumor formation predicted by proliferative characteristics of colonic epithelial cells. J. Natl. Cancer Inst., 70, 279-282. Hall, P.A. and Woods, A.L. (1990) Immunohistochemical markers of cellular prolifertion: achievements, problems and prospects [invited review]. Cell Tissue Kinet., 23, 505-522. McLellan, E.A., Medline, A. and Bird, R.P. (1991) Sequential analysis of the growth and morphological characteristics of aberrant crypt foci: putative preneoplastic lesions. Cancer Res., 51, 5270-5274.
107
13
14
15
16
17
18
19
20
21
22
23
24
McSherry, Mosbach,
C.K., Cohen, B.I., Bokkenheuser. V.D.. E.H., Winter, J., Matoba, N. and Scholes, J.
(1989) Effects of calcium and bile acid feeding on colon tumors in the rat. Cancer Res., 49, 6039-6043. Magnuson, B.A. and Bird, R.P. (1993) Reduction of aberrant crypt foci induced in rat colon with azoxymethane or methylnitrosurea by feeding cholic acid. Cancer Lett.. 68. 15-23. Pereira, M.A. and Khory. M.D. (1991) Prevention by chemopreventive agents of azoxymethane-induced foci of aberrant crypts in rat colon. Cancer Lett.. 61, 27-33. Pretlow, T.P., Barrow, B.J., Ashton, W.S., O’Riordan. M.A., Pretlow, T.G., Jurcisek, J.A. and Stellato, T.A. (1991) Aberrant crypts: putative preneoplastic foci in human colonic mucosa. Cancer Res., 51, 1564- 1567. Reddy, B.S. and Wantanabe, K. (1979) Effect of cholesterol metabolites and promoting effect of lithocholic acid in colon carcinogenesis in germ-free and conventional F344 rats. Cancer Res., 39, 1521-1524. Reddy, B.S., Wantanabe, K., Weisburger, J.R. and Wynder, E.L. (1977) Promoting effect of bile acids in colon carcinogenesis in germ-free and conventional F344 rats. Cancer Res., 37, 3238-3242. Richter, F., Richter, A.. Yang, K. and Lipkin, M. (1992) Cell proliferation in rat colon measured with bromodeoxyuridine, proliferating cell nuclear antigen and [3H]thymidine. Cancer Epidemiol. Biomarkers Prevent.. 1, 561566. Risio, M., Candelaresi, G. and Rossini. F.P. (1993) Bromodeoxyuridine uptake and proliferating cell nuclear antigen expression throughout the colorectal tumor sequence. Cancer Epidemiol. Biomarkers Prevent., 2, 363-367. Roncucci, L., Stamp, D., Medline, A., Cullen, J.B. and Bruce, W.R. (1991) Identification and quantification of aberrant crypt foci and microadenomas in the human colon. Hum. Pathol., 22, 287-294. Sarwal, A.N., Cohen, B.I., Raicht, R.F., Takahashi, M. and Fazzini, E. (1979) Effects of dietary administration of chenodeoxycholic acid on N-methyl-N-nitrosoureainduced colon cancer in rats. Biochim. Biophys. Acta. 574, 423-432. Simanowski, U.A., Seitz, H.K., Cyzan, P., Horner, M., Waldherr, R., Weber, E. and Kommerell, B. (1987) Chronic ursodeoxycholic acid and chenodeoxycholic acid-feeding-induced changes of colon mucosal cell proliferation in rats. J. Natl. Cancer Inst., 79, 163-166. Steel, V.E. and Kelloff, G.J. (1993). Evaluation of a rat colon crypt assay to identify chemopreventive agents compared to rodent tumor assay systems. Proc. Am. Assoc. Cancer Res., 34, 553.
Letters 76 (1994) lOl- ,107
The effect of chenodeoxycholic acid on the development of aberrant crypt foci in the rat colon Laurie A.M. Sutherland,
Ranjana
P. Bird*
Department of Foods and Nutrition, University of Manitoba, Winnipeg,
(Received
12 September
1993; revision
received
4 November
Manitoba
1993: accepted
R3T 2N2.
