Effects of Soy Isoflavone and Endogenous Oestrogen on Breast Cancer in MMTV-erbB2 Transgenic Mice

Abstract

Objective:

Soy isoflavone is associated with modification of breast cancer risk. Effects of dietary isoflavone on breast tissue carcinogenesis under varying endogenous oestrogen contexts were investigated.

Methods:

Five-week-old mouse mammary tumour virus (MMTV)-erbB2 female transgenic mice (n = 180) were divided into three equal groups: low-, normal- and high-oestrogen groups. Each group was then subdivided into an experimental group (given soybean feed) and a control group (given control feed).

Results:

In the high-oestrogen environment, breast cancer incidence was significantly lower in the experimental versus the control group, whereas in the low-oestrogen environment, breast cancer incidence was significantly higher in the experimental versus the control group. There were no between-group differences in mean breast tumour latency, mean largest tumour diameter and breast tumour tissue vascular endothelial growth factor levels.

Conclusions:

Dietary soy isoflavones promote breast cancer at low oestrogen levels but inhibit breast cancer at high oestrogen levels. This effect may only occur during the initiation stage of breast cancer.

Introduction

Endogenous and exogenous factors are both associated with an increased risk of developing breast cancer; the most important factors are genetic predisposition, endogenous oestrogen levels and oestrogen exposure. The associations between oestrogen and risk of breast cancer are based on static analyses of human breast biopsies, population-based parity or hormonal exposure data, and randomized clinical trials using hormone treatments.^{1 – 4} Data suggest that exogenous oestrogens may have different effects on women under different endogenous oestrogen level conditions. ⁵

Soy isoflavones are a class of oestrogenlike compounds previously characterized as natural selective oestrogen receptor modulators. ⁶ Widely used as dietary supplements, ⁷ soy isoflavones have gained much attention in relation to the prevention and treatment of breast cancer, though their impact on breast cancer remains highly controversial. Epidemiological studies indicate that a high dietary intake of soy isoflavones may lower the risk of breast cancer.^{8 – 11} The effects of soy isoflavone in breast tissue are debatable, however, with evidence supporting both oestrogen agonist and antagonist properties.^12,13 Isoflavones elicit clear oestrogenic effects in low-oestrogen rodent and cell culture models^14,15 but may inhibit the effects of 17 β-oestradiol (E2) in systems with higher eostrogen levels.^{15 – 17} These data suggest that the relevant effects of dietary soy isoflavones may strongly depend on the oestrogen context.

In the present study, endogenous oestrogen levels in the mouse mammary tumour virus (MMTV)-erbB2 transgenic mouse mammary cancer model were altered in order to evaluate the interactive effects of dietary soy isoflavones with endogenous oestrogen levels on breast cancer. Tumour incidence, latency and other mouse-related tumour factors were studied; breast cancer angiogenesis was evaluated by measuring vascular endothelial growth factor (VEGF) protein levels in breast tissue. The hypothesis that dietary soy isoflavones would have oestrogen agonist effects in a low-oestrogen context, and oestrogen antagonist effects in a relatively high oestrogen context, was investigated.

Materials and methods

Experimental Animals

This study was approved by the Medical Ethics Committee of the First Affiliated Hospital of Henan University of Science and Technology, Luoyang, China, reference no. 2010001. The study followed the recommendations of the National Institutional Animal Care and Use Committee for the handling, maintenance, treatment and sacrifice of the animals.

The MMTV-erbB2 transgenic mice used in this study were purchased from The Jackson Laboratory (Bar Harbor, ME, USA, license no. 0207 AX1); they were housed in a specific pathogen-free room (n = 3 per cage), maintained at an ambient temperature of 22 ± 2 °C and humidity of 50 ± 10%, with a 12-h light/12-h dark cycle. Mice were allowed free access to tap water and a phyto-oestrogen-free diet, which was used as the control diet in the study and was prepared with cornmeal rather than soybean meal, to prevent any additional components of soy being added to the diet.

A total of 180 5-week-old healthy adult female MMTV-erbB2 transgenic mice, weighing ∼13 g (range 11 – 15 g) were selected for this study. They were divided into three equal groups: low-oestrogen group (subjected to dorsal bilateral ovariectomy, under anaesthesia with 1% pentobarbital sodium, 80 mg/kg), normal-oestrogen group (untreated), and high-oestrogen group (5 μg/injection of E2 given intraperitoneally [i.p] once every 3 days). Two weeks after the ovariectomy and intraperitoneal injection, each group was subdivided into two groups: the experimental group (fed soybean feed) and control group (fed control feed). The mice had free access to feed during the experiment. The ratios of feed components are shown in Table 1.

