Abstract

Conjugated linoleic acid (CLA), a family of isomers of octadecadienoic acid, inhibits rat mammary carcinogenesis, angiogenesis, and lung metastasis from a transplantable mammary tumor. c9,t11-CLA, the predominant isomer in dairy products, and t10,c12-CLA, a component of CLA supplements, are equally effective. The objective of the current studies was to test the efficacy of these two CLA isomers in a clinically relevant breast cancer model. Transgenic mice over-expressing erbB2 in the mammary epithelium were fed control or 0.5% CLA-supplemented diets continuously from weaning. Unexpectedly, t10,c12-CLA stimulated lobular hyperplasia of the mammary epithelium and accelerated mammary tumor development, decreasing median tumor latency to 168 days of age compared with 256 and 270 days in the c9,t11-CLA and control groups, respectively. Metastasis was also increased by t10,c12-CLA, with percentage of tumor-bearing mice with lung metastasis 73, 14 and 31% in the t10,c12-CLA, c9,t11-CLA and control groups, respectively. A second study, in which CLA administration was initiated after puberty, confirmed the stimulatory effect of t10,c12-CLA on mammary tumor development and metastasis. Additionally, t10,c12-CLA, but not c9,t11-CLA, increased the size of the liver, heart, spleen and mammary lymph node. The effects of t10,c12-CLA were not specific to erbB2 transgenic mice, as t10,c12-CLA supplementation increased proliferation in the mammary epithelium of both wild-type FVB and FVB/erbB2 mice. Moreover, the number of terminal end buds, the mammary epithelial structures most sensitive to a carcinogenic insult, was increased 30-fold in FVB wild-type mice fed t10,c12-CLA. These data suggest that it would be prudent to avoid CLA supplements containing the t10,c12-CLA isomer. However, even though c9,t11-CLA was not efficacious in the erbB2 model, its ability to inhibit mammary tumor development in rat models suggests that it may have activity for prevention of some types of breast cancer.

Introduction

One in seven women in the USA is now expected to develop breast cancer in her lifetime. Although improvements in early diagnosis and treatment have resulted in a slight decrease in the breast cancer death rate over the last 10 years, the reality of the high incidence emphasizes the necessity for developing effective, non-toxic, prevention strategies. Animal models have supported the idea that breast cancer incidence may be greatly reduced by various naturally occurring or synthetic agents. One such naturally occurring agent is the fatty acid-conjugated linoleic acid (CLA), a term which refers to a collection of positional and geometric isomers of octadecadienoic acid with conjugated double bonds. A major source of naturally occurring CLA is dairy products and ruminant meat, where c9,t11-CLA is the predominant isomer ( 1 ). CLA can also be produced synthetically, and a mixture of synthetically produced t10,c12-CLA and c9,t11-CLA is currently available in health food stores. This CLA mixture is also being tested clinically for its efficacy in inducing fat loss, improving insulin sensitivity in type II diabetes, improving blood lipids and improving immune function ( 2–6 ).

A number of studies have demonstrated the efficacy of dietary CLA in inhibiting carcinogenesis at different organ sites (reviewed in ref. 7 ). The majority of this work has focused on the rat mammary gland, where CLA was shown to inhibit carcinogenesis induced by both indirect- (dimethylbenz[ a ]anthracene) and direct- ( N -nitroso- N -methyl nitrosourea) acting carcinogens. The effects of CLA were dose dependent in the range between 0.05 and 1% ( 8 , 9 ), with CLA being equally effective when provided in the form of triglyceride or free fatty acid ( 10 ). More recent studies have focused on the identification of the active CLA isomers. CLA-enriched butter fat, containing predominantly c9,t11-CLA, was shown to inhibit rat mammary carcinogenesis ( 11 ). Furthermore, we found that the c9,t11 and t10,c12 isomers of CLA were equally effective in inhibiting the development of N -nitroso- N -methyl nitrosourea-induced preneoplastic lesions and tumors in the rat mammary gland ( 12 ). Hubbard et al. ( 13 ) also reported that each isomer inhibited metastasis in a transplantable mouse mammary tumor model.

The objective of the current studies was to determine whether dietary CLA isomers inhibited mammary carcinogenesis in a model relevant to human breast cancer. Over-expression of erbB2/her2 is observed in 20–30% of human breast cancers, and is associated with cancer progression and a poor prognosis ( 14 ). We therefore chose a transgenic mouse model in which wild-type erbB2 is over-expressed in the mammary epithelium to evaluate the efficacy of each of the CLA isomers to prevent the development and metastatic progression of mammary cancer. Unexpectedly, we found that the t10,c12 isomer of CLA, which is readily available as a major component of CLA supplements, markedly stimulated mammary tumorigenesis and metastasis in this erbB2-over-expressing model.

Materials and methods

Animals and dietary treatments

FVB/N-Tg (MMTVneu)202Mul/J ( 15 ) and FVB/J female mice were obtained from Jackson Laboratories (Bar Harbor, ME). In the first tumorigenesis experiment (CLA from weaning), mice were randomized by weight into three groups of 30 mice per group and placed on a basal AIN-76A diet without or with 0.5% c9,t11-CLA or t10,c12-CLA ( 16 ) at 24 days of age. In the second tumorigenesis experiment (CLA after puberty), mice were randomized into three groups of 25 mice per group and placed on one of the above diets at 68–72 days of age. The diets were fed continuously throughout the experiment. Mice were euthanized when a tumor reached a size of 18–20 mm in the larger diameter. Mammary glands and tumors were removed and prepared for whole-mount analysis and/or fixed in formalin. Lungs of the euthanized mice were perfused with ∼1.5 ml India ink (15% in water) through the trachea. The lungs were then removed, rinsed in water and fixed in Fekete's solution [70% ethanol (v/v), 3.7% paraformaldehyde (v/v) and 0.75 M acetic acid]. For the experiments to compare proliferation and apoptosis in wild-type and transgenic mice, FVB/J (seven mice per group) or FVB/N-Tg (MMTVneu)202Mul/J (five mice per group) were fed control or 0.5% CLA-supplemented diets for 10 days starting at 70 days of age. Mice were maintained in microisolator cages in a temperature- and humidity-controlled environment with a 12 h light–dark cycle, and were given food and water ad libitum . Animals were housed in accordance with the standards set by the National Institutes of Health and the Roswell Park Cancer Institute Animal Care and Use Committee.

Dietary ingredients

The CLA isomers were obtained from Natural ASA (Hovdebygda, Norway) or Larodan Fine Chemicals (Malmö, Sweden); the purity of each isomer was >90%. The other diet ingredients were purchased from the following sources: casein and corn oil, Harlan Teklad (Madison, WI); dextrose, Federal Bakers Supply (Buffalo, NY); AIN-76A vitamin mix, AIN-76 mineral mix, alphacel and DL -methionine, MP Biomedicals (Irvine, CA) and choline bitartrate, Sigma (St Louis, MO).

Preparation of mammary gland whole mounts

Mammary glands four and five were removed in one piece and spread out on slides according to the original anatomical orientation. After air drying on the slide for 10–15 min, the mammary glands were fixed for 15–19 h in Carnoy's fixative [75% (v/v) ethanol and 25% (v/v) acetic acid], washed in 70% (v/v) ethanol and hydrated in H 2 O and then stained with 0.2% (w/v) carmine (Sigma Chemical Company) in 5% (w/v) aqueous AlK(SO 4 ) 2 ·12H 2 O for 18–48 h. Tissues were then dehydrated through 15 min incubations in 70% (v/v), 95% (v/v) and 100% ethanol, defatted in acetone for 1–1.5 h and cleared in xylene for at least 24 h before incubating in Histoclear (National Diagnostics, Atlanta, GA) for 15 min and cover slipping with Permount (Fisher Scientific, Fair Lawn, NJ). Whole mounts were photographed using an Olympus SZ-PT stereoscope or an Olympus BX-40 microscope with an attached Hitachi KP-D50 color digital camera.