5 November
Canada
1993)
Abstract Bile acids are reported to enhance experimentally-induced colonic tumorigenesis. Previously we have reported that cholic acid, a known tumor promoter, actually reduced the number of aberrant crypt foci (ACF). purported preneoplastic lesions (B.A. Magnuson and R.P. Bird, Cancer Lett., 68 (1993), 15-23). This observation was unexpected and has prompted us to explore the effect of other bile acids on the development of ACF. The primary objective of this investigation was to evaluate the effect of feeding varying dosages of chenodeoxycholic acid (CDC) on the induction and growth of ACF and on the proliferative indices of the colonic epithelium. Sprague-Dawley male rats were injected with azoxymethane (+AOM, 20 mgikg) or saline (-AOM). One week later they were randomly allocated to five groups and were fed diets containing CDC at varying levels (0.0. 0.025, 0.05, 0. I and 0.2% by weight) for 2 weeks. After completion of the feeding period the number and crypt multiplicity of ACF were quantified, and three different proliferative indices, inciuding mitotic index, BUDR labelling index (percentage S-phase cells) and proliferating cell nuclear antigen labelling index (percentage cycling cells) were determined. CDC at all dosages increased the number of ACF having the maximum effect at the 0.1% CDC level. A significant dose-related increase in crypt height was noted in CDC-fed + AOM groups when compared with the +AOM control groups. The mitotic indices of colonic crypts were higher (P 5 0.05) only in the 0.025% CDC -AOM group when compared with the 0% CDC -AOM group (5.97 f 0.63 vs. 3.92 + 0.79). The BUDR labelling indices were not altered by CDC feeding (P 2 0.05). PCNA labelling indices increased consistently among the CDC-fed groups. Among the -AOM group the 0.05% CDC group had the maximum value, which was significantly different from the control value (19.21 + 1.92 vs. 10.93 f 0.56, respectively). Among the +AOM groups the PCNA labelling indices increased with increasing levels of CDC. It was concluded that CDC stimulated the development of ACF and altered cell cycle associated events in colonic crypts undergoing neoplastic changes. Key fjords: Aberrant
* Corresponding
crypt: Colon;
Bile acid;
Cell
proliferation;
author.
1.
Abbreviations: ACF, aberrant
crypt foci; AOM, azoxymethane: BUDR, bromodeoxyuridine; CDC. chenodeoxycholic acid; CH, crypt height; LI. labelling index: MI. mitotic index; PCNA, proliferating cell nuclear antigen: PZ, proliferative zone. 0304-3835/94/$06.00 0 1994 Elsevier SSDI 0304-3835(93)03228-W
Preneoplastic
Scientific
Publishers
Ireland
lesions
Introduction Previously
we have reported
that when cholic
acid, a primary bile acid, was fed to SpragueDawley rats the number of aberrant crypt foci, Ltd. All rights reserved
102
proposed pre-neoplastic lesions, was reduced in rat colons [ 141. This finding was unexpected, since it has been demonstrated that feeding cholic acid induces a marked proliferative response in the colonic epithelium [4,6] and enhances the incidence of chemically induced colon cancer [6,13]. These observations raise questions regarding the value of the number and growth of ACF as an early predictor of the disease outcome. These observations also raise questions regarding the mechanism(s) by which cholic acid as well as other bile acids modulate colon cancer development. Chenodeoxycholic acid (CDC), another primary bile acid, is also known to promote colon tumorigenesis [18,22]. The tumor-enhancing effect of CDC has been documented when CDC was administered intrarectally [18] or fed in the diet to animals [22]. In the present study it was of interest to investigate the effect of feeding CDC at varying levels (0, 0.025, 0.05, 0.10 and 0.20% by weight of diet) on the induction and growth of ACF and the proliferative indices in the rat colonic epithelium during a 2-week feeding trial. 2. Materials and methods 2. I. Animals Sixty male weanling Sprague-Dawley rats (Central Breeding Facility, University of Manitoba, Winnipeg) were used. The animals were housed in wire cages with sawdust bedding with a 12-h lightdark cycle and had free access to both food and water. All animals were cared for according to the guidelines of the Canadian Council on Animal Care. 2.2. Chemicals Bromodeoxyuridine (BUDR), colchicine, methylene blue, phosphate-buffered saline (PBS) and chenodeoxycholic acid (CDC) were purchased from Sigma Chemical Co., St. Louis, MO. The Signet immunoperoxidase kit was purchased from Signet Laboratories Inc., Dedham, MA. The PCNA monoclonal antibody was purchased from Dimensions Laboratories Inc., Missisauga, Ont. The anti-BUDR monoclonal antibody was obtained from Becton-Dickson Immunocytometry Systems, San Jose, CA.