Table 1:

Feed composition for the two subgroups within each oestrogen level group (high, normal, low); control feed was fed to control subgroups and soybean feed was fed to experimental subgroups

Feed Type	Component
	Cornmeal	Soybean meal a	Flour	Bran	Sugar	Corn oil	Milk powder
Control feed	40	0	25	20	5	5	5
Soybean feed	0	40	25	20	5	5	5

Data presented as % of total composition.

a

100 g soybean meal contained 0.26 g isoflavone.

Breast Biopsies

From 20 weeks of age, animals were monitored once every 3 days for tumour development. Tumour incidence and latency period were calculated according to the date when a tumour was detected by palpation of the mammary gland (tumour detected when diameter was > 3mm). Single tumours were measured at the maximum diameter. Where multiple tumours were present, the tumour with the largest maximum diameter was recorded. Tumour measurements (maximum diameter) were recorded once every 3 days from the day of first detection. Mice were euthanized using pentobarbital sodium when the volume of the breast tumour was 1.5 cm² or the mouse was 60 weeks of age. The mammary glands were removed and preserved in 10% neutral-buffered formalin, routinely embedded in paraffin wax and sectioned at 4 μm. Sections were stained with haematoxylin and eosin, and evaluated microscopically for epithelial pathology by an expert in human breast pathology.

Immunohistochemical Analysis of VEGF

The paraffin-wax embedded sections (4 μm) were cleared with xylene and rehydrated through a graded ethanol series. The slides were then treated with 0.3% H2O2 for 10 min at room temperature to inhibit endogenous peroxidase activity, placed in 0.01 mM sodium citrate (pH 6.0) and heated in a microwave oven for antigen retrieval. Slides were incubated with mouse monoclonal anti-VEGF primary antibody (1:100 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA) overnight at 4 °C in a humidified atmosphere, then washed three times in 0.1 mM PBS for 2 min. The slides were then incubated for 30 min at room temperature with a secondary horseradish peroxidase-conjugated goat antimouse antibody (Wuhan Boster Biological Technology, Wuhan, China), washed three times in 0.1 mM PBS for 2 min and stained using 3′3-diaminobenzidine. The primary antibodies were replaced with nonspecific rabbit immunoglobulin G for the negative controls.

Immunohistochemical staining results were interpreted by two experienced pathologists and VEGF staining was semiquantitatively assessed by combining the immunohistochemical staining intensity (none, 0; weak, 1; moderate, 2; strong, 3), with the percentage of tumour cells stained (0, 0%; 1, 1 - 10%; 2, 11 - 49%; 3, 50 -100%). Raw data were converted to an immunoreactive score (IRS) by adding together the scores for staining intensity and percentage of tumour cells stained. An IRS of 0-2 was considered ‘-’ (negative), 3 as ‘+’, 4 - 5 as ‘++’, and 6 as ‘+++’. The percentages of mice positive for VEGF immunostaining in each group were also calculated.

Statistical Analyses

Data were analysed using the SPSS^® software package, version 17.0 (SPSS Inc., Chicago, IL, USA) for Windows^®. The latent period and growth of the tumour were expressed as mean ± SD. One-way analysis of variance was used to detect between-group differences. Tumour incidence and VEGF immunoreactivity in the tumour tissues were expressed as frequencies, and were analysed using the χ²-test. A P-value of < 0.05 was considered to be statistically significant.

Results

The number of breast tumours at first detection in each mouse was 1 or 2, with a total of 115 mice (63.9%) developing breast tumours.

The incidence of breast cancer in each of the groups is shown in Table 2. In the high-oestrogen environment, the incidence of breast cancer was significantly lower in the experimental group than in the control group (P < 0.01). In the low-oestrogen environment, incidence of breast cancer was significantly higher in the experimental group than in the control group (P < 0.01). In the normal-oestrogen environment, there was no significant difference in the incidence of breast cancer between the experimental group and the control group.

Table 2:

Incidence of breast cancer in mouse mammary tumour virus (MMTV)-erbB2 transgenic mice with different endogenous oestrogen levels (low-, normal- and high-oestrogen groups) fed either soybean feed (experimental groups) or control feed (control groups)

Incidence/cases	Low-oestrogen group		Normal-oestrogen group		High-oestrogen group
	Experimental n = 30	Control n = 30	Experimental n = 30	Control n =30	Experimental n =30	Control n =30
Tumour cases	28	13	16	22	11	25
Tumour-free cases	2	17	14	8	19	5
Tumour incidence	93.3a,b	43.3 c	53.3	73.3	36.7 a	83.3

Data presented as n mice or % tumour incidence.

