Predictive Markers and Cancer Prevention
Silymarin, a naturally occurring polyphenolic antioxidant flavonoid, inhibits azoxymethane-induced colon carcinogenesis in male F344 rats
Version of Record online: 13 AUG 2002
Copyright © 2002 Wiley-Liss, Inc.
International Journal of Cancer
Volume 101, Issue 5, pages 461–468, 10 October 2002
How to Cite
Kohno, H., Tanaka, T., Kawabata, K., Hirose, Y., Sugie, S., Tsuda, H. and Mori, H. (2002), Silymarin, a naturally occurring polyphenolic antioxidant flavonoid, inhibits azoxymethane-induced colon carcinogenesis in male F344 rats. Int. J. Cancer, 101: 461–468. doi: 10.1002/ijc.10625
- Issue online: 3 SEP 2002
- Version of Record online: 13 AUG 2002
- Manuscript Accepted: 4 JUL 2002
- Manuscript Revised: 24 JUN 2002
- Manuscript Received: 25 JAN 2002
- Ministry of Health, Labor and Welfare of Japan
- Ministry of Education, Science, Sports and Culture of Japan
- High-Technology Center of Kanazawa Medical University
- aberrant crypt foci;
- colon carcinogenesis;
The modifying effect of dietary administration of the polyphenolic antioxidant flavonoid silymarin, isolated from milk thistle [Silybum marianum (L.) Gaertneri], on AOM-induced colon carcinogenesis was investigated in male F344 rats. In the short-term study, the effects of silymarin on the development of AOM-induced colonic ACF, being putative precursor lesions for colonic adenocarcinoma, were assayed to predict the modifying effects of dietary silymarin on colon tumorigenesis. Also, the activity of detoxifying enzymes (GST and QR) in liver and colonic mucosa was determined in rats gavaged with silymarin. Subsequently, the possible inhibitory effects of dietary feeding of silymarin on AOM-induced colon carcinogenesis were evaluated using a long-term animal experiment. In the short-term study, dietary administration of silymarin (100, 500 and 1,000 ppm in diet), either during or after carcinogen exposure, for 4 weeks caused significant reduction in the frequency of colonic ACF in a dose-dependent manner. Silymarin given by gavage elevated the activity of detoxifying enzymes in both organs. In the long-term experiment, dietary feeding of silymarin (100 and 500 ppm) during the initiation or postinitiation phase of AOM-induced colon carcinogenesis reduced the incidence and multiplicity of colonic adenocarcinoma. The inhibition by feeding with 500 ppm silymarin was significant (p < 0.05 by initiation feeding and p < 0.01 by postinitiation feeding). Also, silymarin administration in the diet lowered the PCNA labeling index and increased the number of apoptotic cells in adenocarcinoma. β-Glucuronidase activity, PGE2 level and polyamine content were decreased in colonic mucosa. These results clearly indicate a chemopreventive ability of dietary silymarin against chemically induced colon tumorigenesis and will provide a scientific basis for progression to clinical trials of the chemoprevention of human colon cancer. © 2002 Wiley-Liss, Inc.
Colorectal cancer is the third most common malignant neoplasm in the world.1 In Japan, its incidence has been increasing, and it is now the third leading cause of cancer death. In this context, primary prevention, including chemoprevention, is important.
Silymarin, the collective name for an extract from milk thistle [Silybum marianum (L.) Gaertneri],2 is a naturally occurring polyphenolic flavonoid antioxidant.3 It is composed mainly (−80%, w/w) of silybin (also called silybinin, silibin or silibinin), with smaller amounts of other stereoisomers, such as isosilybin, dihydrosilybin, silydianin and silychristin.4 Silymarin protects experimental animals against the hepatotoxin α-amanitin2 and has a strong antioxidant property.5 Other biologic properties of silymarin and its components have been reported, including inhibition of LOX6 and PG synthetase.7 For over 20 years, silymarin has been used clinically in Europe for the treatment of alcoholic liver disease and as an antihepatotoxic agent.8 As a therapeutic agent, it is well tolerated and largely free of adverse effects.5 It might be a potent anticarcinogen against in vitro9 and in vivo10, 11 carcinogenesis. However, animal chemopreventive studies have been mainly limited to skin,11, 12 and only few studies have involved the digestive organs, including colon.10 In mouse epidermis, silymarin inhibits tumor promoter 12-O-tetradecanoylphorbol-13-acetate-induced ODC activity and mRNA expression, which reflect cell proliferation activity.13 Silymarin inhibits mRNA expression of an endogenous tumor promoter, TNF-α.14 Thus, treatment with silymarin inhibits mouse skin tumorigenesis. Also, at lower nontoxic concentrations, it inhibits transformation in cultured rat tracheal epithelial cells treated with benzo[a]pyrene, by which chemopreventive compounds that act at early stages of the carcinogenic process could be identified.15 The silymarin group of flavonoids (silybin, silychristin and silydianin) inhibits xanthine oxidase.16 Zi et al.17 suggested that silymarin causes G1 arrest in human prostate carcinoma DU145 cell and causes growth inhibition by inactivation of erbB1-SHC signaling pathway leading to up-regulation of Kip1/p27 followed by its increased binding with CDK causing a decrease in CDK- and cyclin-associated kinase activity. These findings led us to evaluate the possible suppressing effects of dietary silymarin on the development of ACF, and early biomarker of colorectal carcinogenesis and colorectal tumors in rats.