Histology, immunohistochemistry and evaluation of metastasis

Paraffin sections of mammary glands and tumors were stained with hematoxylin and eosin (H and E) for evaluation of histology, and immunohistochemistry was performed using 2 or 4 μg/ml rabbit polyclonal erbB2 antibody (#2428, Novus Biologicals, Littleton, CO) or 1 μg/ml polyclonal Ki67 antibody (#500-170) (Novus Biologicals). The ApoTag® Plus Peroxidase In Situ Apoptosis Kit (Chemicon, Temecula, CA) was used on paraffin sections to detect apoptosis. To evaluate metastasis in the lung, the entire lung was sectioned, and mice were scored as either negative or positive for lung metastasis, which is visualized as white nodules on the black ink-stained lungs.

Statistics

Log rank analysis was used to statistically evaluate the Kaplan–Meier curves, with the Holm–Sidak method used to analyze multiple curves. Statistical differences in tumor multiplicity, survival time after development of a palpable tumor, body weight, organ sizes, terminal end bud number and percentage of Ki67- or TUNEL-positive nuclei were tested by one-way analysis of variance, with the Holm–Sidak method used for pairwise multiple comparisons. Differences in lung metastases were tested by chi-square analysis. A P value of <0.05 was considered statistically significant. Where shown, error bars are standard error of the mean.

Results

t10,c12-CLA stimulates mammary tumor development and lung metastasis in erbB2 transgenic mice

The effects of dietary CLA were tested in a model relevant to a significant proportion of breast cancer patients in order to determine if the inhibitory effects of the c9,t11 and t10,c12-CLA isomers could be translated to this group of women. Mice over-expressing wild-type erbB2 in the mammary epithelium (hereby termed ‘erbB2 transgenic mice’) were fed control or CLA-containing diets continuously from 24 days of age. Dietary c9,t11-CLA had no significant effect on the rate at which mammary tumors developed in erbB2 transgenic mice ( Figure 1A ), or on the number of tumors per mouse (data not shown) or on the metastasis to the lung ( Figure 1B ; P = 0.21, not significant). Unexpectedly, however, dietary t10,c12-CLA markedly accelerated mammary tumor development. Latency, or the age at which 50% of the mice had developed a tumor, was only 168 days in the t10,c12-CLA group compared with 270 days in the control ( Figure 1A , P < 0.001). Latency in the c9,t11-CLA group was 256 days (not significant, P = 0.40 compared with control). Tumor multiplicity was not altered by t10,c12-CLA (data not shown); however, lung metastasis was dramatically increased ( Figure 1B , P =0.003). This increased metastasis was not a result of an alteration in primary tumor burden, since all mice were euthanized with a similar tumor size.

Fig. 1.

Mammary tumorigenesis, lung metastasis and body weight of erbB2 transgenic mice fed control, c9,t11-CLA- or t10,c12-CLA-supplemented diets from 24 days of age until being killed. ( A ) Kaplan–Meier plot showing percentage of tumor-free mice in each group at various ages. Of the initial 30 mice in each group, one mouse in the control group and one in the c9,t11-CLA group were killed early due to causes unrelated to the experimental protocol. These mice are not included in the figure. The remaining mice in the control group (29 mice) and all mice in the t10,c12-CLA group (30 mice) were killed with tumor. One mouse in the c9,t11-CLA group was killed without tumor at the end of the experiment; 28 mice in the c9,t11-CLA group were killed with tumor. The control and the c9,t11-CLA groups are each significantly different from the t10,c12-CLA group; there is no statistically significant difference between the control and c9,t11-CLA groups. ( B ) Percentage of tumor-bearing mice with lung metastasis. The numbers above each bar indicate the number of tumor-bearing mice (denominator) with lung metastasis (numerator) in each group. The control and c9,t11-CLA groups are each statistically different from the t10,c12-CLA group. ( C ) Body weights of erbB2 transgenic mice in the three dietary groups. Each point represents the mean ± SEM.

Fig. 1.

Mammary tumorigenesis, lung metastasis and body weight of erbB2 transgenic mice fed control, c9,t11-CLA- or t10,c12-CLA-supplemented diets from 24 days of age until being killed. ( A ) Kaplan–Meier plot showing percentage of tumor-free mice in each group at various ages. Of the initial 30 mice in each group, one mouse in the control group and one in the c9,t11-CLA group were killed early due to causes unrelated to the experimental protocol. These mice are not included in the figure. The remaining mice in the control group (29 mice) and all mice in the t10,c12-CLA group (30 mice) were killed with tumor. One mouse in the c9,t11-CLA group was killed without tumor at the end of the experiment; 28 mice in the c9,t11-CLA group were killed with tumor. The control and the c9,t11-CLA groups are each significantly different from the t10,c12-CLA group; there is no statistically significant difference between the control and c9,t11-CLA groups. ( B ) Percentage of tumor-bearing mice with lung metastasis. The numbers above each bar indicate the number of tumor-bearing mice (denominator) with lung metastasis (numerator) in each group. The control and c9,t11-CLA groups are each statistically different from the t10,c12-CLA group. ( C ) Body weights of erbB2 transgenic mice in the three dietary groups. Each point represents the mean ± SEM.

In spite of the earlier time of tumor appearance and increased metastasis in the t10,c12-CLA group, the tumors did not grow more rapidly in this group. To evaluate this, we determined the survival time of mice in each group, defined as the time between detection of a palpable tumor (2–3 mm) and the time of kill when a tumor reached 18–20 mm in the longer diameter. Mice fed the control, c9,t11- or t10,c12-CLA-supplemented diets were killed 51.7 ± 3.8, 48.6 ± 3.5 and 63.7 ± 4.2 days, respectively, after initial detection of a palpable, measurable tumor. This modest increase in survival time once a palpable tumor had been detected (control versus t10,c12-CLA, P =0.029) suggests that t10,c12-CLA may have been exerting a slight inhibitory effect on tumor growth during this time period.

The increased tumor development in the t10,c12-CLA-fed mice was not secondary to an increase in body weight since mice in this group actually weighed less than those in the control group ( Figure 1C , P < 0.01). Food intake in the t10,c12-CLA-fed group was slightly, but not significantly, increased compared with the other two groups (data not shown).

t10,c12-CLA alters development of the mammary epithelium but does not alter expression or localization of erbB2

In association with the accelerated tumor development, t10,c12-CLA had a marked effect on the mammary epithelium. Most striking, as seen from the mammary whole mounts ( Figure 2A–D ), was a precocious alveolar-like branching and hyperplasia that was continuous along the whole ductal network in this dietary group. This contrasts with the more normal-appearing ductal network in the control and c9,t11-CLA groups of erbB2 transgenic mice. Additionally, in the t10,c12-CLA group, the ducts were very dilated ( Figure 2D ) compared with the control or c9,t11-CLA groups ( Figure 2B and C ), and the epithelium did not grow to the boundary of the mammary fat pad ( Figure 2A ; the numbers under the whole mounts refer to the length of the epithelium from the lymph node). It can also be seen that mammary glands from t10,c12-CLA-fed mice were much smaller ( Figure 2A ), and on viewing the H- and E-stained cross-sections ( Figure 3A ), a markedly reduced adipocyte content [both brown and white adipose tissue] and an increase in fibrocellular stroma (arrows) were noted. No differences were apparent between the control and c9,t11-CLA groups ( Figures 2 and 3 ). Finally, as observed in Figure 2A , the intra-mammary lymph node was significantly larger in the mammary glands from mice fed the t10,c12-CLA-supplemented diet; the lengths were 2.27 ± 0.06, 2.19 ± 0.05 and 3.74 ± 0.13 mm ( n = 16–17) in the control, c9,t11-CLA and t10,c12-CLA groups, respectively. Together with the increased size of the spleen in the t10,c12-CLA group (see below), this suggests that this isomer may cause both systemic and local alterations in lymphoid function.

