Effects of Level and Type of Dietary Faton Incidence of Mammary Tumors Induced in Female Sprague-Dawley Rats by 7,12-Dimethylbenz(a)Anthracene1
K.K.CARROLL2 and H.T.KHOR, Department of Biochemistry,
University of Western Ontario,London 72, Ontario, Canada
ABSTRACT
Female Sprague-Dawley rats on semi-synthetic diets containing 10% and 20% by weight of corn oil developed more mammary adenocarcinomas after treat-ment with a single oral dose of 7,12-di-methylbenz(a)anthracene than similar rats on diets containing only 0.5% or 5% corn oil. Experiments with 10 different fats and oils fed at the 20% level indi-cated that unsaturated fats enhance the yield of adenocarcinomas more than satu-rated fats. Fibroadenomas and adenomas were also found in small numbers in all dietary groups but the yield did not seem to be influenced by level or type of dietary fat. The possible relevance of these findings to the incidence of breast cancer in humans is discussed.
INTRODUCTION
Experiments carried out in a number of dif-ferent laboratories over the past 30 years have provided evidence that rats and mice on high fat diets are more prone to develop mammary tumors than control animals on low fat diets (1,2). This applies both to spontaneous tumors (3,4) and to tumors induced by various means (5-7),and the effect seems not to be dependent on differences in caloric intake (3,7). The pos-sible involvement of dietary fat in the etiology of human breast cancer has also received in-creasing attention in recent years(2,8-10).
Our studies on the effects of dietary fat have all been carried out with mammary tumors induced in female Sprague-Dawley rats by a single oral dose of dimethylbenz(a)anthracene (DMBA)(11). The initial experiments showed that rats on a high fat semisynthetic diet containing 20% corn oil developed more tumors than rats on a corresponding diet containing 20% coconut oil or on a low fat semisynthetic diet containing only 0.5% corn oil (7). In subsequent studies the level of carcinogen in mammary tissue was measured at different time intervals after injection, and although the average level was somewhat higher on the 20% corn oil diet during the first 12 hr,it seemed doubtful that the difference was sufficient to account for the higher tumor yield (12). Furthermore,it was found that the high corn oil diet enhanced the tumor yield when it was fed only after administration of the carcinogen but not when it was fed only before(13).This suggested that the effect was related to develop-ment of the tumors rather than to distribution and metabolism of the carcinogen or other factors concerned with tumor initiation.
The aim of the present experiments was to determine the effect of intermediate levels of corn oil on mammary tumor incidence and to assess the effects of a number of other dietary fats and oils.
MATERIALS AND METHODS
As in earlier studies, the experiments were carried out with 21-to 22-day-old weanling female rats obtained from Sprague-Dawley, Inc.,Madison,Wisconsin. They were placed on the semisynthetic test diets as soon as received and were housed two to a cage in a tempera-ture-controlled, well-ventilated room with arti-ficial lighting controlled to give alternating 12 hr periods of light and darkness. The propor-tions of major ingredients in the diets are shown in Table I. The composition of the vitamin supplements and the sources of dietary materials have been reported previously(13). The diets were normally prepared in 2 kg lots and were stored in a cold room at 4 C.
DIETARY FAT AND MAMMARY TUMORS
The corn oil, butter and lard used for the experiments were purchased locally. Other fats and oils were donated by Procter and Gamble, Ltd., Hamilton, Ontario. The fatty acid compo-sition of the different dietary fats is shown in Table II.For these analyses,the fats were trans-methylated by refluxing for 2 hr with a mixture prepared by adding 1 vol of acetyl chloride dropwise to 9 vol of reagent grade methanol (19). The methyl esters were then extracted with petroleum ether and analyzed by gas liquid chromatography in a Beckman GC-45 with hydrogen flame detector, using a column of 15% EGSS-X on Chromosorb P. The results were quantitated by integration and were checked by use of National Institutes of Health standard mixtures of fatty acid methyl esters.