In the present study, the possible modifying effect of silymarin on the development of AOM-induced rat colon carcinogenesis was investigated using the short-term bioassay of ACF, which are putative preneoplastic lesions for the development of colonic tumors.18, 19 Further, the effect on the activity of the detoxifying enzymes GST and QR was investigated as certain inducers of GST and/or QR are possible cancer chemopreventive agents.20 In the long-term bioassay, biomarkers such as PCNA labeling index and apoptotic index were measured in colonic neoplasms. Polyamine levels in the surrounding colonic mucosa were also examined since polyamine is intimately involved in growth-related processes, including colon tumorigenesis,21 and its metabolism is a target for cancer chemoprevention.22 Colonic PGE2 level and β-glucuronidase activity were also determined. Increased β-glucuronidase activity in either microflora or colonic mucosa plays an important role in colon carcinogenesis induced by carcinogenic hydrazine derivatives, including AOM and DMH.23 Inhibitors of this enzyme inhibit colon carcinogenesis.24 Silymarin and silybin are inhibitors of β-glucuronidase.25
MATERIAL AND METHODS
Animals, diets and chemicals
Male F344 rats were obtained at 4 weeks of age from Japan SLC (Hamamatsu, Japan). Wire cages were used to house 3 or 4/cage. After their arrival from the supplier, they were quarantined and fed a basal diet (CE-2; Clea, Tokyo, Japan) for 1 week. Rats were maintained in an experimental room under controlled conditions of 23 ± 2°C (SD), 50 ± 10% humidity and a 12 hr light/dark cycle. Drinking water and diet were supplied ad libitum. AOM (Sigma, St. Louis, MO) was used to induce colonic ACF and neoplasms. Silymarin was purchased from Sigma-Aldrich (Tokyo, Japan). Experimental diets were made on a weekly basis by mixing silymarin at doses of 100, 500 and 1,000 ppm (w/w) with a powdered CE-2 basal diet and stored in a cold room (<4°C) until used.
ACF assay (short-term experiment)
After quarantine for 1 week, 80 rats aged 5 weeks were divided into 9 groups (Fig. 1). Starting at 6 weeks of age, animals in groups 1–7 were s.c.-injected with AOM (20 mg/kg body weight) once a week for 2 weeks to induce colonic ACF. They received injections of AOM between 9:00 and 10:00 AM. Rats in groups 2–4 were given the basal diet, CE-2, containing 100, 500 or 1,000 ppm silymarin, respectively, for 4 weeks, beginning at 5 weeks of age. Animals in groups 5–7 were given the basal diet mixed with 100, 500 or 1,000 ppm silymarin, respectively, for 4 weeks, beginning 2 weeks after the last injection of AOM. Animals in group 8 did not receive AOM and were fed the diet containing silymarin (1,000 ppm) for 8 weeks. Group 9 served as an untreated control. During the study, animals were carefully observed daily for clinical signs of toxicity (emaciation, hematuria, diarrhea, bloody stool and nervous activity). They were weighed daily. Animals were killed at week 4 or 8 by ether overdose to evaluate the modifying effects of dietary silymarin on the development of AOM-induced ACF. Colons were removed from cecum to anus, flushed with 0.9% NaCl solution, slit open longitudinally, flattened between a filter paper and fixed in 10% buffered formalin at room temperature. Colonic ACF were counted on the mucosa stained for 1 min in a solution of 0.5% methylene blue (Sigma) dissolved in distilled water. ACF counting was done under a light microscope at a magnification of x40. The number of ACF and their location in the colon were recorded along with the number of crypts in each focus.