Fig. 2.

Mammary gland whole mounts from erbB2 transgenic mice demonstrate a marked inhibition of epithelial ductal elongation concurrent with lobular hyperplasia in mice fed t10,c12-CLA- but not in mice fed c9,t11-CLA-supplemented diets from weaning to time of kill. In all the t10,c12-CLA-fed mice, the epithelium failed to grow to the end of the fat pad, but did undergo significant hyperplasia and lobular differentiation. The numbers under the whole-mount photos are the mean ± SEM ( n = 16–17) of the epithelial length from the mammary lymph node, as indicated by the asterisk beside the control whole mount. Whole-mount photos ( A ) were taken under the ×1 objective. Magnifications of the whole mounts ( B , C and D , both upper and lower panels) were each taken under the ×4 objective and show normal ducts with some budding in the control (B) and c9,t11-CLA (C) groups, and dilated ducts with extensive lobular hyperplasia in the t10,c12-CLA group (D).

Fig. 2.

Mammary gland whole mounts from erbB2 transgenic mice demonstrate a marked inhibition of epithelial ductal elongation concurrent with lobular hyperplasia in mice fed t10,c12-CLA- but not in mice fed c9,t11-CLA-supplemented diets from weaning to time of kill. In all the t10,c12-CLA-fed mice, the epithelium failed to grow to the end of the fat pad, but did undergo significant hyperplasia and lobular differentiation. The numbers under the whole-mount photos are the mean ± SEM ( n = 16–17) of the epithelial length from the mammary lymph node, as indicated by the asterisk beside the control whole mount. Whole-mount photos ( A ) were taken under the ×1 objective. Magnifications of the whole mounts ( B , C and D , both upper and lower panels) were each taken under the ×4 objective and show normal ducts with some budding in the control (B) and c9,t11-CLA (C) groups, and dilated ducts with extensive lobular hyperplasia in the t10,c12-CLA group (D).

Fig. 3.

Mammary gland histology is dramatically altered in erbB2 transgenic mice fed t10,c12-CLA, but neither CLA isomer alters expression or localization of erbB2 protein within the mammary epithelium. ( A ) H and E staining of mammary glands from mice fed control (i) or CLA-supplemented (ii, c9,t11-CLA; iii, t10,c12-CLA) diets from weaning to time of kill. Photos are representative of 19–20 H and E sections per group, and were taken under the ×10 objective. ( B ) Immunohistochemical detection of erbB2 (brown staining) in ducts (i–iii) and lobules (iv–vi) of mice fed control (i and iv), c9,t11-CLA- (ii and v) or t10,c12-CLA- (iii and vi) supplemented diets. Immunohistochemistry was performed on mammary glands of seven mice per group and the photos shown are representative of all fields on each of the seven slides per group. All slides were counterstained with hematoxylin (blue staining). Photos were taken under the ×20 objective. WAT, white adipose tissue; BAT, brown adipose tissue. The black arrows in the t10,c12-CLA group point to fibroblastic stroma. The green arrow in the control group points to BAT.

Fig. 3.

Mammary gland histology is dramatically altered in erbB2 transgenic mice fed t10,c12-CLA, but neither CLA isomer alters expression or localization of erbB2 protein within the mammary epithelium. ( A ) H and E staining of mammary glands from mice fed control (i) or CLA-supplemented (ii, c9,t11-CLA; iii, t10,c12-CLA) diets from weaning to time of kill. Photos are representative of 19–20 H and E sections per group, and were taken under the ×10 objective. ( B ) Immunohistochemical detection of erbB2 (brown staining) in ducts (i–iii) and lobules (iv–vi) of mice fed control (i and iv), c9,t11-CLA- (ii and v) or t10,c12-CLA- (iii and vi) supplemented diets. Immunohistochemistry was performed on mammary glands of seven mice per group and the photos shown are representative of all fields on each of the seven slides per group. All slides were counterstained with hematoxylin (blue staining). Photos were taken under the ×20 objective. WAT, white adipose tissue; BAT, brown adipose tissue. The black arrows in the t10,c12-CLA group point to fibroblastic stroma. The green arrow in the control group points to BAT.

We considered the possibility that the increased tumorigenesis in mice fed the diet supplemented with t10,c12-CLA was a result of an increased expression of erbB2. However, immunohistochemical detection of erbB2 in mammary glands from each of the groups demonstrated no changes in expression or subcellular localization of erbB2 ( Figure 3B ). Most cells within the mammary ducts and lobules expressed erbB2 along the lateral, basal and apical membranes, and no differences in staining intensity were noted. Compared with ducts and lobular ductules, erbB2 expression in tumors was highly heterogeneous and not significantly different among the groups (data not shown). These data suggest that the effects of t10,c12-CLA were not a result of changes in erbB2.

t10,c12-CLA increased the size of the liver, heart and spleen

In addition to stimulating mammary tumorigenesis, absolute as well as relative weights of the liver, heart and spleen were significantly increased in mice fed the diet supplemented with t10,c12-CLA ( Table I ). No differences were noted between the control and c9,t11-CLA groups. When we examined H and E sections of the spleen, an increase in the size of the white pulp was observed in the t10,c12-CLA group, and could account for the majority of the increased spleen size in this group. In particular, the spleens of mice fed t10,c12-CLA showed an enlargement of follicular areas, which was due to an increase in primary follicles rather than germinal centers, which were noticeably lacking. The expansion of the white pulp in the absence of germinal center development, which was abundant in control and c9,t11-CLA-fed mice, suggests that t10,c12-CLA may suppress the development of a secondary immune response in these tumor-bearing mice. Finally, the livers in the t10,c12-CLA group were much lighter in color, and upon H and E staining, liver epithelium was heavily vacuolated in appearance, indicative of fatty liver (data not shown).

Table I.

Effect of long-term dietary CLA on mouse organ and body weights

Variable Control diet c9,t11-CLA diet t10,c12-CLA diet 
Body weight (g)  35.4 ± 1.0 a  34.0 ± 1.0 a  29.1 ± 0.7 b 
Liver weight (g)  1.53 ± 0.04 a  1.57 ± 0.04 a  2.59 ± 0.11 b 
Liver weight (g/100 g body wt)  4.37 ± 0.12 a  4.66 ± 0.11 a  8.89 ± 0.30 b 
Heart weight (mg)  132 ± 4 a  129 ± 3 a  177 ± 7 b 
Heart weight (mg/100 g body wt)  374 ± 11 a  385 ± 14 a  615 ± 18 b 
Spleen weight (mg)  149 ± 9 a  146 ± 8 a  252 ± 14 b 
Spleen weight (mg/100 g body wt)  423 ± 23 a  437 ± 28 a  858 ± 35 b 
Variable Control diet c9,t11-CLA diet t10,c12-CLA diet 
Body weight (g)  35.4 ± 1.0 a  34.0 ± 1.0 a  29.1 ± 0.7 b 
Liver weight (g)  1.53 ± 0.04 a  1.57 ± 0.04 a  2.59 ± 0.11 b 
Liver weight (g/100 g body wt)  4.37 ± 0.12 a  4.66 ± 0.11 a  8.89 ± 0.30 b 
Heart weight (mg)  132 ± 4 a  129 ± 3 a  177 ± 7 b 
Heart weight (mg/100 g body wt)  374 ± 11 a  385 ± 14 a  615 ± 18 b 
Spleen weight (mg)  149 ± 9 a  146 ± 8 a  252 ± 14 b 
Spleen weight (mg/100 g body wt)  423 ± 23 a  437 ± 28 a  858 ± 35 b 

Mean ± SEM. Means in a row without a common letter are statistically different. N = 28–29, 26–28 and 29–30 for control, c9,t11-CLA and t10,c12-CLA groups, respectively. Tissues were obtained from the transgenic erbB2 mice used in the first tumorigenesis study, in which the control or CLA-supplemented diets were fed from weaning.