At 50 days of age, all rats were given a single dose of 5 mg DMBA in 0.25 ml of sesame oil by stomach tube. As in previous experiments, the rats were transferred to commercial diet for two days before and one day after administra-tion of DMBA to minimize effects of the experimental diets on absorption of the carci-nogen. The DMBA was obtained from Eastman Organic Chemicals, Rochester, N.Y., and was purified by recrystallization from methanol-water as described previously(13).
The diets were continued for four months after administration of the DMBA and the rats were palpated regularly for mammary tumors. At the end of this time they were killed with chloroform and the tumors were preserved in 10% buffered formalin for sectioning and staining with hematoxylin and eosin. To mini-mize variations due to the use of different lots of animals at different times, the rats were allotted randomly in groups of 10 and the dif-ferent experimental diets were all fed at the same time. The complete set of experiments was then repeated twice to bring the total to 30 animals per group. The only exceptions were the diets containing 10% corn oil and 20% lard. These were not included in the first feeding trial,and 20 rats were therefore allotted to each of these diets in a subsequent experiment.
The apparent digestibility of the different dietary fats was determined on groups of six rats during the course of one of the dietary experiments. The animals had been on diet about three months when these experiments were caried out. Feces were collected daily for five consecutive days and pooled for each group on a daily basis. For extraction of fecal lipids (20),2g of dry feces were ground in a mortar and transferred to 100 ml centrifuge tubes with 10 ml of 96% ethanol. To the mixture was added four to six drops of concentrated HCI and 2 ml of water and the lipids were extracted with two portions of 40 ml petroleum ether (30-60 C). The combined extract was transfer-red to a weighed beaker and dried,first under nitrogen and then overnight in a desiccator over Os. The Coefficient of Apparent Digesta-bility was calculated as fat ingested-fat excreted
fat ingested ested- fat excreted x 100.
RESULTS
Effect of Different Levels of Dietary Fat
on Mammary Tumor Incidence
The cumulative palpable mammary tumor incidence in female rats fed diets containing different levels of corn oil is shown in Figure 1 and the results at autopsy are summarized in Table III. The groups on diets containing 10% or 20% by weight of corn oil gave similar results and the tumor incidence was higher than with diets containing 0.5% or 5% corn oil. The dif-ference between the groups on 10% corn oil and 0.5% corn oil was significant (P<0.05)and the group on 10% corn oil had significantly more tumors per rat than either of the low fat groups (P<0.05). Most of the tumors were adenocarcinomas and the differing incidence on high and low fat diets was seen only in this type of tumor(Table IV).
Effect of Different Types of Dietary Fat on Mammary Tumor Incidence
0.06
9.9L
0.0L
Table V shows the effect of different dietary fats, fed at a level of 20% by weight of the diet, on incidence of mammary tumors induced by DMBA. More than 85% of the animals developed tumors in all groups except those on
9.9T6.6EZ
SS.0
FIG.1.Effect of diets containing different levels of corn oil on the cumulative palpable mammary tumor incidence in rats treated with DMBA. 20% corn oil diet; 10%;→5%;· 0.5%.
LIPIDS, VOL.6, NO.6
K.K.CARROLL AND H.T.KHOR
TABLE IV
Incidence of Different Types of Mammary Tumors in Rats
on Semisynthetic Diets Containing Different Levels of Corn Oil
Adenocarcinomas
Corn oil
diets Palpable Nonpalpabl e Total Palpable Nonpalpa ble Total Palpable Nonpalpab le Total tallow and rapeseed oil. However, there tended to be more tumors per rat when unsaturated fats were fed and this is reflected in the tend toward higher tumor yields with increasing un-saturation in the dietary fat (Table VI).Rape-seed oil is the most obvious exception, and the low yield with this oil may be related to its high content of monounsaturated C2o and C22 fatty acids (Table II). As before, most of the tumors were adenocarcinomas and the higher yield on diets containing unsaturated fats was confined to this type of tumor (Table VI).