Assay of GST and QR activities
Twenty-six male rats aged 5 weeks were used for the GST and QR assay. They were gavaged with silymarin at a dose level of 0, 40, 200 or 400 mg/kg body weight in 0.5 ml of 5% gum arabic (Sigma) for 5 consecutive days. All rats were killed by decapitation 30 min after the last gavage. Livers and colons were excised immediately. Livers were perfused with saline to remove blood and minced into small pieces. The colon was slit longitudinally and washed with PBS (pH 7.4), and the mucosa was collected by scraping the mucosal surface using a stainless steel disposable microtome knife (S35; Feather Safety Razor, Seki, Japan). Aliquots of minced liver and of mucosal scrapings were processed for the cytosolic fraction. GST and QR activity was determined using CDNB or DCNB for GST26 and NADH/menadione for QR as substrates.27 All spectrographic assays were based on absorption at 340 nm, and all samples were measured in triplicate. One unit of enzyme activity is defined as the amount of enzyme catalyzing the conversion of 1 μmol of substrate to product per min at 25°C. Cytosolic protein concentrations were determined by the Bradford28 method using BSA as the standard.
Long-term experiment for evaluating modulatory effects of dietary silymarin
For the long-term study (Fig. 2), 120 rats aged 5 weeks were randomly divided into 7 groups. Groups 1–5 received 3 weekly s.c. injections of AOM (15 mg/kg body weight). Rats in groups 2 and 3 were fed diets containing 100 or 500 ppm silymarin for 4 weeks, respectively, commencing 1 week before the first dose of AOM. Groups 4 and 5 were fed diets mixed with 100 and 500 ppm silymarin, respectively, for 30 weeks of the postinitiation phase, starting 1 week after the last administration of AOM. Group 6 was fed the diet mixed with 500 ppm silymarin for the entire study period (34 weeks). Group 7 served as an untreated control. All rats were carefully observed daily for clinical welfare and weighed weekly until they reached 14 weeks of age and every 4 weeks thereafter. Consumption of the experimental diets was also recorded to estimate the intake of test compounds. The experiment was terminated 34 weeks after commencement, and all animals were killed by ether overdose to assess the incidence of large bowel tumors and ACF. At autopsy, all organs were macroscopically inspected for the presence of pathologic lesions; and then liver, kidney, heart, lung and digestive tract including intestine were isolated and fixed in 10% buffered formalin. Liver and kidney were weighed and processed for histopathologic examination. The intestine was excised, opened longitudinally, flushed clean with saline and examined for the presence of tumors. After removing colonic tumors, colons were fixed in 10% buffered formalin and then processed for ACF analysis and histopathologic examination by conventional methods. Five rats randomly selected from each group were used for ACF assessment. Neoplasms in the intestine were diagnosed on hematoxylin and eosin–stained sections of 3 μm, according to the criteria described by Ward.29 If tumor cells with tubular formation invaded the submucosa, the tumor was diagnosed as adenocarcinoma. When tumor cells did not invade the submucosa, the tumor was diagnosed as adenoma. Other organs, those with macroscopic changes in kidney, heart, lung and digestive tract, were also examined histopathologically after fixation with 10% buffered formalin.
Measurement of PCNA and apoptotic indices in colonic neoplasms
All colonic tumors were used for measurement of PCNA labeling and apoptotic indices. Tumor tissues fixed in 10% buffered formalin for 2 days were embedded in paraffin. Serial cross-sections of 3 μm were cut and mounted onto gelatin-coated glass slides. PCNA and apoptotic cells were determined by immunohistochemistry. A mouse primary MAb against PCNA (PC10, 1:50 dilution; Dako, Kyoto, Japan) and a rabbit polyclonal primary antibody against ssDNA (1:300 dilution, Dako) were applied to the sections according to the manufacturer's protocol (Dako LSAB 2 kit/HRP). All incubation steps were carried out for 15 min at 37°C. Negative controls were prepared by omitting primary antibodies. As secondary antibody, goat polyclonal antirabbit/mouse immunoglobulin labeled with biotin (Dako) was used. The labeled streptavidin-biotin immunoperoxidase technique was applied for immunohistochemistry. All nuclei that densely immunoreacted with PCNA or ssDNA antibody were regarded as PCNA- or ssDNA-positive. PCNA and apoptotic indices were determined by counting the number of positive cells among at least 200 cells in the tumor and are indicated as percentages.
β-Glucuronidase activity of scraped colonic mucosa without tumors from 4 rats of each group was determined spectrophotometrically by measuring the release of phenolphthalein from sodium phenolphthalein-β-glucuronide as described by Goldin and Gorbach.30 Protein was determined by the Bradford28 method using BSA as the standard.
Measurement of PGE2 level
For PGE2 determination, scraped colonic mucosa without tumors of 4 rats from each group was homogenized in 400 μl of PBS on ice. After centrifugation at 10,000g for 5 min, supernatants were diluted at a ratio of 1:9 and PGE2 concentration was measured using a commercial experimental kit (Cayman, Ann Arbor, MI) according to the protocol of the manufacturer. Samples were assayed in triplicate. Protein concentrations for tissue samples were determined using the Bradford28 method.
