| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Supplement |
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711
Address all correspondence and requests for reprints to: Suzanne E. Fenton, Ph.D., U.S. Environmental Protection Agency, ORD, National Health and Environmental Effects Research Laboratory, Reproductive Toxicology Division, MD-67, Research Triangle Park, North Carolina 27711. E-mail: fenton.suzanne{at}epa.gov.
Abstract
Breast cancer is the most common non-skin cancer among women in this country. Breast cancer risk is significantly influenced by genetics, but over 70% of the women that are diagnosed have noninherited or sporadic cancer. The risk of breast cancer is thought to be modified by lifestyle and environment. Exposures to certain chemicals and hormone-mimicking or endocrine-disrupting compounds (EDCs) are suspected of contributing to increased breast cancer incidence as well as precocious puberty in the United States. Studies of EDC effects in rodents indicate that multiple toxicants can alter mammary gland development, with or without changing other markers of puberty. EDCs can cause transient and persistent effects on mammary gland development depending on dose, exposure parameters, and whether exposure was during critical periods of gland growth or differentiation. Adverse effects from these abnormal developmental patterns include the presence of carcinogen-sensitive structures in greater numbers or for longer periods in the gland and inhibited functional differentiation leading to malnutrition or increased mortality of their offspring. Developmental toxicants of the mammary gland could lead to an increase in the incidence of mammary tumors if they alter circulating or tissue-localized hormone levels, gland receptor expression patterns, hormone transport, or metabolism that results in altered response to endogenous hormones or growth factors. Environmental disruptors of rodent mammary gland development must be identified for informed decisions in epidemiological studies aimed at identification of environmental factors contributing to breast cancer risk, altered breast development during puberty, or inability to produce sufficient breast milk.
THE RODENT MAMMARY gland has been used for decades as a model of the human breast. Animal models have been used to study both developmental events, including the function of genes in development via the use of knockout mouse technology, and the mechanisms, treatment, and potential prevention of breast cancer. These models have many fundamental benefits (beyond cost and ethical considerations), such as the ability to delete, suppress, or overexpress a gene in the tissue of interest, use of a large tissue pool to provide the statistical power necessary to substantiate a finding, the fact that rats and humans produce similar types of tumors in their mammary tissue (1, 2), and that the rodent and human undergo mammary gland development at a similar biological pace (albeit the absolute time is quite different) with a few exceptions (reviewed in Ref. 3). Rats and mice have been used to test endogenous components (those that are produced naturally by the body) or events for their ability to confer sensitivity to or protection from mammary tumor development (e.g. protective properties of pregnancy and adverse effects of 17ß-estradiol). In recent years, animal models have been used to test exogenous agents (accidental or unperceived exposures from the environment) for their mammary carcinogenicity or conferred sensitivity to other chemicals, in addition to their developmental effects on the mammary tissue. This review summarizes the developmental effects of exogenous components, such as endocrine-disrupting compounds (EDCs), on the mammary gland and highlights epidemiological examples that relate to findings in the rodent. Discussion of critical periods of exposure that may make the mammary gland more or less susceptible to additional carcinogenic/chemical exposures is also included.
Critical Periods of Mammary Gland Development
The epithelial bud and ductal outgrowth of the mouse and rat mammary gland begins to form during late gestation, approximately 6–7 d before birth (4). However, the little information available on fetal breast epithelial development suggests that the human tissue begins ductal development early in gestation, about embryonic wk (EW) 12–14 (5, 6). Both the rodent and human mammary epithelia grow at an isometric rate (at the same rate as the body) until just before puberty, although there is a short burst just before birth (5, 6, 7). Peripubertal growth of the mammary tissue, controlled by the rapidly changing hormonal milieu, is exponential, and the fat pad rapidly fills with epithelium to manifest the adult form of the gland. The gland will stay in this form, with minor changes depending on the stage of the estrous (rodent) or menstrual (human) cycle, or will undergo dramatic differentiation during pregnancy (7). The comparison of human and rodent mammary gland developmental periods is reviewed in Table 1
(modified from Ref. 3).
|
, the prenatal development of the mammary epithelial sprout (when the primary ducts form at the bud site coincident with the nipple) serves as a critical event. The mammary epithelial bud receives signals from the surrounding fat pad to form primary ducts and begin extension into the fat before birth (4). Interference of this occurrence could lead to altered timing of mammary development or formation of the glandular structures (altered number of primary ducts or blind ducts or unusual presence of nipple/areola), leaving lasting effects on the gland.
|
Finally, the gland undergoes a third critical period of development during pregnancy. During this time, the gland prepares itself for functional lactation. Interruption of this process can lead to mortality or malnutrition of the offspring. This is particularly important in wildlife and domestic species that rely solely on maternal nourishment for reproduction. Pregnancy has been shown to be protective of breast tissue to later-life disease (cancer) (reviewed in Ref. 9). Whether this is because of some permanent change in receptor status, cell morphology, or signaling mechanisms necessary for cancer initiation or progression or is simply because of the final pregnancy-induced differentiation of breast TEBs is under investigation.