DISCUSSION
The results of these experiments confirm our previous findings (7,13) that young female Sprague-Dawley rats treated with DMBA develop more mammary carcinomas on a semi-synthetic diet containing 20% corn oil than on a comparable diet containing only 0.5% corn oil. The tumor yield was not increased appre-ciably by raising the level of corn oil from 0.5% to 5%, but a further increase to 10% gave a tumor yield similar to that obtained with the 20% corn oil diet (Fig. 1, Tables III and IV). These findings are in general agreement with the results of Silverstone and Tannenbaum(3), who investigated the effect of diets containing different levels of partially hydrogenated cot-tonseed-soybean oil on the incidence of spontaneous mammary tumors in mice.They found that the tumor incidence increased as the level of fat in the diet was raised from 2-4% up to 12-16%, but beyond this level additional fat seemed to have little effect. Both in our experi-ments and in those of Silverstone and Tannen-baum, about 90% of the animals developed tumors when the diet contained 10-12% fat and it might be argued that there is not much room for further increase at higher levels. A smaller dose of DMBA could have been used in our studies to give a lower tumor incidence, but the 5 mg dose was chosen because it appeared to give the best differential in tumor yield between rats on high and low fat diets(13). It may also be noted that the number of tumors per rat showed a significant increase as the level of corn oil was raised from 5% to 10%, but no further increase in going from 10% to 20% (Table III).
Our initial studies on dietary fat in relation to incidence of DMBA-induced mammary cancer in rats indicated that coconut oil had much less effect on tumor yield than corn oil (7). Other workers have shown that edible oils such as olive oil (4), Crisco (5), and lard(6)can enhance mammary carcinogenesis in the rat, but no systematic studies have been carried out with different dietary fats. Our investigation was therefore expanded to include a variety of other edible fats and oils. The results indicated that, in general, rats on diets containing unsatu-rated fats developed more adenocarcinomas after treatment with DMBA than rats on similar diets containing saturated fats (Table VI).In fact,the tumor yields with dietary fats such as coconut oil and tallow were much the same as those obtained with low fat diet. (Table IV). However, although the total yield was lower with saturated fats, in most cases the per-centage of animals developing tumors was about the same as for unsaturated fats. The observed differences were thus due mainly to differences in the number of tumors per rat (Table V).
Rapeseed oil was exceptional in giving a low tumor yield although it contains a relatively high proportion of unsaturated fatty acids. Much of the unsaturation, however, is accounted for by the C2o and C22 monoenes, eicosenoic acid and erucic acid (Table II).Rape-seed oil has a lower coefficient of digestibility than most other dietary fats and oils (21)and is known to depress growth in rats (22) and to increase the concentration of cholesterol in adrenals and ovaries (23). Possibly the low tumor yield is related to some of these effects.
In the present experiments, the animals in the rapeseed oil group had the lowest average weight at autopsy although it was not much below that of other dietary groups (Table V). The apparent digestibility was also lower than for any other fat except tallow (Table V), but not as low as the value reported in the literature K.K.CARROLL AND H.T.KHOR
(24).Erucic acid appears to be responsible for most of the observed effects of rapeseed oil in animals (22,25) and the relatively small effects on growth and digestibility in the present experiments may be due to the use of an oil containing a lower percentage of erucic acid (16) or to feeding the oil at a lower level than those used in earlier studies.
Lard appeared to be more effective than other solid fats tested and olive oil, which con-tains oleic acid as the major unsaturated fatty acid (Table II), gave a tumor yield comparable to that of oils with a high content of linoleic acid (Table VI). The report of Harman (26) that increasing the level of a-tocopherol in a diet containing 20% corn oil decreased the tumor yield in rats treated with DMBA suggests that lipoperoxidation may be a factor.A recent study by Poling et al. (27) indicated that the incidence of spontaneous mammary tumors in rats fed heat-treated fats was much the same as in rats fed the untreated fats, but the peroxide values of the fats were little changed by the heating procedure. Our own studies have indi-cated that the effect is probably related to development rather than initiation of the tumors and it seems possible that the distribu-tion and metabolism of steroid hormones may be altered by the level and type of fat in the diet.Further experiments are being carried out to investigate this possibility.