Scraped colonic mucosa without tumors was stored at −70°C until measured. Proteins were extracted from the mucosa and tissue polyamine levels in the extraction determined by the methods described previously.31
Data (mean ± SD) were analyzed by 1-way ANOVA followed by the Bonferroni/Dunn post-hoc test or Kruskal-Wallis nonparametric 1-way ANOVA with Bonferroni correction. Tumor incidence was compared using the χ2 test (when one of the expected frequencies was >6) or Fisher's exact probability test (when one of the expected frequencies was <5). p < 0.05 was considered significant.
ACF assay (short-term study)
Results of the ACF assay are summarized in Tables I and II. An 8-week period of silymarin feeding alone (group 8) did not cause any clinical signs of toxicity, though at week 4 treatment of silymarin together with AOM (groups 2–4) reduced body weight gain and liver weight (Table I). As shown in Table II, initiation feeding with silymarin at all doses significantly reduced the occurrence of total ACF (p < 0.001). Also, the frequency of large ACF consisting of 4 or more crypts was significantly reduced by feeding with silymarin at all dose levels (p < 0.001 or p < 0.01) (Table II). Similarly, at week 8, silymarin postinitiation feeding at all dose levels significantly inhibited the total number of ACF (p < 0.05 or p < 0.001) (Table II). Inhibition rates caused by postinitiation feeding with silymarin were 26% at 100 ppm, 23% at 500 ppm and 63% at 1,000 ppm. However, postinitiation feeding with silymarin did not significantly affect the number of large ACF (Table II).
|Group||Treatment||Week 41||Week 81|
|Body weight (g)||Liver weight (g)||Body weight (g)||Liver weight (g)|
|1||AOM alone||218 ± 8 (8)||10.8 ± 0.9||269 ± 13 (8)||11.3 ± 1.7|
|2||AOM + 100 ppm silymarin||198 ± 32 (8)||8.8 ± 0.82||—||—|
|3||AOM + 500 ppm silymarin||196 ± 73 (8)||9.3 ± 1.63||—||—|
|4||AOM + 1,000 ppm silymarin||193 ± 62 (8)||9.8 ± 0.94||—||—|
|5||AOM 100 ppm silymarin||—||—||259 ± 8 (8)||11.8 ± 1.5|
|6||AOM 500 ppm silymarin||—||—||259 ± 9 (8)||12.9 ± 1.6|
|7||AOM 1,000 ppm silymarin||—||—||249 ± 123 (8)||11.0 ± 0.8|
|8||1,000 ppm silymarin||218 ± 12 (4)||11.5 ± 0.5||246 ± 11 (4)||10.4 ± 1.1|
|9||No treatment||210 ± 5 (4)||10.5 ± 0.9||268 ± 15 (4)||12.5 ± 1.7|
|Group||Treatment||Week 41||Week 81|
|Total ACF/colon||1 crypt (%)||2 crypts (%)||3 crypts (%)||4 or more crypts (%)||Total ACF/colon||1 crypt (%)||2 crypts (%)||3 crypts (%)||4 or more crypts (%)|
|1||AOM alone||140 ± 17 (8)||44.8 ± 3.8||32.5 ± 4.1||10.8 ± 1.7||12.0 ± 1.2||156 ± 21 (8)||37.8 ± 3.8||28.6 ± 2.2||20.2 ± 2.7||13.4 ± 3.6|
|2||AOM + 100 ppm silymarin||74 ± 132 (8)||48.3 ± 2.7||36.4 ± 5.1||10.6 ± 2.4||5.2 ± 1.42||—||—||—||—||—|
|3||AOM + 500 ppm silymarin||64 ± 102 (8)||46.4 ± 4.0||37.1 ± 4.4||12.5 ± 1.5||4.2 ± 3.33||—||—||—||—||—|
|4||AOM + 1,000 ppm silymarin||50 ± 72 (8)||51.1 ± 5.0||32.4 ± 2.1||12.0 ± 3.2||4.4 ± 3.13||—||—||—||—||—|
|5||AOM 100 ppm silymarin||—||—||—||—||—||116 ± 204 (8)||38.8 ± 5.7||33.0 ± 3.03||19.4 ± 3.0||15.0 ± 2.3|
|6||AOM 500 ppm silymarin||—||—||—||—||—||120 ± 54 (8)||33.9 ± 5.0||35.2 ± 3.32||17.9 ± 4.2||15.1 ± 3.0|
|7||AOM 1,000 ppm silymarin||—||—||—||—||—||58 ± 62 (8)||47.0 ± 2.92||28.4 ± 6.5||14.7 ± 2.12||11.2 ± 4.5|
GST and QR activities
GST and QR activities in the liver and colon of rats gavaged with silymarin are given in Table III. Dosing with 40, 200 and 400 mg/kg body weight of silymarin significantly elevated liver GST (GST-CDNB) (p < 0.001). Also, gavage with 400 mg/kg body weight of silymarin significantly increased liver QR activity (p < 0.01). GST-CDNB activity in the colonic mucosa of rats gavaged with 40, 200 and 400 mg/kg body weight of silymarin was significantly higher than that of rats with solvent alone (p < 0.001 or p < 0.05). Colonic QR activity of these rats was also elevated by gavage with silymarin, but the increase was not statistically significant.