P < 0.05 between-group comparisons:

a

versus control group in same oestrogen environment;

b

versus other experimental groups in normal- and low-oestrogen environments;

C

versus other control groups in normal and low-oestrogen environments; χ²-test.

The mean ± SD breast cancer period of latency was 257.51 ± 46.72 days. There were no significant between-group differences in the mean tumour latency period (Fig. 1) or in the mean tumour maximum diameter (Fig. 2). All tumour tissues were shown to be adenocarcinomas following histopatho -logical assessment. The tumour cells were densely arranged, with clear boundaries in the intra- and extracellular regions. Cell nuclei were deeply stained. The tissue had an atypical morphology, with areas of necrosis surrounded by inflammatory cells. These characteristics confirmed the tissue to be breast cancer (Figs 3A and 3B).

Figure 1:

Tumour latency in mouse mammary tumour virus (MMTV)-erbB2 transgenic mice with different endogenous oestrogen levels (low, normal and high), fed either soybean feed (experimental group) or control feed (control group). Data presented as mean ± SD; no statistically significant between-group differences (P ≥ 0.05); one-way analysis of variance

Figure 2:

Tumour diameter in mouse mammary tumour virus (MMTV)-erbB2 transgenic mice with different endogenous oestrogen levels (low, normal and high), fed either soybean feed (experimental group) or control feed (control group). Data are mean ± SD; no statistically significant between-group differences (P ≥ 0.05); oneway analysis of variance

Figure 3:

Representative photomicrographs of breast tumour tissue in mouse mammary tumour virus (MMTV)-erbB2 transgenic mice with different endogenous oestrogen levels (low, normal and high), fed either soybean feed (experimental group) or control feed (control group). Haematoxylin and eosin stained tumour tissue: (A) original magnification × 200; (B) original magnification × 400. Immunohistochemical staining for vascular endothelial growth factor (VEGF) showing (C) VEGF-negative cells and (D) VEGF-positive cells (original magnification × 400)

Representative examples of positive and negative VEGF immunohistochemical staining are shown in Figs 3C and 3D, and the incidence of VEGF-positive breast tumour tissue in each group is shown in Table 3. There were no significant between-group differences.

Table 3:

Effect of soybean isoflavone on tumour tissue vascular endothelial growth factor (VECF) protein levels in mouse mammary tumour virus (MMTV)-erbB2 transgenic mice with different endogenous oestrogen levels (low-, normal- and high-oestrogen), fed either soybean feed (experimental group) or control feed (control group)

VECF status	Low-oestrogen group		Normal-oestrogen group		High-oestrogen group
	Experimental n = 28	Control n = 13	Experimental n = 16	Control n = 22	Experimental n = 11	Control n = 25
Positive	17(60.7)	7 (53.8)	9 (56.3)	15 (68.2)	6 (54.5)	16 (64.0)
Negative	11 (39.3)	6 (46.2)	7 (43.8)	7(31.8)	5 (45.5)	9 (36.0)

Data presented as n mice or % incidence of VEGF-positive tumour tissue.

No statistically significant between-group differences (P ≥ 0.05), χ²-test.

Discussion

The phyto-oestrogen soy isoflavone is an important component of soy and is the main component of human dietary oestrogen-like supplements. Studies show that oestrogen exposure is an important determinant of breast cancer risk, but its effect on breast cancer is influenced by a variety of conditions. Results of in vivo studies into the effects of endogenous hormone levels are inconclusive. ¹⁸ The present study used the stable breast cancer transgenic mouse model MMTV-erbB2, and altered the endogenous oestrogen environment of the mice by i.p. injection of oestrogen or by performing ovariectomy. The mice were fed with soybean or a control diet without isoflavones. The study showed that dietary soy isoflavones affected the development of breast cancer under different endogenous oestrogen levels in this established murine model. In a high-oestrogen environment, the incidence of breast cancer in the presence of dietary soy isoflavones was significantly lower than in the control group fed without dietary isoflavones. Conversely, in a low-oestrogen environment, the incidence of breast cancer in the presence of dietary soy isoflavones was significantly higher than in the control group. These findings establish a relationship between the endogenous oestrogen environment and the effects of isoflavones in the breast, and suggest that soy intake may not be safe for all women.