Delaying CLA feeding until after puberty does not abrogate the stimulatory effect of t10,c12-CLA on mammary tumorigenesis

To eliminate a potential confounding effect of t10,c12-CLA on the mammary epithelium during pubertal development, a second experiment was performed in which erbB2 transgenic mice were fed control or CLA-supplemented diets starting after puberty at 68–72 days of age. Under these conditions, a statistically significant stimulatory effect of t10,c12-CLA on mammary tumorigenesis and lung metastasis was again seen ( Figure 4 ). Median times of mammary tumor development in this study were 271, 263 and 207 days of age in the control, c9,t11-CLA and t10,c12-CLA groups, respectively ( P < 0.001 for t10,c12-CLA versus control or c9,t11-CLA). Once a palpable tumor was detected, survival times in each of the groups were 39.6 ± 2.7, 40.9 ± 2.6 and 42.5 ± 2.8 days, respectively (not statistically different), suggesting comparable tumor growth rates in each of the dietary groups. Tumor multiplicity was similar in all three groups (data not shown).

Fig. 4.

Diets supplemented with t10,c12-CLA after puberty, but not with c9,t11-CLA, accelerate mammary tumor development and increase lung metastasis of erbB2 transgenic mice. Diets were initiated at 68–72 days of age, and fed continuously. ( A ) Kaplan–Meier plot showing percentage of tumor-free mice in each group at various ages. Of the initial 25 mice in each group, one mouse in the control group, four in the c9,t11-CLA group and two in the t10,c12-CLA group were killed early due to causes unrelated to the experimental protocol. These mice are not included in the figure. The remaining mice in the control (24 mice), c9,t11-CLA (21 mice) and t10,c12-CLA (23 mice) groups were killed with tumor. The control and the c9,t11-CLA groups are each significantly different from the t10,c12-CLA group; there is no statistically significant difference between the control and c9,t11-CLA groups. ( B ) Percentage of tumor-bearing mice with lung metastasis. The numbers above each bar indicate the number of tumor-bearing mice (denominator) with lung metastasis (numerator) in each group. Evaluation of metastasis could not be performed in one control mouse and one t10,c12-CLA mouse. The control and c9,t11-CLA groups are each statistically different from the t10,c12-CLA group.

Fig. 4.

Diets supplemented with t10,c12-CLA after puberty, but not with c9,t11-CLA, accelerate mammary tumor development and increase lung metastasis of erbB2 transgenic mice. Diets were initiated at 68–72 days of age, and fed continuously. ( A ) Kaplan–Meier plot showing percentage of tumor-free mice in each group at various ages. Of the initial 25 mice in each group, one mouse in the control group, four in the c9,t11-CLA group and two in the t10,c12-CLA group were killed early due to causes unrelated to the experimental protocol. These mice are not included in the figure. The remaining mice in the control (24 mice), c9,t11-CLA (21 mice) and t10,c12-CLA (23 mice) groups were killed with tumor. The control and the c9,t11-CLA groups are each significantly different from the t10,c12-CLA group; there is no statistically significant difference between the control and c9,t11-CLA groups. ( B ) Percentage of tumor-bearing mice with lung metastasis. The numbers above each bar indicate the number of tumor-bearing mice (denominator) with lung metastasis (numerator) in each group. Evaluation of metastasis could not be performed in one control mouse and one t10,c12-CLA mouse. The control and c9,t11-CLA groups are each statistically different from the t10,c12-CLA group.

As was observed in the first experiment, mammary glands were smaller in mice fed t10,c12-CLA ( Figure 5 ), and length of the epithelium from the lymph node toward the fat pad boundary was significantly decreased from 19.5 ± 0.7 and 19.1 ± 1.0 mm in the control and c9,t11-CLA groups, respectively, to 12.6 ± 1.0 mm in the t10,c12-CLA group. Since dietary CLA was not initiated until after puberty, this suggests that t10,c12-CLA may have induced apoptosis of the epithelium, as we demonstrated previously in vitro , as well as in pre-malignant lesions in the rat mammary gland in vivo ( 17 , 18 ). Extensive ductal and alveolar hyperplasia with dilated ducts, and apparent vacuole formation (arrows), was also seen in mammary glands of mice fed the diet supplemented with t10,c12-CLA, but not in those fed the control or c9,t11-CLA diets ( Figure 5 ).

Fig. 5.

Mammary gland whole mounts from erbB2 transgenic mice fed the control, c9,t11-CLA- and t10,c12-CLA-supplemented diets. Diets were initiated at 68–72 days of age, and fed continuously (same experiment as shown in Figure 4 ). Note the extensive budding and alveolar hyperplasia from the ducts in the t10,c12-CLA group. The arrows point to vacuole-like structures within the ducts from the t10,c12-CLA group, and the arrowhead to an alveolar lumen. Photos were taken from mammary whole mounts using the ×1 (left side) or ×4 (right side) microscope objectives.

Fig. 5.

Mammary gland whole mounts from erbB2 transgenic mice fed the control, c9,t11-CLA- and t10,c12-CLA-supplemented diets. Diets were initiated at 68–72 days of age, and fed continuously (same experiment as shown in Figure 4 ). Note the extensive budding and alveolar hyperplasia from the ducts in the t10,c12-CLA group. The arrows point to vacuole-like structures within the ducts from the t10,c12-CLA group, and the arrowhead to an alveolar lumen. Photos were taken from mammary whole mounts using the ×1 (left side) or ×4 (right side) microscope objectives.

Dietary t10,c12-CLA stimulates proliferation of the mammary epithelium of both FVB/erbB2 transgenic and FVB wild-type mice

To evaluate whether erbB2 over-expression was necessary for the stimulatory effect of t10,c12-CLA, a short-term experiment was undertaken in which the effects of CLA on the Ki67 labeling index were compared in the mammary epithelium of FVB/erbB2 transgenic and FVB wild-type mice. Ki67 is a proliferation marker that is expressed throughout the active phases of the cell cycle, but not in G 0 , and it has been shown to be an independent prognostic indicator in breast cancer ( 19 ). Mice were fed the control or CLA-containing diets for 10 days, starting at 70 days of age, the time at which CLA supplementation was initiated in the second mammary tumorigenesis study. The 10 day feeding period was chosen based on our earlier studies demonstrating that t10,c12-CLA induced changes in the mammary stroma within 3 days ( 16 ). Table II shows that t10,c12-CLA significantly increased the Ki67 labeling index in the terminal end buds, ducts and lobules of the mammary epithelium in both FVB wild-type and FVB/erbB2 mice, although the effect was more dramatic in the transgenic mice. Moreover, extensive budding of the epithelium and dilatation of the ducts were observed after only 10 days of feeding the FVB wild-type mice the t10,c12-CLA-supplemented diet, and the number of terminal end buds was markedly increased ( Figure 6 ). Apoptosis was low in both wild-type and transgenic mice, and in the wild-type mice was not significantly altered by either CLA isomer ( Table II ). A significant increase in apoptosis was observed in the FVB/erbB2 transgenic mice fed the t10,c12-CLA-supplemented diet; however, the percentage of cells undergoing apoptosis in this group was considerably less than those undergoing proliferation ( Table II ).