EDCs and Mammary Gland Development
There are relatively few studies that have reported on the effects of EDCs (or lack thereof) on mammary gland development. There are several possible reasons: 1) the primary focus of traditional 2-yr bioassay studies is to look for potential carcinogens in animals dosed in adulthood, after the most sensitive windows, for effects in the mammary gland; 2) the mammary gland is not a tissue suggested to be evaluated in tiered testing strategies for EDC effects on reproductive or endocrine tissues after peripubertal exposures (not looking for the effects at all); 3) only a small portion of government-based funding has prioritized the area of EDCs to cover these type of studies, even though expertise and ideas are available (e.g. EPA STAR grants support EDC research); 4) lack of information on the importance of this area of study; or some combination thereof. This review is meant to inform on the importance of this area of study but is not a breast cancer review. The relationship between critical-period exposure and later-life disease susceptibility is an important area of discussion.
Of the several EDCs that have been reported to affect mammary gland development in a rodent model, there is probably more information on the potent aryl-hydrocarbon receptor ligand 2,3,7,8-tetrachlorodibenzo-p-dioxin (dioxin) than any other. This persistent EDC, recognized as a component of Agent Orange (10), is of note because its effects in humans have been (or are being) evaluated because of occupational or accidental environmental exposures. One report (11), looking specifically at pubertal development in a cohort of 200 Belgian adolescents (aged 15.8–19.6 yr old) exposed to polychlorinated aromatic hydrocarbons (including dioxin) via pollution demonstrated a significant delay in breast development in girls that was associated with a doubling of serum dioxin concentrations. The young women did not exhibit delayed menarche, but both girls and boys exhibited delayed adrenarche associated with serum contaminant levels. Another study evaluating pubertal timing in girls exposed to polybrominated biphenyls during gestation and lactation (12) found no change in mammary developmental timing but suggested that exposure caused earlier menstruation (average of 1 yr earlier).
Early-life exposure to dioxin in animal models leads to delayed mammary gland development when exposed during critical periods of mammary development. Brown and co-workers (13) demonstrated, in Sprague-Dawley rats, that female offspring exposed to 1 µg/kg dioxin on gestation d 15 (GD15) and cross-fostered to control dams after birth have delayed mammary differentiation at 50 d old. Lewis et al. (14) demonstrated stunted mammary development in 11-wk-old Holtzman rats after a single 1 µg/kg dose of dioxin on GD15 and ovariectomy at 9 wk. Another study (15) demonstrated that Long-Evans rat offspring (a third strain) exposed to 1 µg/kg dioxin on a single GD had delayed epithelial migration through the fat pad of the gland that was persistent into adulthood. The abnormal development was evident as early as postnatal d 4 (PND4), characterized by stunted longitudinal growth and fewer primary ducts and lateral branches. A critical period for the long-lasting developmental effect of dioxin on the mammary gland was determined to be around GD15, because single doses administered as late as GD20 (as well as several days of the postnatal period) had relatively little effect on the morphological development of the pup glands at weaning (15). This critical period corresponds to breast bud development taking place in the first trimester of women’s pregnancies, EW10–13 (Table 1
), a time when some women may not even know they are pregnant, and others are probably not aware of the sensitive developmental processes going on in their growing baby. Furthermore, when dioxin-exposed female rat offspring were bred and allowed to raise their litters, the mammary glands of female offspring in the second generation were also found to be significantly smaller than those of control animals. Although the abnormal development was apparent in the mammary epithelium of dioxin-exposed offspring, a transplantation study of affected epithelia into cleared fat pads of control rats (and vice versa) demonstrated a significant contribution from the stromal compartment of the gland for the abnormal epithelial development (15).
Another period of time during which dioxin-like compounds are reported to affect mammary gland development is during the peripubertal period (16). Animals exposed to multiple doses of 2.5 µg/kg dioxin (PND25, 27, 29, and 31) had inhibited mammary epithelial outgrowth and fewer TEBs on PND32 as a result of their exposure. Dioxin doses as low as 1 µg/kg on GD15 are known to delay puberty in rat offspring (17, 18) without altering serum estradiol levels. Yet, the effects of dioxin on the mammary gland, after identical exposure, persist into adulthood (15).