Breast cancer is a hormone-related tumour. A study assessing the risk of breast cancer according to hormone levels in the blood showed a positive relationship between serum E2 levels and breast cancer. ¹⁹ Results from a study of women on the island of Guernsey ²⁰ suggested that mean baseline blood hormone levels were higher in women diagnosed with breast cancer than in those who remained free from breast cancer. A statistically significant inverse association was found between the internal oestrogen environment and the incidence of breast cancer, in MMTV-erbB2 transgenic mice in the present study. For mice fed a normal diet without soy isoflavones, the incidence of breast cancer in a low-oestrogen environment was significantly lower than that in a normal oestrogen environment. These data suggest that endogenous oestrogen levels are an important factor for the occurrence of breast cancer, which is consistent with published studies. ²¹

The oestrogenic activity of isoflavones is well documented. ²² Numerous studies have confirmed that isoflavones bind oestrogen receptors, ²³ induce distinct changes in oestrogen receptor conformation, ²⁴ and elicit a variety of oestrogen receptor-mediated effects.^25,26 The structure of the isoflavone is similar to that of E2; thus, it is capable of binding to the oestrogen receptor and modulating the effect of E2. ²⁷ Isoflavones have, however, far lower affinity for the oestrogen receptors than E2, leading to different effects, depending on endogenous oestrogen levels. The mechanism of action may be related to the oestrogen-like and anti-oestrogen effect of the isoflavones. In a high-oestrogen environment, binding of isoflavones and their metabolites to oestrogen receptors inhibits E2 binding, exerting an anti-oestrogen effect. In a low-oestrogen environment, which has little or no E2, serum isoflavone concentrations may exceed E2 concentrations following a soy meal. Although isoflavone has far lower affinity for oestrogen receptors than E2, ²² it still exerts an oestrogen-like effect. One study found a significant negative correlation between consumption of soy isoflavone and risk of developing breast cancer in premenopausal Japanese women, but not in postmenopausal women. ²⁸ A meta-analysis of published studies investigating the relationship between soy isoflavones and breast cancer before 2004 ²⁹ included a stratified analysis according to menopausal status. The study showed a stronger inverse association between soy isoflavone exposure and breast cancer risk in premenopausal women than in postmenopausal women, and suggested that soy isoflavones have a stronger protective effect in premenopausal women, with relatively weak or no effect on postmenopausal women. Endogenous oestrogen levels vary with different physiological stages and between women of the same physiological stage, with obesity being the most important factor underlying high oestrogen levels. ³⁰ Women with a high body mass index are more likely to have high oestrogen levels and may benefit most from dietary soy isoflavones. Recent studies from the Shanghai Breast Cancer Study support this idea, indicating significantly lower breast ³¹ and endometrial ³² cancer risk specifically, in more obese women with a high intake of soy.

Inconsistencies in the present study between tumour incidence and latency period, tumour size and expression of VEGF may raise questions regarding the relevance of the mouse model used. Between-group differences were only found in tumour incidence, with no significant between-group differences in tumour latency period and average maximum tumour diameter. The presence of VEGF protein in tumour tissues, as a measure of the proliferation and angiogenesis of the tumour, was also evaluated. No significant between-group differences were observed, indicating that isoflavones may have no function in the proliferation, growth and (therefore) prognosis of breast cancer. Tumour occurrence and development involves complex processes. Isoflavones may affect the initial stages of carcinogenesis, or increase susceptibility to the development of breast cancer. Inconsistencies in tumour incidence and latency period, size of the tumours and expression of VEGF between the present and the former study ³³ need to be verified. Discrepancies between studies may relate to oestrogen context, isoflavone dose and formulation, and species differences. Rodents, unlike primates, have complete loss of oestrogen with ovariectomy, and this total sex-steroid deficiency may sensitize tissues and obscure potential hormone interactions.

In conclusion, the present study shows that the effects of soy isoflavones on breast cancer depend on endogenous oestrogen levels, promoting the occurrence and development of breast cancer at low-oestrogen levels and inhibiting breast cancer growth at high-oestrogen levels. In different physiological periods of a woman's life, especially after menopause, endogenous oestrogen levels are significantly lower than normal, and exposure to soy isoflavones is associated with an increased risk of breast cancer. ³⁴ This study supports the rationale of a relationship between soy isoflavone exposure and breast cancer that is dependent on endogenous oestrogen levels. The data suggest a need to use soy isoflavones more rationally in finding effective ways to prevent and treat breast cancer.