|Organ||Enzyme||Enzyme activity (mean ± SD, mU/mg protein)1|
|0 mg||40 mg||200 mg||400 mg|
|Liver||GST-CDNB||802.1 ± 49.5 (6)||986.3 ± 71.22 (6)||1032.2 ± 63.02 (7)||1351.7 ± 165.62 (7)|
|GST-DCNB||34.3 ± 2.8 (6)||37.0 ± 1.7 (6)||41.6 ± 4.03 (7)||44.1 ± 5.23 (7)|
|QR||131.0 ± 33.5 (6)||137.3 ± 18.5 (6)||148.3 ± 23.8 (7)||240.8 ± 54.13 (7)|
|Colon||GST-CDNB||142.5 ± 4.7 (6)||155.9 ± 4.62 (6)||157.8 ± 11.94 (7)||178.3 ± 12.82 (7)|
|QR||510.8 ± 38.4 (6)||521.0 ± 18.7 (6)||553.4 ± 27.1 (7)||557.4 ± 78.1 (7)|
During the experiment, clinical signs of toxicity, low survival and poor condition were not observed in any group. This was confirmed by histologic examination in liver, kidney, heart and lungs. Mean daily intake of a diet with or without silymarin was between 22.7 and 23.6 g (mean daily intake of silymarin/rat: 100 ppm diet, 2.35 mg; 500 ppm diet, 11.45 mg). Mean body and liver weights in all groups are shown in Table IV. There were no significant differences in mean body weights among groups. Mean liver weight of rats in group 1 (AOM alone) was significantly lower than that of rats in group 7 (untreated, p < 0.01). Mean and relative liver weights of groups 2 (AOM + 100 ppm silymarin), 3 (AOM + 500 ppm silymarin), 4 (AOM 100 ppm silymarin) and 5 (AOM 500 ppm silymarin) were significantly higher than group 1 (P < 0.001). The mean relative liver weight (g/100 g body weight) of group 1 was significantly lower than group 7 (P < 0.001).
|Group||Treatment||Number of rats examined||Body weight (g)||Liver weight (g)||Relative liver weight (g/100 g body weight)|
|1||AOM||20||359 ± 18||11.8 ± 0.82||3.43 ± 0.223|
|2||AOM + 100 ppm silymarin||19||350 ± 21||14.9 ± 2.24||4.20 ± 0.504|
|3||AOM + 500 ppm silymarin||19||365 ± 13||16.3 ± 1.84||4.44 ± 0.434|
|4||AOM 100 ppm silymarin||18||347 ± 15||13.8 ± 1.84||3.96 ± 0.434|
|5||AOM 500 ppm silymarin||20||347 ± 18||14.1 ± 1.74||4.11 ± 0.494|
|6||500 ppm silymarin||12||345 ± 17||13.7 ± 2.2||3.96 ± 0.50|
|7||None||12||356 ± 20||14.1 ± 2.3||4.23 ± 0.43|
Incidence and multiplicity of the intestinal neoplasms
Macroscopically, most tumours developed in the large intestine and some in the small intestine of rats in groups 1–5. Large intestinal tumors were mainly located in the middle and distal colon. No tumors were found in any organs of animals in groups 6 and 7. Colonic tumors were macroscopically sessile or pedunculated and histologically tubular adenoma, tubular adenocarcinoma or signet-ring cell carcinoma, with a high incidence of tubular adenocarcinoma. Some rats had both adenoma and adenocarcinoma in the intestine. A few rats given AOM had mesenchymal renal tumors and/or preneoplastic hepatocellular lesions, but these pathologic lesions were not observed in groups 6 and 7. The incidence and multiplicity of intestinal tumors are shown in Tables V and VI, respectively. The frequency of large intestinal adenocarcinoma in groups 3 (11%) and 5 (15%) was significantly lower than in group 1 (55%, p < 0.01). The incidence of small intestinal adenocarcinoma did not significantly differ among groups 1–5. As presented in Table VI, a significant reduction in the multiplicity of colonic adenocarcinomas (number of carcinomas/rat) in groups 3 (0.11 ± 0.31) and 5 (0.15 ± 0.36) was observed compared to group 1 (0.75 ± 0.79, p < 0.01).