A third critical period of mammary gland development that is significantly affected by dioxin is development of the lactating gland. Studies in C57BL/6 mice, dosed with 5 µg/kg dioxin on GD0, 7, and 14 demonstrated significant abnormal development of the pregnant gland as early as GD9, with the most dramatic effects seen on the day of parturition (19). Dioxin exposure caused a decreased milk protein gene expression in the gland and early postnatal mortality in a group of dams allowed to deliver their offspring. Importantly, when mammary glands from age-matched nonpregnant mice given the same doses of dioxin were evaluated (with identical time between doses), there was no effect of dioxin. The impact of dioxin, or the related polychlorinated biphenyls (PCBs), on lactational performance in other species has been described (reviewed in Ref. 15). This is yet another example of how important it is to consider the timing of exposure and critical periods of development in the gland of interest. The impact of EDCs on breast milk production in women is not known, but many laboratories are currently investigating the environmental toxicants that are found in breast milk as a means of understanding exposures and potential risk to mother and child (3).
There are other examples of EDCs that have been shown to affect the development of the mammary gland. Atrazine, a high-use chlorotriazine herbicide brought to the attention of risk assessors because of its effects on mammary tumor formation in Sprague-Dawley rats (20), has been reported to cause abnormal mammary epithelial development in rats (21, 22). This nonlipophilic chemical is very different from dioxin (half-life in humans, 6–11 yr; rodent, 10–30 d; from http://ntp.niehs.nih.gov/ntp/roc/toc11.html) (23), having a far shorter half-life in people, rats, and the environment (day or days in mammals to several months in soil) (24). The reports on atrazine revealed that a 3-d exposure during a critical fetal period of mammary development can give rise to the most severe and long-lasting effects in Long-Evans rats, equivalent to that seen after a 7-d exposure (22). Cross-fostering studies revealed that both in utero and lactational exposures are needed to develop the persistent effects of the compound on the mammary gland of the offspring, suggesting that atrazine may be stored in the mammary gland of the dam during exposure and transferred to the offspring during lactation (21). Furthermore, if animals demonstrating abnormal mammary gland development after gestational atrazine exposure are bred, and allowed to rear offspring, the weight gain of their pups is significantly inhibited (12–25%), as is the mammary gland development of the pups (22). This is also a second example of an environmental toxicant that alters mammary gland development in multiple generations. The lowest doses of dioxin and atrazine able to cause these developmental abnormalities are yet to be determined, but it should be mentioned that the doses used in these animal studies are significantly higher than reported human serum or urine levels, respectively.
Low-dose rodent studies using bisphenol A (BPA), a hormonally active compound found in plastics and dental sealants, have been performed (25, 26). This compound has a very short half-life, proposed to be about 90 min in rats (27). These studies suggest that low-dose (25 ng/kg·d) gestational or perinatal BPA administration to CD-1 mice caused a stimulation of mammary growth, with increased TEB early and more lateral branches per gland in the adult, compared with animals exposed to 0 or 250 ng/kg·d BPA. The higher dose of BPA had an opposite effect; reduced lateral and longitudinal growth compared with control (25, 26). In other studies using higher BPA doses (0.5 and 10 mg/kg·d; four daily sc injections starting on GD15), transient increases in mammary differentiation were noted (28). However, when the same lab repeated this experiment (four daily sc injections, 10 mg/kg) in 15-d-old CD-1 mice, they found no effect of treatment (29). These same studies revealed that zearalenone can also stimulate long-lasting proliferative effects on the mammary epithelium when exposure is transplacental (28). This information suggests, again, that the latter part of gestation is a critical period of development for the fetal mammary gland in rodents, correlating with first-trimester development in women (Table 1).
Only a handful of other environmental toxicants (nutritional, pharmaceutical, metals) have been evaluated for their effects on mammary gland development. Some heavy metals are being evaluated for reproductive tissue effects (arsenic, uranium), but cadmium is the only one (to date) with published effects on mammary proliferation (30). These metals are hypothesized to have estrogen-like attributes and are of interest because of their natural contamination of ground water.