L.A.M.
Sutherland, R.P. Bird/Cancer
Letr. 76 (1994) 101-107
2.3. Diets The basal diet was semi-synthetic AINdiet 0, 25, 50, 100 or 200 mg [1,2] which contained CDC1100 g of diet (0, 0.025, 0.05, 0.10 and 0.20% CDC). 2.4. Study design Sixty rats were randomly divided into two groups (30 rats/group). One group of rats was injected with a single dose of AOM (20 mg/kg) (+AOM), and the other thirty served as a vehicle control and were injected with saline (-AOM) One week after the injection the rats were randomly allocated to one of five experimental groups. These groups were fed one of the diets containing 0.00. 0.025,0.05,0.10 or 0.20% CDC. Twenty-four hour food intake records were recorded for three separate periods. Body weights were recorded weekly. Twenty-one days after initial injection time the animals were killed by CO2 asphyxiation. Twoand-a-half hours prior to termination all animals received an injection of colchicine (1 mgikg), and 1 h prior to termination all animals received an injection of bromodeoxyuridine (BUDR) (50 mgkg). After termination the colons were removed, flushed with phosphate-buffered saline (PBS), slit open from the caecum to the anus and fixed flat between filter paper in 70% ethanol. 2.5. Quantijkation of aberrant crypt foci After a minimum of 24 h in 70% ethanol, the colons were placed in a 0.2% methylene blue solution dissolved in PBS for approximately 5- 10 min, then placed mucosal side up on a microscope slide and viewed with a light microscope. Criteria used to define and quantify ACF have been described previously [3]. 2.6. Assessment of proliferative indices A 2-cm2 piece of tissue from each animal was embedded in paraffin and processed for histology. Haematoxylin and eosin sections were evaluated for the presence of metaphase cells in complete longitudinal sections of colonic crypts as described previously [4]. Assessment of the BUDR and PCNA-labelled cells was carried out by immunohistochemical techniques [ 191 employing a Signet immunoperoxidase staining kit utilized according
L.A.M.
Sutherland, R. P. Bird/ Cancer Lett. 76 (1994)
iOl-107
to the manufacturer’s instructions. The antiBUDR monoclonal antibody and the PCNA monoclonal antibody were diluted 1:40. 2.7. Statistical analysis Statistical analysis of the data was performed by analysis of variance and by the new Duncan’s multiple range test. A P value 5 0.05 was considered significant. 3. Results LEVEL
Throughout the study, no significant differences were found in body weights or food consumption of either the AOM-injected groups or the noninjected groups fed varying levels of CDC. The mean aberrant crypt foci (ACF) in the AOM-injected groups are shown in Fig. 1. There was a significant increase in the number of ACF in the 0.05,O. 1 and 0.2% CDC groups compared with the control group (0% CDC). In terms of crypt multiplicity, ACF with one and two crypts per focus were significantly higher in the 0.1 and 0.2% CDC groups when compared with the control group (Table 1). For all of the groups the majority of ACF were found in the 3-l l-cm regions of the fixed colon (Fig. 2). Among the groups receiving higher levels of CDC (0.05, 0.10 and 0.20%) the majority of ACF occurred closer to the rectal end (3-7 cm).
Table I Distribution
of aberrant
crypt Number
foci with varying
crypt
OF CDC
(X)
Fig. I. Effect of CDC feeding in mean total number of ACF/colon/group. An asterisk symbolizes those values that are significantly different from the control value (P 5 0.05).