|Group||Treatment (number of rats examined)||Number of rats with tumors at|
|Entire intestine||Small intestine||Large intestine|
|1||AOM (20)||14 (70%)||5 (25%)||12 (60%)||4 (20%)||1 (5%)||4 (20%)||13 (65%)||4 (20%)||11 (55%)|
|2||AOM + 100 ppm silymarin (19)||10 (53%)||4 (21%)||8 (42%)||3 (16%)||1 (5%)||2 (11%)||9 (47%)||2 (11%)||8 (42%)|
|3||AOM + 500 ppm silymarin (19)||72 (37%)||3 (16%)||43 (21%)||3 (16%)||1 (5%)||2 (11%)||44 (21%)||2 (11%)||24 (11%)|
|4||AOM 100 ppm silymarin (18)||8 (44%)||3 (17%)||7 (39%)||1 (6%)||0 (0%)||1 (6%)||8 (44%)||3 (17%)||7 (39%)|
|5||AOM 500 ppm silymarin (20)||10 (35%)||5 (25%)||6 (30%)||8 (15%)||2 (10%)||7 (35%)||62 (30%)||6 (30%)||34 (15%)|
|6||500 ppm silymarin (12)||0||0||0||0||0||0||0||0||0|
|Group||Treatment (number of rats examined)||Multiplicity (number of tumors/rat) of intestinal tumors at|
|Entire intestine||Small intestine||Large intestine|
|1||AOM (20)||1.30 ± 1.10||0.30 ± 0.56||1.00 ± 1.05||0.30 ± 0.64||0.05 ± 0.22||0.25 ± 0.54||1.00 ± 0.89||0.25 ± 0.54||0.75 ± 0.79|
|2||AOM + 100 ppm silymarin (19)||0.95 ± 1.10||0.21 ± 0.52||0.74 ± 1.02||0.26 ± 0.64||0.05 ± 0.22||0.21 ± 0.52||0.68 ± 0.80||0.16 ± 0.49||0.53 ± 0.68|
|3||AOM + 500 ppm silymarin (19)||0.37 ± 0.482||0.16 ± 0.36||0.21 ± 0.412||0.16 ± 0.36||0.05 ± 0.22||0.11 ± 0.31||0.21 ± 0.412||0.11 ± 0.31||0.11 ± 0.312|
|4||AOM 100 ppm silymarin (18)||0.67 ± 0.82||0.17 ± 0.37||0.50 ± 0.69||0.06 ± 0.23||0||0.06 ± 0.23||0.61 ± 0.76||0.17 ± 0.37||0.44 ± 0.60|
|5||AOM 500 ppm silymarin (20)||0.95 ± 1.07||0.40 ± 0.49||0.55 ± 0.67||0.50 ± 0.74||0.10 ± 0.30||0.40 ± 0.58||0.45 ± 0.743||0.30 ± 0.46||0.15 ± 0.362|
Frequency of ACF at the end of study
The incidence of ACF at the end of the study is shown in Table VII. ACF developed in rats treated with silymarin together with or after AOM exposure (groups 1–5), whereas no ACF were found in rats that were not dosed with AOM (groups 6 and 7). The frequency of ACF/colon in groups 3 and 5 was significantly smaller than that of group 1 (p < 0.05). Total aberrant crypts per colon in groups 2–5 were significantly lower than in group 1 (p < 0.05 or p < 0.01).
|Group||Treatment (number of rats examined)||Total number of ACF/colon||Aberrant crypts/colon||Aberrant crypts/focus|
|1||AOM (5)||94 ± 33||220 ± 39||2.74 ± 0.38|
|2||AOM + 100 ppm silymarin (5)||68 ± 19||149 ± 382||2.19 ± 0.43|
|3||AOM + 500 ppm silymarin (5)||55 ± 152||121 ± 453||2.20 ± 0.44|
|4||AOM 100 ppm silymarin (5)||64 ± 22||138 ± 323||2.16 ± 0.54|
|5||AOM 500 ppm silymarin (5)||53 ± 192||116 ± 513||2.19 ± 0.51|
|6||500 ppm silymarin (5)||0||0||0|
PCNA and apoptotic indices in colonic neoplasms
As indicated in Table VIII, the PCNA-positive index of adenocarcinomas in rats of groups 4 and 5 was significantly smaller than that in group 1 (p < 0.001). The apoptotic index of adenocarcinomas in rats of groups 2, 4 and 5 was significantly larger than that in group 1 (p < 0.001, p < 0.01 and p < 0.001, respectively). However, these parameters for cell proliferation did not significantly differ among adenomas in rats of groups 1–5.