Table II.

Effect of CLA feeding for 10 days on proliferation and apoptosis in MECs from FVB wild-type and FVB/erbB2 transgenic mice

Mouse strain Mammary structure  Variable * Control diet c9,t11-CLA diet t10,c12-CLA diet 
FVB TEB Ki67  13.6 ± 10.7 a  18.2 ± 6.4 a  45.8 ± 1.9 b 
FVB Ducts Ki67  19.0 ± 1.7 a  5.4 ± 1.3 b  31.6 ± 1.7 c 
FVB Lobules Ki67  15.5 ± 2.4 a  8.7 ± 2.4 a  38.4 ± 1.7 b 
FVB TEB TUNEL  1.22 ± 0.51 a  0.83 ± 0.43 a  3.04 ± 0.54 a 
FVB Ducts TUNEL  1.22 ± 0.28 a  0.73 ± 0.22 a  2.50 ± 0.65 a 
FVB Lobules TUNEL  2.52 ± 0.76 a  1.03 ± 0.32 a  1.93 ± 0.56 a 
FVB/erbB2 TEB Ki67  2.74 ± 0.74 a  2.56 ± 0.46 a  38.8 ± 3.3 b 
FVB/erbB2 Ducts Ki67  7.17 ± 1.36 a  4.32 ± 0.85 a  22.4 ± 2.1 b 
FVB/erbB2 Lobules Ki67  3.06 ± 0.55 a  5.24 ± 0.87 a  26.0 ± 2.0 b 
FVB/erbB2 TEB TUNEL  0.51 a ± 0.18   0.43 ± 0.15 a  2.39 ± 0.39 b 
FVB/erbB2 Ducts TUNEL  0.35 ± 0.15 a  0.55 ± 0.21 a  3.47 ± 0.58 b 
FVB/erbB2 Lobules TUNEL  0.89 ± 0.22 a  0.08 ± 0.06 b  2.41 ± 0.42 c 
Mouse strain Mammary structure  Variable * Control diet c9,t11-CLA diet t10,c12-CLA diet 
FVB TEB Ki67  13.6 ± 10.7 a  18.2 ± 6.4 a  45.8 ± 1.9 b 
FVB Ducts Ki67  19.0 ± 1.7 a  5.4 ± 1.3 b  31.6 ± 1.7 c 
FVB Lobules Ki67  15.5 ± 2.4 a  8.7 ± 2.4 a  38.4 ± 1.7 b 
FVB TEB TUNEL  1.22 ± 0.51 a  0.83 ± 0.43 a  3.04 ± 0.54 a 
FVB Ducts TUNEL  1.22 ± 0.28 a  0.73 ± 0.22 a  2.50 ± 0.65 a 
FVB Lobules TUNEL  2.52 ± 0.76 a  1.03 ± 0.32 a  1.93 ± 0.56 a 
FVB/erbB2 TEB Ki67  2.74 ± 0.74 a  2.56 ± 0.46 a  38.8 ± 3.3 b 
FVB/erbB2 Ducts Ki67  7.17 ± 1.36 a  4.32 ± 0.85 a  22.4 ± 2.1 b 
FVB/erbB2 Lobules Ki67  3.06 ± 0.55 a  5.24 ± 0.87 a  26.0 ± 2.0 b 
FVB/erbB2 TEB TUNEL  0.51 a ± 0.18   0.43 ± 0.15 a  2.39 ± 0.39 b 
FVB/erbB2 Ducts TUNEL  0.35 ± 0.15 a  0.55 ± 0.21 a  3.47 ± 0.58 b 
FVB/erbB2 Lobules TUNEL  0.89 ± 0.22 a  0.08 ± 0.06 b  2.41 ± 0.42 c 

Mean ± SEM. All mammary structures within each section were evaluated. Mice ( n = 7 per FVB group and n = 5 per FVB/erbB2 group) were fed control or CLA-supplemented diets for 10 days, starting at 70 days of age. For the control, c9,t11-CLA and t10,c12-CLA groups, respectively, the number of mammary structures counted were as follows. FVB mice (Ki67): TEBs, 5, 5, 84; ducts, 86, 88, 86 and lobules, 32, 20, 44. FVB/erbB2 mice (Ki67): TEBs, 60, 58, 57; ducts, 62, 60, 66 and lobules, 50, 53, 57. FVB mice (TUNEL): TEBs, 10, 17, 82; ducts, 87, 87, 89 and lobules, 17, 10, 37. FVB/erbB2 mice (TUNEL): TEBs, 59, 61, 60; ducts, 66, 71, 72 and lobules, 52, 43, 59. Means in a row without a common letter are statistically different.

*

Percentage of Ki67- or TUNEL-positive nuclei.

Fig. 6.

Mammary gland whole mounts from FVB wild-type mice fed the control, c9,t11-CLA- and t10,c12-CLA-supplemented diets for 10 days starting at 70 days of age. Extensive budding and dilated ducts can be seen in the t10,c12-CLA group. The numbers in each panel refer to the number of terminal end buds (mean ± SEM of seven mice per group). Each photo in the left panels of the three dietary groups was taken under the ×1 objective. Each photo in the right panels of the three dietary groups was taken under the ×4 objective.

Fig. 6.

Mammary gland whole mounts from FVB wild-type mice fed the control, c9,t11-CLA- and t10,c12-CLA-supplemented diets for 10 days starting at 70 days of age. Extensive budding and dilated ducts can be seen in the t10,c12-CLA group. The numbers in each panel refer to the number of terminal end buds (mean ± SEM of seven mice per group). Each photo in the left panels of the three dietary groups was taken under the ×1 objective. Each photo in the right panels of the three dietary groups was taken under the ×4 objective.

Discussion

Dietary CLA, fed as a mixed isomer preparation containing equal amounts of c9,t11-CLA and t10,c12-CLA, as well as the individual isomers, has been shown previously to inhibit carcinogen-induced rat mammary tumorigenesis, as well as the growth and/or metastases of mouse mammary tumors in syngeneic and xenograft models (reviewed in ref. 7 ). Together, these studies were strongly supportive of the notion that dietary CLA might have clinical application in inhibiting the development and metastatic spread of breast cancer, as well as reducing residual disease. However, results from the current experiments demonstrating that the t10,c12 isomer of CLA increases mammary tumor development and metastasis in mice over-expressing wild-type erbB2 in the mammary epithelium clearly dispel this notion for the t10,c12-CLA isomer. Moreover, our data suggest that the detrimental effect of t10,c12-CLA is independent of the time at which supplementation is initiated, since tumorigenesis was accelerated in both the prepubertal and postpubertal experimental protocols, with a mean time of tumor appearance 144 and 137 days, respectively, after the mice were started on the t10,c12-CLA-containing diets. The effect was isomer specific, since c9,t11-CLA did not alter tumorigenesis or metastasis.

Human breast cancer and erbB2/her2

ErbB2/her2 is naturally expressed in normal breast epithelium of humans ( 20 , 21 ), as well as in rat ( 22 ) and mouse ( 23 ) mammary glands. Over-expression of erbB2 is observed in 20–30% of human breast cancers, and is associated with a poor prognosis ( 14 , 24 , 25 ). This increased expression of erbB2 is seen in ductal carcinoma in situ , an early stage of breast cancer development, and interestingly its frequency is higher in ductal carcinoma in situ than in invasive breast cancers ( 26–30 ). No over-expression has been noted in hyperplastic or dysplastic lesions ( 28 ). Although our experiments examined the effect of dietary CLA in a model in which the majority of the mammary epithelial cells (MECs) over-express erbB2, and as a result could overestimate the effect of t10,c12-CLA in human breast cancer, the high frequency of erbB2 over-expression (40–77%) in ductal carcinoma in situ ( 26–30 ), together with the high incidence of breast cancer, suggests that a large population of women with clinical or preclinical disease may be susceptible to the deleterious effect of t10,c12-CLA.