Examination of statistical data for human populations of different countries has disclosed a strong positive correlation between dietaryfat intake and age-adjusted mortality from breast cancer (2,8,9). The findings in experimental animals may not be applicable to humans, but considered in the light of this statistical evi-dence,they suggest the possibility that the inci-dence of breast cancer might be lowered by decreasing fat intake. The high fat diets used in our studies contain about the same level of fat (20% by weight = approximately 40% of total calories) as typical American diets,whereas a low-fat diet which was effective in decreasing the incidence of mammary tumors in rats is comparable in fat content (5% by weight = approximately 10% by calories) to diets in countries such as Japan where the death rate from breast cancer is much lower than in America. The fact that unsaturated fats appeared to enhance the yield of mammary tumors to a greater extent then saturated fats in the experiments with rats also suggests that caution should be exercised in recommending a
LIPIDS, VOL.6,NO.6 large-scale shift to more highly unsaturated fats in human diets.
ACKNOWLEDGMENTS
This work was supported by the National Cancer Institute of Canada. R.Rasmussen assisted with care and feeding of the animals; Anne Varga prepared his-tological sections.
REFERENCES
1. Tannenbaum,A.,and H. Silverstone,in“Cancer”Vol 1,Edited by R.W.Raven,Butterworth & Co., Ltd.,London,1957,p.306-334.
2. Carroll, K.K., E.B. Gammal and E.R. Plunkett, Canad. Med. Ass. J. 98:590-594 (1968).
3.Silverstone,H.,and A. Tannenbaum,Cancer Res. 10:448-453(1950).
4. Benson, J.,M. Lev and C.G. Grand, Ibid. 16:135-137(1956).
5.Dunning,W.F.,M.R.Curtis and M.E.Maun,Ibid. 9:354-361(1949).
6.Engel, R.W., and D.H. Copeland,Ibid. 11:180-183(1951).
7.Gammal,E.B.,K.K.Carroll and E.R. Plunkett, Ibid.27:1737-1742(1967).
8. Lea,A.J.,Lancet 2:332-333(1966).
9.Wynder, E.L.,Cancer 24:1235-1240(1969).
10.Hems,G.,Brit. J. Cancer 24:226-234(1970).
11.Huggins,C.,and N.C.Yang,Science 137:257-262 (1962).
12.Gammal,E.B.,K.K.Carroll and E.R. Plunkett, Cancer Res. 28:384-385 (1968).
13.Carroll,K.K., and H.T. Khor, Ibid. 30:2260-2264 (1970).
14.Luddy,F.E., R.A. Barford, S.F. Herb and P. Magidman, JAOCS 45:549-552 (1968).
15.Herb, S.F., P. Magidman and R.W. Riemen-schneider, Ibid. 37:127-129(1960).
16. Downey, R.K.,B.M.Craig and C.G.Youngs,Ibid. 46:121-123(1969).
17. Jellum, M.D., and R.E. Worthington, Ibid. 43:661-664 (1966).
18.Putt, E.D.,B.M.Craig and R.B. Carson,Ibid. 46:126-129(1969).
istry,"Third Edition, D.C. Heath & Co.,Boston, 1955,p.328.
20. Friedner, S., and S. Moberg, Clin. Chim.Acta 18:345-349(1967).
21.Deuel, Jr.,H.J.,"The Lipids,"Vol.2,Interscience Publishers Inc.,New York,1955,p.222.
22.Thomasson, H.J., and J. Boldingh, J. Nutr. 56:469-475(1955).
23. Carroll, K.K., and R.L. Noble,Endocrinology 51:476-486(1952).
24.Deuel,Jr.,H.J.,A.L.S.Cheng and M.G. More-house,J.Nutr.35:295-300(1948).
25.Carroll, K.K., Canad. J. Biochem. Physiol. 40:1115-1122(1962).
26. Harman, D.,Clin. Res. 17:125 (1969).
27.Poling,C.E.,E.Eagle,E.E. Rice, A.M.A.Durand and M.Fisher, Lipids 5:128-136(1970).
[Received January 5,1971]7,12-Dimethylbenz[a]anthracene