|Group||Treatment||PCNA-positive index (%)||Apoptotic index (%)|
|Adenoma/number of tumors examined||Adenocarcinoma/number of tumors examined||Adenoma/number of tumors examined||Adenocarcinoma/number of tumors examined|
|1||AOM alone||33.5 ± 3.5/4||55.9 ± 6.9/11||4.75 ± 1.71/4||3.54 ± 1.02/11|
|2||AOM + 100 ppm silymarin||26.5 ± 3.5/2||48.9 ± 8.9/8||5.05 ± 0.35/2||5.68 ± 1.16/82|
|3||AOM + 500 ppm silymarin||26.0 ± 5.0/2||32.0 ± 8.5/2||5.90 ± 0.80/2||3.70 ± 0.30/2|
|4||AOM 100 ppm silymarin||24.7 ± 10.8/3||39.6 ± 4.6/72||5.90 ± 0.86/3||5.29 ± 0.78/73|
|5||AOM 500 ppm silymarin||27.3 ± 6.5/6||29.5 ± 4.6/32||6.33 ± 1.17/6||8.20 ± 0.78/32|
β-Glucuronidase activity, PGE2 level and polyamine content in colonic mucosa
The data on assays of β-glucuronidase activity, PGE2 level and polyamine content in the colonic mucosa determined at the end of the study are summarized in Table IX. β-Glucuronidase activity, PGE2 level and total polyamine content in group 1 were significantly greater than in group 7 (p < 0.01 or p < 0.05). β-Glucuronidase activity in groups 2–5 was significantly lower than that in group 1 (p < 0.01 or p < 0.001). The PGE2 level in groups 3 and 5 was significantly lower than in group 1 (p < 0.01). Total polyamine content in groups 3–5 was significantly smaller than that in group 1 (p < 0.001, p < 0.05 and p < 0.01, respectively).
|Group||Treatment (number of rats examined)||β-Glucuronidase activity (hourly nmol/mg protein)||PGE2 level (pg/mg protein)||Polyamine content (nmol/mg protein)|
|1||AOM alone (4)||0.46 ± 0.082||426 ± 592||0.53 ± 0.29||3.46 ± 0.232||8.09 ± 0.723||12.08 ± 0.153|
|2||AOM + 100 ppm silymarin (4)||0.12 ± 0.024||404 ± 52||0.19 ± 0.14||2.29 ± 0.424||8.30 ± 0.42||10.78 ± 0.82|
|3||AOM + 500 ppm silymarin (4)||0.19 ± 0.045||234 ± 414||0.12 ± 0.19||3.24 ± 1.14||7.11 ± 1.21||9.47 ± 0.135|
|4||AOM 100 ppm silymarin (4)||0.20 ± 0.035||358 ± 79||0.12 ± 0.18||3.04 ± 0.21||6.74 ± 1.14||9.90 ± 1.116|
|5||AOM 500 ppm silymarin (4)||0.15 ± 0.045||199 ± 854||0.14 ± 0.15||3.04 ± 0.21||6.22 ± 0.856||9.40 ± 0.734|
|6||500 ppm silymarin (4)||0.06 ± 0.02||201 ± 49||0.06 ± 0.18||3.01 ± 0.61||5.94 ± 0.39||9.01 ± 0.31|
|7||None (4)||0.05 ± 0.02||192 ± 73||0||2.70 ± 0.20||6.51 ± 1.07||9.21 ± 1.04|
Our present results clearly indicate that dietary feeding of silymarin effectively suppresses the occurrence of colonic ACF and adenocarcinomas induced by AOM when administered during or after the carcinogen treatment. The results described here are basically in agreement with those of Gershbein,10 who found that dietary feeding of silymarin (0.1%) during the entire period of DMH-induced rat intestinal carcinogenesis significantly inhibited the development of large and small intestinal adenocarcinomas. We did not observe the inhibitory effect of silymarin on the incidence of small intestinal neoplasms. This may due to their low incidence and the use of a different carcinogen. Silymarin inhibited the growth of human breast32 and prostate17 cancer cell lines. A chemopreventive effect of silymarin on mouse bladder carcinogenesis has been found.33 Thus, silymarin may possess cancer chemopreventive ability in multiple organs.
Several mechanisms by which chemopreventive agents exert their inhibitory effects on tumorigenesis could be considered. AOM is an intermediate of the colonic carcinogen DMH and is metabolized by cytochrome P-450 2E1 and, possibly, P-450 1A, as well as by the phase II carcinogen-detoxifying enzyme GST.34, 35 However, silymarin has no influence in liver P-450 2E1.36 We37 and others38, 39 have reported that certain chemopreventive agents inhibit the development of ACF and carcinoma induced by AOM through induction of GST and QR. Also, epidemiologic observations suggest that consumption of certain cruciferous vegetables reduces the risk of colon cancer in individuals with GSTM1 null type.40 Our results on GST and QR activities in liver and colon could explain the decrease in ACF formation and the colon cancer development in rats given silymarin during the initiation phase.