Several dietary components known for their estrogen agonist or antagonist properties have been evaluated for their effects on mammary gland development, with a variety of outcomes reported depending on the delivery route (injection vs. oral), species evaluated, developmental time of delivery, exposure dose (many have U-shaped response curves), and diet of the animals on study (reviewed in Ref. 31). Among these compounds is genistein, a heavily studied component of soy. Genistein has been reported to have no effect on morphology of the mouse gland after gestational/lactational gavage exposure (32), to cause hyperplastic lesions in the adult rat gland after gestational/lactational dietary exposure (33) or in utero exposure (eight daily injections midpregnancy) (34), to dose-dependently increase the incidence of dimethylbenz[a]anthracene (DMBA)-induced mammary tumors after in utero exposure (35), or to confer protection from DMBA-induced mammary carcinogenesis after peripubertal exposure of female rats either by injection or via the diet (36, 37). This is an area of study on mammary gland development where few solid conclusions have been reached because of the variety of discrepancies in exposure conditions. The importance of dose for xenoestrogens is raised again in a recent study (38) that demonstrates differing effects of high- and low-dose diethylstilbestrol and tamoxifen on mammary gland proliferation and differentiation after neonatal exposure. Furthermore, the importance of reporting the dietary makeup of rodent study chow is also increasing, because whey (vs. casein) protein diets can alter rat mammary gland development (39). The same holds true for the use of soy-based diets.
Adverse Developmental Effects and Disease
A striking effect of the developmental perturbations in the mammary gland after environmental exposures is altered TEB differentiation (either hastened or delayed). The TEBs are well documented as the structure of the rodent mammary gland most sensitive to the effects of carcinogens (1) and when affected can give rise to adenomas and ductal carcinomas similar to those seen in women. Humans and rodents alike possess TEBs during the peripubertal period. Studies on brief gestational exposure to dioxin or atrazine (15, 21, 22) suggest that those EDCs can delay differentiation 2- to 3-fold compared with controls. This would leave the mammary gland vulnerable to the effects of carcinogens for a significantly longer developmental period. Girls that undergo precocious puberty may also be vulnerable to carcinogens for extended periods if they demonstrate longer time to final Tanner-stage breast development (start early but end at mean age for population). With this said, early or rapid breast development may confer protection from later-life disease.
Environmental exposures that promote or delay differentiation of the mammary epithelia may cause uncommon numbers of TEBs to be present at inopportune times. For example, studies evaluating the effects of dioxin on chemically induced tumor development in rodent models have found that toxicant exposure on GD15 (13) or PND18 (40) increased the number of mammary tumors in the adult rodent. Brown et al. (13) reported more TEBs present in dioxin-exposed animals on PND50 (the day that chemical carcinogen was delivered) compared with controls; some may translate that to mean that dioxin increased the number of TEBs in the gland, but careful examination of this event over time in Long-Evans rats (15) determined that dioxin causes an impairment of TEB development and delays both their formation and differentiation. Once they are present in the gland, they are there for a longer period of time. This finding might explain the increased mammary tumors after carcinogen exposure; more TEBs mean increased targets for carcinogen action. Interestingly, the stromal portion of the mammary gland is not only important for dioxin-induced developmental events (15), but mouse mammary fibroblasts lacking dioxin receptor demonstrate impaired tumorigenicity (nearly four times less than control) in a mouse xenograft model (41). Those findings suggested that the dioxin receptor signaling pathway may be involved in an angiogenic or extracellular matrix mediated response needed for tumor formation.
Few studies of the effects of atrazine on chemically induced mammary tumors have been conducted. Sprague-Dawley rats bearing chemically induced tumors or not were ovariectomized and were treated with atrazine (0, 5, 50, or 500 ppm in diet) to evaluate the chemical’s effect on tumor cell proliferation (42). Only in the group of animals that had no tumors after ovariectomy, they found a significant increase in the percent incidence of mammary tumors at the highest dose of dietary atrazine (500 ppm) and in the number of tumors per rat in the two highest atrazine dose groups (average per rat was less than one). Another study used Ha-ras transgenic rats, chemically induced on PND50 to form mammary tumors (43), that were also treated with 0, 5, 50, and 500 ppm atrazine in their diet. The lowest doses produced higher incidences of tumors than control, but a dose response was lacking. Other studies in Sprague-Dawley rats fed atrazine for nearly 2 yr demonstrated a dose response in increased incidence and earlier appearance of spontaneously forming adenomas and carcinomas, but this effect was due to precocious reproductive senescence (mode of action) caused by the herbicide (20, 44). This finding deems evaluation of the estrous cycle in aging rodents very important, especially in Sprague-Dawley rat tumor studies, because the atrazine mode of action is not thought to be applicable to women (44). The mode or mechanism of action for other mammary developmental toxicants is under investigation.