The percentage of ACF consisting of 1 to 5 crypt(s)/foci was similar in all groups. Generally, CDC at all levels increased the mitotic indices in the -AOM group, with the exception of the 0.2% CDC level, but only the 0.025% CDC level significantly increased the mitotic index when compared with the control (Table 2). Among the +AOM groups, CDC at all levels did not increase the mitotic indices but tended to increase the size of the proliferative zones in an inconsistent manner. For crypt height (CH) there was a dose-related increase in CH, with a significant increase occurring in the +AOM group
multiplicity
of crypts/focus
‘X CDC
ACF-I
ACF-2
ACF-3
ACF-4
ACF-5
0.00
r&It *Il.7 81.7Tf ~8.2 lO8.5t* f 14.8 114.0* f 20.0 I l2.3* f 13.7
36.8t *4.9 50.2 t* l 9.l 62.2t* f 11.6 68.8* iz5.6 64x* zt9.1
6.3 *0.8 IO.2 l I.7 12.0 zt2.6 I I.2 zt2.5 12.3 f I.8
I.0 *0.5 1.8 l 0.7 2.5 l I.1 I.8 +0.6 0.7 *0.5
0.0 *o.o 0.3 l 0.3 0.2 ZtO.2 0.0 *o.o 0.3 *0.2
0.025 0.05 0.10 0.20
ACF-1 to ACF-5 represent with different superscripts
ACF with one to five crypts. Values are mean f S.E.M.. n = 6 animals/group. are significantly different (P 5 0.05).
Means within each column
L.A.M.
2
30
2
25
2
20
2
15
-101
’
’
’
’
’
’
’
’
’
’
0
2
4
6
6
10
12
14
16
18
20
csntimeter
of colon
CDC on crypt
heightl,
Treatment
group
size of proliferative
zoneb and mitotic
-AOM CDC Level (%)
0.00 0.025 0.05 0.10 0.20
Crypt
R.P. Bird/Cancer
Letr. 76 (1994)
101-107
only at the 0.1 and 0.2% CDC levels. There was no consistent effect of the level of CDC on crypt height in the -AOM groups. This dose-related increase in crypt height in the injected groups and the inconsistent increase in CH in the non-injected group were found to be similar when these parameters were determined while quantifying BUDR and PCNA-labelled cells. There was no significant effect of feeding CDC on the BUDR labelling indices or size of the proliferative zone in the +AOM or -AOM groups (Table 3). The size of the proliferative zone containing PCNA-labelled cells was significantly higher for the 0.05 and 0.1% groups in both the -AOM and +AOM groups when compared with the corresponding control groups (Table 4). The CDC feeding increased the PCNA labelling indices in the -AOM as well as in the +AOM groups. However, this effect was more pronounced among the +AOM group, which had significantly higher PCNA labelling indices at all levels of CDC, with a maximum value at the 0.05% CDC level when compared with the 0% CDC control group.
Fig. 2. Distribution of ACF along the length of the colon of rats fed varying levels of CDC; (0) control, (0) 0.025%, (V) 0.05%, (r) 0.10% and (0) 0.20%.
Table 2 Effect of feeding
Sutherland,
activityC of rat colonic
cryptsd.
+AOM Prolif. zone
Mitotic
Prolif. zone
Mitotic
(CW
height
(PZ)
(MI)
index
(CW
Crypt
height
(PZ)
(MI)
29.l* *I.1 25.4t ho.8 26.7t* l 0.9 29.2* Zto.9 28.3* Zto.7
8.7t* f 1.6 10.4* Zto.9 9.5*t kO.6 10.4’ Zko.7 6.6t Zt1.0
3.9V ~0.8 6.0’ ~0.6 4.97; +0.6 5.3t* l 0.5 2.8$ Zto.4
25.9t *I.3 27.6*t *I.2 27.8t* *0.4 29.6* Zto.5 29.5* ho.6
7.1t f 1.2 9.2t* l 0.9 10.6t* ~0.8 7.3t f 1.3 12.1* l 1.2
3.4 Zto.5 3.8 *0.4 4.0 *0.5 3.2 Yko.5 4.4 *0.5
index
The effect of CDC on each parameter for -AOM and +AOM groups was assessed separately. Means within each column with different superscripts are significantly different (P 5 0.05). ‘Number of cells per mid-axial crypt. bProliferative zone is assessed by enumerating the highest location of the metaphase cells in each crypt, then averaging those values to determine the size of the proliferative zone/group. ‘Number of metaphase cells/l00 cells/group. dValues are mean + S.E.M., n = 6 animals/group.
L.A.M.