Silymarin inhibits increased cell proliferation caused by a radical-generating promoter.13 In the current study, expression of cell proliferation biomarkers such as PCNA labeling index and polyamine level in the colonic adenocarcinoma was significantly inhibited by feeding with this mixture. In addition, dietary silymarin increased the apoptotic index in colonic malignancy. Silymarin feeding reduced polyamine levels in the colonic mucosa surrounding tumors. Thus, it is likely that the inhibition of AOM-induced colonic adenocarcinoma formation for animals consuming silymarin is due in part to the alteration of cell proliferation in the colonic mucosa and neoplasms.
Elevated levels of fecal, microfloral and colonic mucosal β-glucuronidase in rats41 and humans42 are associated with diets high in animal protein and fat. Although there is no evidence that AOM is metabolized by bacterial or colonic mucosal β-glucuronidase, this enzyme is largely responsible for the hydrolysis of glucuronide conjugates in the colon and thus important in the generation of toxic and carcinogenic substances.30, 43 Inhibition of AOM-induced colon tumorigenesis by inhibitors of β-glucuronidase was reported in rats.24, 44 Silymarin and its component silybin protect against CCl4-induced liver toxicity by inhibiting β-glucuronidase activity in the liver and in intestinal bacteria in rats.25 In the present study, dietary administration of silymarin lowered β-glucuronidase activity in the colonic mucosa exposed to AOM, suggesting that silymarin may play a role in preventing the formation of active intermediates from AOM in the colonic mucosa. This may also explain the anti-initiating effect of silymarin on AOM-induced colon tumorigenesis. The significance of this enzyme should be considered in light of its importance in the etiology of colon cancer.
Eicosanoids have been implicated in colon carcinogenesis. Arachidonic acid, a precursor of several biologically active eicosanoids, is the most abundant polyenoic fatty acid found in the phospholipids of mammalian tissues. It is metabolized both by the COX pathway, which produces PGs, thromboxane and prostacyclin, and by the LOX pathway, which leads to the formation of leukotrienes and lipoxins. Arachidonic acid products synthesized via both pathways could modulate colon carcinogenesis,45 and some inhibitors,46 including LOX inhibitors of the arachidonic acid cascade, possess chemopreventive activity in colon carcinogenesis.47 In this context, the inhibition of PG biosynthesis,7 LOX6 and COX-248 after silymarin dosing is of interest and important when considering possible mechanisms behind the chemopreventive effect of silymarin found in the present study. Here, silymarin feeding reduced the biosynthesis of PGE2 in colonic mucosa exposed to AOM, suggesting that silymarin may affect both pathways of arachidonate metabolism.
In summary, dietary administration of silymarin significantly suppressed the development of AOM-induced rat colonic carcinoma, in conjunction with modulation of cell growth in the colonic adenocarcinoma and induction of the phase II enzymes QR and GST in the liver and large intestine, and reduced the levels of β-glucuronidase and PGE2 in the colorectal mucosa. Zhao and Agarwal,49 in a study on the tissue distribution of silibinin (silybin), the major active constituent of silymarin, revealed that free and conjugated silibinin are distributed in stomach, liver, pancreas, lung, prostate and skin in conjunction with phase II enzyme (GST and QR) induction in mice. They did not examine the distribution of silibinin in the intestine but found elevated GST and QR activity in the small intestine of mice, as seen in the colonic mucosa of rats in our study. Our group has demonstrated chemopreventive ability of silymarin in urinary bladder carcinogenesis.33 Therefore, further studies to assess the chemopreventive ability of silymarin are needed in different carcinogenesis models. The results described here and reported by others suggest that silymarin has cancer chemopreventive effects in several organs through several mechanisms.
We thank Ms. S. Yamamoto and Ms. A. Nagaoka for secretarial assistance and the staff of the Research Animal Facility of Kanazawa Medical University for taking good care of the animals.
- 21Polyamines in blood-cells as a cancer marker. Lancet 1979; ii: 912., , .
- 32Anticarcinogenic effect of a flavonoid antioxidant, silymarin, in human breast cancer cells MDA-MB 468: induction of G1 arrest through an increase in Cip1/p21 concomitant with a decrease in kinase activity of cyclin-dependent kinases and associated cyclins. Clin Cancer Res 1998; 4: 1055–64., , .
- 48Significant inhibition by the flavonoid antioxidant silymarin against 12-O-tetradecanoylphorbol-13-acetate-caused modulation of antioxidant and inflammatory enzymes, and cyclooxygenase 2 and interleukin-1α expression in SENCAR mouse epidermis: implications in the prevention of stage I tumor promotion. Mol Carcinogenesis 1999; 26: 321–33., , .