Several other xenoestrogens and organochlorines, specifically PCBs and 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE), have been suspected of involvement in mammary tumor development. In two separate studies, Desaulniers and co-workers (40, 45) found no significant effect of organochlorine mixtures, delivered on 5 d between birth and weaning, on either mammary gland development or carcinogen-induced tumor development. Another study also demonstrated no effect of prenatal exposure to biologically relevant levels of organochlorine mixture on mammary gland development but that this mixture may interact with dietary genistein to promote ductal hyperplasia (34). Organochlorines have also been the focus of several large case-control studies on breast cancer risk (cohorts of middle-aged women). Results of multiple rodent and epidemiological studies have failed to demonstrate a consistent significant correlation between PCB/DDE (organochlorine) exposure and development of breast cancer (reviewed in Refs. 31 and 46). In fact, a review of five case-control studies (total of over 1400 cases and 1600 controls), which included the Nurses Study, demonstrated no positive relationship between serum PCB or DDE levels and risk of breast cancer (47). However, this developmental exposure review may address the reason for the lack of relationship of body burden and disease in the case-control studies. The studies in rodents described here focus on critical periods of mammary development (in utero, puberty, and pregnancy) and demonstrate effects of environmental compounds during these times. Dioxin, for example, has little if any noticeable effect on the mammary gland of the adult animal yet serious and long-lasting effects if exposure is during a critical period of development (15, 19). Case-control studies lack the ability to evaluate exposures to these women during their own critical breast developmental windows. Longitudinal cohorts are the best way to determine the early life factors or exposures necessary for increased breast cancer risk, even though study results may be painfully slow and costly.
There are many environmental compounds for which there is no information about human or rodent mammary tissue effects, yet levels of these compounds are on the rise in either the U.S. population or wildlife species or both. For example, many chemicals known to alter endocrine hormone levels affect rodent pup weight gain or mortality (potential effect on the dam’s milk production) or other reproductive tissues in the pup after gestational exposure remain to be evaluated for mammary gland developmental effects, not to mention tumor development. Some chemical classes that fall into this category include perchlorates, phthalates, perfluorinated alkyl acids, water disinfection by-products, alkylphenols, other heavy metals, and polybrominated flame retardants.
Recommendations
To translate rodent data on mammary gland effects to human risk assessment, more information is needed on the kinds, doses, and timing of toxicant exposures that affect this important tissue. Repetition of findings in the rodent model from more than one lab, as is the case with the dioxin effects on mammary gland development (13, 14, 15), will help epidemiologists determine which chemical or chemical classes to focus ever-limiting study funds on. Furthermore, determination of sources of these exposures (and potentially the mixture of exposures) will assist epidemiologists and risk assessors. Studies on rodent mammary gland development have shown that the majority of the effects are found after exposure during critical periods of development (bud outgrowth, puberty-induced exponential growth, and pregnancy). Studies that focus solely on exposure of mature animals or adult women and try to evaluate exposures once the disease or adverse health outcome has occurred may be missing important details on developmental effects of the exposure. Future studies on environmental exposures affecting mammary or breast development and their sensitivity to carcinogens must include a careful evaluation of dose- response studies delivered during these sensitive periods of gland development. Many animal studies are conducted with exposure doses that would result in a body burden higher than that found in humans; future work should strive to use doses relevant to humans or at least measure body burden in the rodent studies so that information can be used in risk assessment.
As we begin to better understand the sources of environmental exposures known to affect breast development, pregnant and lactating women and peripubertal adolescents should be informed of these potential sources of exposure by their health education teachers, pediatricians, obstetricians, lactation consultants, or midwives. A substantial learning curve lies ahead to accomplish this goal, but it is imperative that exposure during critical periods of growth of the mammary gland (as well as other reproductive tissues) be limited to decrease or prevent an increase in future adverse reproductive effects (e.g. cancer, fertility, inflammatory disease, and miscarriage).
Footnotes
Portions of these data within this manuscript were presented at the EDC Forum sponsored by The Endocrine Society at the 87th Annual Meeting of The Endocrine Society, San Diego, CA, June 2005.
Disclaimer: The information in this document has been funded by the U.S. Environmental Protection Agency. It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.
First Published Online May 11, 2006
Abbreviations: BPA, Bisphenol A; DDE, 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene; EDC, endocrine-disrupting compound; EW, embryonic week; GD, gestational day; PCB, polychlorinated biphenyl; PND, postnatal day; TEB, terminal end bud.
Received September 2, 2005.
Accepted for publication January 3, 2006.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