Sutherland,
R. P. Bird/ Cancer Lett.
Table 3 Effect of feeding chenodeoxycholic
acid on size of proliferative
Treatment
105
76 (‘1994) 101-107
zone” and BUDR
labelling
indicesb
of rat colonic
crypts’.
group
-AOM
+AOM
CDC level (%)
Prolif.
Label. index (LI)
Prolif. zone (PZ)
Label. index (LI)
0.00
5.1 ho.8
3.6 *0.8
0.025
4.7 f 1.0 2.9 *0.6
3.1 ho.6 2.8 *0.5
4.1 f 1.3 4.0 f 1.2
3.5 *I..5 2.2 *0.7
8.9* l 1.4 4.1t + 1.8 10.6* *0.9 4.8t +0.8 10.7* f I.1
4.6 *to.3 3.3 f 1.2 3.9 *0.2 2.6 *0.4 4.5 *to.8
0.05 0.10 0.20
zone (PZ)
The effect of CDC on each parameter for -AOM and +AOM groups was assessed separately. ferent superscripts are signiticantly different (P 5 0.05). “The proliferative zone is assessed by enumerating the highest location of the BUDR-labelled those values to determine the size of the proliferative zone/group. bNumber of BUDR-labelled cells/l00 cells/group. ‘Values are mean f S.E.M., n = 6 animals/group.
Table 4 Effect of feeding chenodeoxycholic Treatment
acid on size of proliferative
zone a and PCNA
labelling
Means within each column cells in each crypt,
indices’
of rat colonic
with dif-
then averaging
crypts’
group
+AOM
-AOM
CDC level (‘X))
Prolif. zone (PZ)
Label. index (LI)
Prolif. zone (PZ)
Label. index (LI)
0.00
10.6t
I0.9T kO.6 16.8t* *2.0 19.2* f 1.9 i5.8?* l 3.1 13.5t* +0.6
13.8t
Il.5$ l 1.0 l7.91_ f 1.4 26.6* l 2.5 23.2t* *2.2 24.1t* *2.7
0.025 0.05 0.10 0.20
10.6 13.5*t f 1.3 16.1* f 1.5 16.2* f 1.6 13.0*t kO.8
*0.7 17.3’ hO.9 16.6* l 0.7 I7.4f *0.7 17.3* * 1.4
The effect of CDC on each parameter for -AOM and +AOM groups was assessed separately. ferent superscripts are significantly different (P 5 0.05). “The proliferative zone is assessed by enumerating the highest location of the PCNA-labelled those values to determine the size of the proliferative zone/colon. bNumber of PCNA-labelled cells/l00 cells/group. ‘Values are mean + S.E.M.. n = 6 animals/group.
Means within each column ceils in each crypt.
with dif-
then averaging
106
4. Discussion In the present study, dietary feeding of CDC resulted in a dose-related increase in the number of ACF in the rat colon. CDC feeding did not increase mitotic and BUDR labelling indices in carcinogen-treated animals; however, an increase in mitotic indices was observed in the normal colonic epithelium. Analysis of the number and distribution of cells exhibiting PCNA revealed that significant increases occurred in CH, PZ and LI, with higher levels of CDC (0.05, 0.1 and 0.2) in both the injected (+AOM) and non-injected (-AOM) colons when compared with the 0% CDC group. This effect was more pronounced in the +AOM group. Aberrant crypt foci have been identified in rodent [3] and human colons [16,21] and are proposed to be pre-neoplastic lesions [3]. The ACF system has been used extensively by several other investigators to screen and identify potential modulators of colon carcinogenesis [ 15,241. It has been demonstrated that the genotypic and phenotypic features of ACF are those to be expected in preneoplastic lesions [12]. The ability of CDC to increase the number of ACF in the rat colon is in agreement with the tumor-enhancing effect of CDC [7,8,22]. CDC is one of the primary bile acids derived from cholesterol metabolism and is currently used clinically to dissolve gallstones [5]. Chenodeoxycholic acid is thought to exert its effect through its metabolism into lithocholic acid by colonic microflora. Lithocholic acid has been reported to be a colonic tumor promoter and also augments the genotoxicity of known carcinogens 1171. The exact mechanism(s) by which CDC or lithocholic acid may enhance the number of ACF or tumor incidence is unknown. Measurements of the proliferative indices of colonic epithelium have been used frequently in animal studies to assess risk for cancer development [ 10,191. Several approaches have been taken in this regard, including the quantification of mitotic and/or BUDR labelling indices (as a measure of Sphase cells) and the quantification of cells exhibiting proliferating cell nuclear antigen (PCNA) [ 191. It has been reported that there is general correspondence between BUDR and PCNA labelling indices in normal-appearing crypts, the latter In the present study being slightly higher [ 191.