Why does t10,c12-CLA stimulate mammary carcinogenesis in erbB2 transgenic mice?

Our observation that t10,c12-CLA accelerates mammary tumor development and increases metastasis, and yet does not alter survival time after a palpable tumor is detected, is something of a conundrum, but may reflect the multifaceted activity of this CLA isomer. For example, we demonstrated previously that in wild-type mice, t10,c12-CLA inhibits angiogenesis (as does c9,t11-CLA), concurrent with a decrease in the pro-angiogenic and tumor growth factors vascular endothelial growth factor and leptin ( 16 , 31 ). This could limit tumor expansion. On the other hand, the acceleration of tumor development, as well as the increased metastasis to the lung, suggests that t10,c12-CLA has modified the mammary epithelium and/or its stromal environment, resulting in an increased sensitivity to tumor development. This notion is supported by the dramatically altered morphological appearance of the mammary epithelium and stroma in mice fed the t10,c12-CLA-supplemented diet. Indeed, the extensive epithelial budding, which was continuous along the entire ductal network, suggested that the proliferative capacity of the epithelium was increased at least at early times after the mice received this CLA isomer. In fact, we found this to be the case, as the Ki67 labeling index was increased in the erbB2 transgenic mice after only 10 days of feeding t10,c12-CLA. This increased proliferation could account for the earlier time of tumor appearance in this group.

Unexpectedly, feeding of t10,c12-CLA for only 10 days also stimulated the Ki67 labeling index in the mammary epithelium of wild-type mice of the same strain and dramatically increased the number of terminal end buds, an epithelial structure known to be sensitive to carcinogens ( 32 ). This suggests that activation of erbB2 is not required for the stimulatory effect of t10,c12-CLA and that proliferative signals resulting from t10,c12-CLA supplementation and erbB2 over-expression may converge downstream to enhance tumor development. The mechanism by which this may occur is currently not known; however, two observations point to fruitful areas of future research. First, the intra-mammary lymph node, which is immediately adjacent to the epithelium, was visibly enlarged, and as a result, there may be changes in cytokine secretion that could impact growth of the epithelium. Second, the fibroblastic stroma surrounding the mammary ductal epithelium in the t10,c12-CLA-supplemented mice is reminiscent of the reactive stroma that surrounds breast tumors, and it is tempting to speculate that it plays an important role in modulating the tumorigenic response of the erbB2-over-expressing epithelium and the proliferative response of wild-type mouse epithelium. In support of this possibility, it is significant that t10,c12-CLA has no readily discernible effect on the rat mammary stroma ( 33 and P.Masso-Welch and M.M.Ip, unpublished data), and a mixture of CLA isomers containing approximately equal amounts of t10,c12-CLA and c9,t11-CLA inhibited, rather than stimulated rat mammary epithelial proliferation and carcinogenesis ( 33 ).

The pregnancy-like increase in lobular–alveolar development of the mammary epithelium in the t10,c12-CLA group of erbB2 mice was unexpected. A possible explanation is that the marked increased in the fibroblastic stroma surrounding the ducts ( Figure 3 ), which we also observed in wild-type CD2F1 mice ( 16 ), could result in an enhanced deposition of extracellular matrix, a known stimulator of alveolar differentiation ( 34 ). However, since contact with the extracellular matrix is required but not sufficient ( 35 , 36 ) for alveolar development, additional changes in the local or systemic hormonal or growth factor milieu would have to be postulated.

An intriguing consequence of the increased differentiation may be the induction of a MEC population that is functionally equivalent to the pregnancy-induced MEC sub-population remaining in the mammary gland after post-weaning involution of the epithelium ( 37 , 38 ). These parity-induced MECs are pluripotent and capable of self-renewal ( 38 ), and importantly, were shown to be the specific targets for mammary tumorigenesis in the mouse MMTV-erbB2 over-expression model ( 39 ). The relevance of this concept to human breast cancer should also be considered. Henry et al. ( 39 ) have pointed out that transgenic mouse models of the type used in our study may represent a good model for the transient increased risk of breast cancer after a full-term pregnancy. If our data were translatable to the human situation, this could imply that some women consuming t10,c12-CLA during and immediately after a full-term pregnancy might be at an increased risk of breast cancer, especially if they have pre-existing erbB2-over-expressing cells within the epithelium.

Is it safe for humans to consume t10,c12-CLA?

CLA is readily available in health food stores and supermarkets as a mixture of the c9,t11 and t10,c12 isomers, and thus the safety of each of these isomers has to be considered. From the observations reported in the current paper, together with other reports in the literature (reviewed in refs 40–42 ), it is evident that there are a number of reasons to make a strong recommendation against dietary supplements containing the t10,c12 isomer of CLA. Thus, in addition to the marked stimulatory effect of t10,c12-CLA on mammary carcinogenesis and metastasis in erbB2 transgenic mice, this isomer also modestly stimulated intestinal tumorigenesis in the adenomatous polyposis Coli min/+ mouse model ( 43 ). It is possible that the adverse effects are related to over-expression (erbB2) or mutation (APC); however, since these changes are not known prior to biopsy and diagnosis, we cannot in good conscience recommend t10,c12-CLA for cancer prevention. On the other hand, c9,t11-CLA, although not effective in erbB2 transgenic mice, has been shown to be both safe and effective in other breast cancer models.

Second, the increased size of several organs in mice fed t10,c12-CLA is a warning signal of potential problems with this isomer. Previous reports have demonstrated hepatic steatosis, as well as increased size of the liver and spleen in non-tumor-bearing mice fed t10,c12-CLA ( 44 , 45 ), so the changes we observed in the erbB2 transgenic mice are not secondary to tumor development. It is not known whether similar changes occur in humans. When a mixture of CLA isomers containing approximately equal amounts of c9,t11-CLA and t10,c12-CLA was given to healthy overweight or obese subjects for 12 or 24 months, one study (6 g CLA/day) reported no changes in serum levels of liver enzymes ( 3 ), although another study showed a very modest increase in serum aspartate aminotransferase in subjects receiving ∼3.5 g CLA per day ( 2 , 5 ). These studies also suggested a modest change in serum cholesterol profiles (a slight increase in low-density lipoprotein cholesterol and/or a slight decrease in high-density lipoprotein cholesterol), indicative of a possible change in liver function. Although it is not known if CLA induces splenomegaly in humans as it does in mice, a slight increase in circulating lymphocytes ( 2 ) and a decrease in infection rate ( 3 ) after long-term CLA administration suggests such changes may occur and have a biological consequence. Overall, however, there have been no consistent changes in immune function in a number of clinical studies ( 42 ).

Third, to our knowledge, ours is the first study to report an increase in the size of the heart. In contrast, 3 week supplementation with 1% t10,c12-CLA ( 45 ) or 5 month supplementation with 1% of a CLA mixture containing t10,c12-CLA ( 46 ) did not show any increase in mouse heart weight. Whether this difference reflects the timing of the supplement or the use of a purified isomer in our study remains to be determined. Nevertheless, the potential relevance of this observation to the human population should be considered. Finally, additional concerns have been raised in clinical studies, although not investigated in our current study. In men with metabolic syndrome, supplementation with t10,c12-CLA, but not with a CLA mixture containing c9,t11-CLA and t10,c12-CLA, increased lipid peroxidation, C-reactive protein, plasma insulin and glucose, and decreased insulin sensitivity ( 47 , 48 ). Moreover, a CLA mixture containing t10,c12-CLA decreased insulin sensitivity and increased fasting glucose levels in type II diabetics ( 6 ), suggesting another population in which CLA supplementation may be detrimental.