L.A.M.
Sutherland, R.P. Bird/Cancer
Leu. 76 (1994)
101-107
the proliferative indices of colonic epithelia of animals treated with a carcinogen (+AOM) or saline (-AOM) and fed CDC were measured by all three different methods. Feeding CDC resulted in an increase in the crypt height, suggesting that this bile acid did exert a hyperplastic response. Interestingly, this observation only occurred in the +AOM group, suggesting that the carcinogentreated colons responded differently to CDC feeding. Our observation that 0.2% CDC seemed to decrease the mitotic index in the colonic crypts supports the observation by Simanowski and coworkers [23], who also demonstrated that a 1% CDC treatment reduced the proliferative response in the rat colon. These effects are incongruous to the proposed role for cell proliferation in mediating the biological effects of bile acid [9]. The mechanisms by which CDC feeding increased the number of PCNA-labelled cells and the significance of these changes in the development of ACF and colon cancer remain to be evaluated. Proliferating cell nuclear antigens are nuclear proteins functioning as auxiliary proteins of DNA polymerase [20]. These proteins may be regulated differently in normal and neoplastic tissue [20]. Furthermore, increased expression of PCNA has been associated with DNA excision repair synthesis [ll]. It is noteworthy that the effects of CDC on the induction of ACF and proliferative indices of the colon are quite different from those reported for cholic acid [ 141. It has been reported previously that cholic acid is highly mitogenic and reduces the number of ACF, while it is capable of enhancing tumor incidence. In the present study CDC was able to increase the number of ACF without exerting a distinct mitogenic response. These findings allude to the fact that the ability of various bile acids to enhance tumor growth may not necessarily be mediated via increased cell proliferation. It is proposed that bile acids may serve as an important group of compounds with structural and biological heterogenicity which may reveal cellular and biochemical events important in the initiation and post-initiation stages of colon carcinogenesis. 5. Acknowledgments This study was supported by the Natural Sciences and Engineering Research Council of
L.A.M.
Sutherland, R.P. Bird/Cancer
Lett. 76 (1994)
101-107
Canada (NSERC) and the National Cancer Institute of Canada. L. Sutherland is the recipient of the undergraduate NSERC award. 6. References 1
2
3
4
5
6
7
8
American Institute of Nutrition (1977) Report of the AIN ad hoc committee on standards for nutrition studies. J. Nutr., 107, 1340-1348. American Institute of Nutrition (1980) Second report of the ad hoc committee on standards for nutrition studies. J. Nutr., 110, 1726. Bird, R.P. (1987) Observation and quantification of aberrant crypts in the murine colon treated with a colon carcinogen: preliminary findings. Cancer Lett., 37, 147-151. Bird, R.P., Schneider, R., Stamp, D. and Bruce, W.R. (1986) Effect of dietary calcium and cholic acid on the proliferative indices of murine colonic epithelium. Carcinogenesis, 7, 1657-1661. Cohen, B.1. and Raicht, R.F. (1981) Effects of bile acids on colon carcinogenesis in rats treated with a carcinogen. Cancer Res., 41, 3759-3760. Cohen, B.I., Raicht, R.F., Deschner. E.E. Takahashi, M., Sarwal, A.N. and Fazzini, E. (1980) Effect of cholic acid feeding in N-methyl-N-nitrosourea induced colon tumors and cell kinetics in rats. J. Natl. Cancer Inst., 64, 573-578. Cyzan, P., Seitz, H.K., Waldherr, R., Stiehl, A., Raedsch, R. and Kommerell, B. (1982) Chenodeoxycholic acid but not ursodeoxycholic acid enhances colon carcinogenesis in the rat. In: Bile Acids and Cholesterol in Health and Disease, pp. 343-45. Editors: G. Paumgartner, A. Stiehl, G. Gerok. MTP Press, Lancaster, PA. Danzinger, R.G., Hofmann, A.F., Schoenfteld, L.J. and Thistle, J.L. (1972) Dissolution of cholesterol gallstones by chenodeoxycholic acid. N. Engl. J. Med., 286, l-8. Deschner, E.E. and Raicht, R.F. (1979) Influence of bile in kinetic behavior of colonic epithelial cells of the rat. Digestion, 19, 322-327. Deschner, E.E., Long, F.C., Hakissian, M. and Herrmann, S.L. (1983) Differential susceptibility of AKR, C57BL/6 and CFI mice to 1,2-dimethylhydrazine-induced colonic tumor formation predicted by proliferative characteristics of colonic epithelial cells. J. Natl. Cancer Inst., 70, 279-282. Hall, P.A. and Woods, A.L. (1990) Immunohistochemical markers of cellular prolifertion: achievements, problems and prospects [invited review]. Cell Tissue Kinet., 23, 505-522. McLellan, E.A., Medline, A. and Bird, R.P. (1991) Sequential analysis of the growth and morphological characteristics of aberrant crypt foci: putative preneoplastic lesions. Cancer Res., 51, 5270-5274.