At this point, it is important to assess the dose of CLA used in our studies, with those used in clinical trials. Based on the assumption that a 25 g mouse consumes 4.4 g food per day, a mouse eating a diet supplemented with 0.5% CLA will consume 22 mg CLA per day. This is equivalent to 880 mg CLA/kg mouse per day, or when normalized to a human equivalent dose based on body surface area to 71.5 mg/kg body wt per day [to normalize to a human equivalent dose on a body surface area basis, the mouse dose in mg/kg body weight is divided by 12.3; this assumes a 60 kg human subject (http://www.fda.gov/cber/gdlns/dose.htm)]. In two long-term CLA human trials, daily intake of the t10,c12 component of the CLA mixture ranged from 1.476 ( 2 ) to 2.256 g ( 3 ), which, for a 60 kg human subject is equivalent to 24.6 and 37.6 mg/kg body wt, respectively, or approximately one half of the amount the mice in our study received.

In summary, the data presented in this paper show that a dose of the t10,c12 isomer of CLA that is readily achievable in women taking a CLA supplement could potentially have adverse effects in a clinically significant population of women, namely, those at risk of developing a breast cancer in which erbB2/her2 is over-expressed. Moreover, the ability of dietary t10,c12-CLA to stimulate proliferation of wild-type mammary epithelium suggests that the epithelium may have an increased sensitivity to a carcinogenic or oncogenic insult. Together with the data demonstrating that this isomer affects major organs, at least in mice, and as discussed above may be harmful in certain types of obesity and diabetes in humans, we believe it would be prudent to avoid supplements containing t10,c12-CLA. On the other hand, although c9,t11-CLA was not able to overcome the strong oncogenic stimulus from erbB2 over-expression in the model used here, based on its proven ability to inhibit mammary epithelial proliferation and tumorigenesis in the rat ( 11 ) and growth and metastasis of mammary tumors in other mouse models ( 7 ), together with its lack of toxicity, we believe that supplements containing c9,t11-CLA may be both safe and efficacious in breast cancer prevention.

Abbreviations

    Abbreviations
  • CLA

    conjugated linoleic acid

  • E

    eosin

  • H

    hematoxylin

  • MEC

    mammary epithelial cell

We gratefully acknowledge Dr Dale Bauman of the Department of Dairy Science at Cornell University (Ithaca, NY), Dr Andreas Menzel at Loders Croklaan (Wormerveer, The Netherlands), who independently confirmed the purity of the different lots of CLA used in our studies, and Mary Vaughan who performed the immunohistochemistry. These studies were supported by the Susan G.Komen Breast Cancer Foundation BCTR0201524 (M.M.I.), by National Institutes of Health CA61763 (C.I.), by an Avon-American Association for Cancer Research Scholar Award (X.M.) and by the shared resources of the National Cancer Institute Roswell Park Cancer Center Support Grant CA16056.

Conflict of Interest Statement : None declared.