107
13
14
15
16
17
18
19
20
21
22
23
24
McSherry, Mosbach,
C.K., Cohen, B.I., Bokkenheuser. V.D.. E.H., Winter, J., Matoba, N. and Scholes, J.
(1989) Effects of calcium and bile acid feeding on colon tumors in the rat. Cancer Res., 49, 6039-6043. Magnuson, B.A. and Bird, R.P. (1993) Reduction of aberrant crypt foci induced in rat colon with azoxymethane or methylnitrosurea by feeding cholic acid. Cancer Lett.. 68. 15-23. Pereira, M.A. and Khory. M.D. (1991) Prevention by chemopreventive agents of azoxymethane-induced foci of aberrant crypts in rat colon. Cancer Lett.. 61, 27-33. Pretlow, T.P., Barrow, B.J., Ashton, W.S., O’Riordan. M.A., Pretlow, T.G., Jurcisek, J.A. and Stellato, T.A. (1991) Aberrant crypts: putative preneoplastic foci in human colonic mucosa. Cancer Res., 51, 1564- 1567. Reddy, B.S. and Wantanabe, K. (1979) Effect of cholesterol metabolites and promoting effect of lithocholic acid in colon carcinogenesis in germ-free and conventional F344 rats. Cancer Res., 39, 1521-1524. Reddy, B.S., Wantanabe, K., Weisburger, J.R. and Wynder, E.L. (1977) Promoting effect of bile acids in colon carcinogenesis in germ-free and conventional F344 rats. Cancer Res., 37, 3238-3242. Richter, F., Richter, A.. Yang, K. and Lipkin, M. (1992) Cell proliferation in rat colon measured with bromodeoxyuridine, proliferating cell nuclear antigen and [3H]thymidine. Cancer Epidemiol. Biomarkers Prevent.. 1, 561566. Risio, M., Candelaresi, G. and Rossini. F.P. (1993) Bromodeoxyuridine uptake and proliferating cell nuclear antigen expression throughout the colorectal tumor sequence. Cancer Epidemiol. Biomarkers Prevent., 2, 363-367. Roncucci, L., Stamp, D., Medline, A., Cullen, J.B. and Bruce, W.R. (1991) Identification and quantification of aberrant crypt foci and microadenomas in the human colon. Hum. Pathol., 22, 287-294. Sarwal, A.N., Cohen, B.I., Raicht, R.F., Takahashi, M. and Fazzini, E. (1979) Effects of dietary administration of chenodeoxycholic acid on N-methyl-N-nitrosoureainduced colon cancer in rats. Biochim. Biophys. Acta. 574, 423-432. Simanowski, U.A., Seitz, H.K., Cyzan, P., Horner, M., Waldherr, R., Weber, E. and Kommerell, B. (1987) Chronic ursodeoxycholic acid and chenodeoxycholic acid-feeding-induced changes of colon mucosal cell proliferation in rats. J. Natl. Cancer Inst., 79, 163-166. Steel, V.E. and Kelloff, G.J. (1993). Evaluation of a rat colon crypt assay to identify chemopreventive agents compared to rodent tumor assay systems. Proc. Am. Assoc. Cancer Res., 34, 553.