References

1.
Parodi
PW
Cows' milk fat components as potential anticarcinogenic agents
J. Nutr.
 , 
1997
, vol. 
127
 (pg. 
1055
-
1060
)
2.
Gaullier
JM
, et al.  . 
Conjugated linoleic acid supplementation for 1 y reduces body fat mass in healthy overweight humans
Am. J. Clin. Nutr.
 , 
2004
, vol. 
79
 (pg. 
1118
-
1125
)
3.
Whigham
LD
, et al.  . 
Safety profile of conjugated linoleic acid in a 12-month trial in obese humans
Food Chem. Toxicol.
 , 
2004
, vol. 
42
 (pg. 
1701
-
1709
)
4.
Tricon
S
, et al.  . 
Effects of cis -9, trans -11 and trans -10, cis -12 conjugated linoleic acid on immune cell function in healthy humans
Am. J. Clin. Nutr.
 , 
2004
, vol. 
80
 (pg. 
1626
-
1633
)
5.
Gaullier
JM
, et al.  . 
Supplementation with conjugated linoleic acid for 24 months is well tolerated by and reduces body fat mass in healthy, overweight humans
J. Nutr.
 , 
2005
, vol. 
135
 (pg. 
778
-
784
)
6.
Moloney
F
, et al.  . 
Conjugated linoleic acid supplementation, insulin sensitivity, and lipoprotein metabolism in patients with type 2 diabetes mellitus
Am. J. Clin. Nutr.
 , 
2004
, vol. 
80
 (pg. 
887
-
895
)
7.
Ip
MM
, et al.  . 
Prevention of mammary cancer with conjugated linoleic acid: role of the stroma and the epithelium
J. Mammary Gland Biol. Neoplasia
 , 
2003
, vol. 
8
 (pg. 
101
-
116
)
8.
Ip
C
, et al.  . 
Mammary cancer prevention by conjugated dienoic derivative of linoleic acid
Cancer Res.
 , 
1991
, vol. 
51
 (pg. 
6118
-
6124
)
9.
Ip
C
, et al.  . 
Conjugated linoleic acid suppresses mammary carcinogenesis and proliferative activity of the mammary gland in the rat
Cancer Res.
 , 
1994
, vol. 
54
 (pg. 
1212
-
1215
)
10.
Ip
C
, et al.  . 
Effect of timing and duration of dietary conjugated linoleic acid on mammary cancer prevention
Nutr. Cancer
 , 
1995
, vol. 
24
 (pg. 
241
-
247
)
11.
Ip
C
, et al.  . 
Conjugated linoleic acid-enriched butter fat alters mammary gland morphogenesis and reduces cancer risk in rats
J. Nutr.
 , 
1999
, vol. 
129
 (pg. 
2135
-
2142
)
12.
Ip
C
, et al.  . 
Conjugated linoleic acid isomers and mammary cancer prevention
Nutr. Cancer
 , 
2002
, vol. 
43
 (pg. 
52
-
58
)
13.
Hubbard
NE
, et al.  . 
Effect of separate conjugated linoleic acid isomers on murine mammary tumorigenesis
Cancer Lett.
 , 
2003
, vol. 
190
 (pg. 
13
-
19
)
14.
Slamon
DJ
, et al.  . 
Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene
Science
 , 
1987
, vol. 
235
 (pg. 
177
-
182
)
15.
Guy
CT
, et al.  . 
Expression of the neu protooncogene in the mammary epithelium of transgenic mice induces metastatic disease
Proc. Natl Acad. Sci. USA
 , 
1992
, vol. 
89
 (pg. 
10578
-
10582
)
16.
Masso-Welch
PA
, et al.  . 
Isomers of conjugated linoleic acid differ in their effects on angiogenesis and survival of mouse mammary adipose vasculature
J. Nutr.
 , 
2004
, vol. 
134
 (pg. 
299
-
307
)
17.
Ip
MM
, et al.  . 
Conjugated linoleic acid inhibits proliferation and induces apoptosis of normal rat mammary epithelial cells in primary culture
Exp. Cell Res.
 , 
1999
, vol. 
250
 (pg. 
22
-
34
)
18.
Ip
C
, et al.  . 
Induction of apoptosis by conjugated linoleic acid in cultured mammary tumor cells and premalignant lesions of the rat mammary gland
Cancer Epidemiol. Biomarkers Prev.
 , 
2000
, vol. 
9
 (pg. 
689
-
696
)
19.
Scholzen
T
, et al.  . 
The Ki-67 protein: from the known and the unknown
J. Cell Physiol.
 , 
2000
, vol. 
182
 (pg. 
311
-
322
)
20.
Press
MF
, et al.  . 
Expression of the HER-2/neu proto-oncogene in normal human adult and fetal tissues
Oncogene
 , 
1990
, vol. 
5
 (pg. 
953
-
962
)
21.
Gompel
A
, et al.  . 
Epidermal growth factor receptor and c- erb B-2 expression in normal breast tissue during the menstrual cycle
Breast Cancer Res. Treat.
 , 
1996
, vol. 
38
 (pg. 
227
-
235
)
22.
Darcy
KM
, et al.  . 
Changes in ErbB2 (her-2/neu), ErbB3, and ErbB4 during growth, differentiation, and apoptosis of normal rat mammary epithelial cells
J. Histochem. Cytochem.
 , 
2000
, vol. 
48
 (pg. 
63
-
80
)
23.
DiAugustine
RP
, et al.  . 
EGF-related peptides and their receptors in mammary gland development
J. Mammary Gland Biol. Neoplasia
 , 
1997
, vol. 
2
 (pg. 
109
-
117
)
24.
Berger
MS
, et al.  . 
Correlation of c-erbB-2 gene amplification and protein expression in human breast carcinoma with nodal status and nuclear grading
Cancer Res.
 , 
1988
, vol. 
48
 (pg. 
1238
-
1243
)
25.
Slamon
DJ
, et al.  . 
Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer
Science
 , 
1989
, vol. 
244
 (pg. 
707
-
712
)
26.
Gusterson
BA
, et al.  . 
Immunohistochemical distribution of c-erbB-2 in infiltrating and in situ breast cancer
Int. J. Cancer
 , 
1988
, vol. 
42
 (pg. 
842
-
845
)
27.
Somerville
JE
, et al.  . 
c-erbB-2 overexpression and histological type of in situ and invasive breast carcinoma
J. Clin. Pathol.
 , 
1992
, vol. 
45
 (pg. 
16
-
20
)
28.
Allred
DC
, et al.  . 
Overexpression of HER-2/neu and its relationship with other prognostic factors change during the progression of in situ to invasive breast cancer
Hum. Pathol.
 , 
1992
, vol. 
23
 (pg. 
974
-
979
)
29.
Ho
GH
, et al.  . 
In microdissected ductal carcinoma in situ , HER-2/neu amplification, but not p53 mutation, is associated with high nuclear grade and comedo histology
Cancer
 , 
2000
, vol. 
89
 (pg. 
2153
-
2160
)
30.
Hoque
A
, et al.  . 
Her-2/neu gene amplification in ductal carcinoma in situ of the breast
Cancer Epidemiol. Biomarkers Prev.
 , 
2002
, vol. 
11
 (pg. 
587
-
590
)
31.
Masso-Welch
PA
, et al.  . 
Inhibition of angiogenesis by the cancer chemopreventive agent conjugated linoleic acid
Cancer Res.
 , 
2002
, vol. 
62
 (pg. 
4383
-
4389
)
32.
Russo
J
, et al.  . 
Experimentally induced mammary tumors in rats
Breast Cancer Res. Treat.
 , 
1996
, vol. 
39
 (pg. 
7
-
20
)
33.
Thompson
H
, et al.  . 
Morphological and biochemical status of the mammary gland as influenced by conjugated linoleic acid: implication for a reduction in mammary cancer risk
Cancer Res.
 , 
1997
, vol. 
57
 (pg. 
5067
-
5072
)
34.
Darcy
KM
, et al.  . 
Mammary organoids from immature virgin rats undergo ductal and alveolar morphogenesis when grown within a reconstituted basement membrane
Exp. Cell Res.
 , 
1991
, vol. 
196
 (pg. 
49
-
65
)
35.
Darcy
KM
, et al.  . 
Prolactin and epidermal growth factor regulation of the proliferation, morphogenesis and functional differentiation of normal rat mammary epithelial cells in three dimensional primary culture
J. Cell Physiol.
 , 
1995
, vol. 
163
 (pg. 
346
-
364
)
36.
Darcy
KM
, et al.  . 
Hydrocortisone and progesterone regulation of the proliferation, morphogenesis and functional differentiation of normal rat mammary epithelial cells in three dimensional primary culture
J. Cell Physiol.
 , 
1995
, vol. 
163
 (pg. 
365
-
379
)
37.
Wagner
KU
, et al.  . 
An adjunct mammary epithelial cell population in parous females: its role in functional adaptation and tissue renewal
Development
 , 
2002
, vol. 
129
 (pg. 
1377
-
1386
)
38.
Boulanger
CA
, et al.  . 
Parity-induced mouse mammary epithelial cells are pluripotent, self-renewing and sensitive to TGF-β1 expression
Oncogene
 , 
2005
, vol. 
24
 (pg. 
552
-
560
)
39.
Henry
MD
, et al.  . 
Parity-induced mammary epithelial cells facilitate tumorigenesis in MMTV-neu transgenic mice
Oncogene
 , 
2004
, vol. 
23
 (pg. 
6980
-
6985
)
40.
Larsen
TM
, et al.  . 
Efficacy and safety of dietary supplements containing CLA for the treatment of obesity: evidence from animal and human studies
J. Lipid Res.
 , 
2003
, vol. 
44
 (pg. 
2234
-
2241
)
41.
Terpstra
AH
Effect of conjugated linoleic acid on body composition and plasma lipids in humans: an overview of the literature
Am. J. Clin. Nutr.
 , 
2004
, vol. 
79
 (pg. 
352
-
361
)
42.
Tricon
S
, et al.  . 
Conjugated linoleic acid and human health: a critical evaluation of the evidence
Curr. Opin. Clin. Nutr. Metab. Care
 , 
2006
, vol. 
9
 (pg. 
105
-
110
)
43.
Rajakangas
J
, et al.  . 
Adenoma growth stimulation by the trans -10, cis -12 isomer of conjugated linoleic acid (CLA) is associated with changes in mucosal NF-kappaB and cyclin D1 protein levels in the Min mouse
J. Nutr.
 , 
2003
, vol. 
133
 (pg. 
1943
-
1948
)
44.
Clément
L
, et al.  . 
Dietary trans-10,cis-12 conjugated linoleic acid induces hyperinsulinemia and fatty liver in the mouse
J. Lipid Res.
 , 
2002
, vol. 
43
 (pg. 
1400
-
1409
)
45.
Yamasaki
M
, et al.  . 
Immunoglobulin and cytokine production from spleen lymphocytes is modulated in C57BL/6J mice by dietary cis-9, trans-11 and trans-10, cis-12 conjugated linoleic acid
J. Nutr.
 , 
2003
, vol. 
133
 (pg. 
784
-
788
)
46
Tsuboyama-Kasaoka
N
, et al.  . 
Conjugated linoleic acid supplementation reduces adipose tissue by apoptosis and develops lipodystrophy in mice
Diabetes
 , 
2000
, vol. 
49
 (pg. 
1534
-
1542
)
47.
Risérus
U
, et al.  . 
Supplementation with conjugated linoleic acid causes isomer-dependent oxidative stress and elevated C-reactive protein—a potential link to fatty acid-induced insulin resistance
Circulation
 , 
2002
, vol. 
106
 (pg. 
1925
-
1929
)
48.
Risérus
U
, et al.  . 
Treatment with dietary trans10,cis12 conjugated linoleic add causes isomer-specific insulin resistance in obese men with the metabolic syndrome
Diabetes Care
 , 
2002
, vol. 
25
 (pg. 
1516
-
1521
)