| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Albert Einstein College of Medicine
Departments of Developmental and Molecular Biology and
Address all correspondence and requests for reprints to: Jeffrey W. Pollard, Ph.D., Departments of Developmental and Molecular Biology and Obstetrics & Gynecology and Women’s Health, Albert Einstein College of Medicine, 1300 Morris Park Avenue, New York, New York 10461. E-mail: pollard{at}aecom.yu.edu
 |
Introduction |
|---|
 Top Introduction References |
|---|
and ß] that
are ligand activated transcription factors. Historically, the rodent
reproductive tract has been one of the most studied estrogen target
tissues, in part due to the fact that the primary role for estrogens is
in the regulation of reproductive processes. Estradiol-17ß
(E2) in both the adult rat and mouse uterus stimulates
epithelial cell proliferation, increases vascular permeability
and water imbibition, and changes the expression of many genes
(1, 2). In addition, E2 also results in an influx into the
uterine stroma of leukocytes, including eosinophils and macrophages (3, 4). Indeed, the early uterine responses to estrogens look a lot like an
inflammatory response characterized by edema and leukocyte recruitment,
and analogies have been drawn between the two processes (5). This
increase in uterine wet weight has provided a reliable assay for
estrogens and antiestrogens for many years. Recent studies of germ-line
mutations in the ER receptor gene have indicated that all these
pleiotropic effects of E2 in the uterus have as their
primary cause the interaction of E2 with the ER (6). Early studies in the 1950s showed variations in response to estrogens in wet weight and accumulation of total protein, RNA, and DNA in the uterus according to mouse strain (7, 8). More recently, members of the Teuscher research group and collaborators have shown, in rats, the differential accumulation of CD4+ and Ia+ cells (9) and, in the mouse, variations in the eosinophil recruitment into the uterus according to strain (10). In the latter case, using differential exclusion mapping, they identified a locus, Est1, on chromosome 4 that controlled the degree of E2-regulated influx of eosinophils into the uterus. In a paper published in this issue of Endocrinology (11), these studies have been extended to the more classical response of E2- regulated uterine wet weight and, using this assay, the authors have conclusively shown genetic control of estrogen responsiveness (11). Using C57BL/6J and C3H/HeJ, which are high (3.5-fold) and low (2.2-fold) responders to E2, respectively, and genome exclusion mapping using a (C57BL/6J x C3H/HeJ) x C3H/HeJ back cross population, they identified two loci Est2and Est3 that controlled the phenotypic variation in uterine wet weight. These loci were different from the previously described Est1. Further analysis also identified another locus that interacted with the other Est loci and this was designated as Est4. Linkage analysis to microsatellite markers mapped the quantitative trait loci (QTL), Est2 and Est3, to chromosome 5 and 11, respectively, and studies in mice heterozygous at single or double loci suggested that there is epistasis or interaction between these two loci. The authors concluded "that this functional synergism implies that an intimate association between the products of these loci alone or in combination with other loci may be involved in controlling the higher responder phenotype."
What could be the causes of these interactions and their resultant
phenotypic effect? Unfortunately, the exact molecular mechanisms of
E2 regulation of uterine wet weight increases are unknown.
As stated above, the E2-ER interaction is the primary
event in estrogen action. However, neither locus maps to the ER (or
for that matter ERß) gene. Nevertheless, the ER is involved in
complexes with many other proteins and several of these including,
SRC-1, TIF2/GRIP1, p/CIP and CBP, enhance its ligand-induced
transcriptional activity in reporter gene assays (12). Therefore, one
possible mechanism for the variation in response to E2 is
an alteration in the concentrations of these coactivator proteins
resulting in subtle alterations in transcriptional efficiency and/or
target gene selection. The ER has also been shown to be the target
of several growth factor pathways (12). For example, epidermal growth
factor (EGF) can stimulate uterine epithelial cell proliferation in
ovariectomized mice but only in the presence of ER (13, 14). EGF
signaling can result in ligand-independent transactivation of the ER
possibly through phosphorlyation of the receptor, and presumably this
cross-talk between the signaling pathways activates the ER and is
the cause of the proliferative response of uterine epithelial cells to
EGF (13). Mutation of the three amino-terminal phosphorylation sites in
the ER results in a significant reduction in transcriptional activity,
and differential phosphorylation of these sites have been reported in
different cell types in response to E2 binding (15, 16).
Clearly increases or decreases in receptor activity, and therefore
presumably estrogenic response, could be modulated by differential
phosphorylation that result from growth factor signaling in the cells.
Therefore, variations in the expression of the growth factor or in the
intensity of signaling to its downstream effectors caused by
Est2 and/or Est3 could explain the variations in
estrogen responsiveness. It would be very interesting to explore ER
phosphorylation patterns in the uterus of the different inbred strains
used in the present study.
Many of the actions of estrogen are also mediated through the local synthesis of cytokines, growth factors, and chemokines (5, 17, 18). Thus, recruitment of macrophages into the uterus at proestrus is largely under the regulation of E2-induced uterine epithelial synthesis of the mononuclear phagocytic growth factor, colony stimulating factor-1 (19). Similarly, eosinophil recruitment appears to be due to the E2-induced expression of eotaxin, an eosinophil chemoattractant (20). Furthermore, there is uterine synthesis of other, so-called immune cytokines (lymphokines), and this appears to be under the regulation of E2 early in the uterine response (5, 17, 18, 21). These cytokines, along with others such as vascular endothelial growth factor (VEGF) (22) or Epo (23), may be the cause of the vascular permeability and resultant water imbibition that constitutes a significant proportion of the wet weight response. Clearly, modulations in the synthesis of these cytokines/chemokines by Est loci could result in altered responsiveness to E2.
Using microsatellite markers, Roper and colleagues (11) determined
the map position of Est2 and Est3 on chromosome 5
and 11, respectively. Interestingly, each QTL maps to a region that
encodes E2-regulated genes. These include the seratonin
receptor 5a and interleukin-6 (IL-6) on chromosome 5 and procollagen,
type 1; 1, integrin 3; colony-stimulating factor, granulocyte;
retinoic acid receptor (RAR) ; thyroid hormone receptor ; and the
familial breast cancer gene, Brca1, on chromosome 11.
Particularly intriguing is the localization of the retinoic acid
receptor and Brca1 to Est3. Familial
mutations in BRCA1 results in a very substantially elevated
risk of estrogen-induced breast cancer, and these families also show
increased susceptibility to ovarian cancer (24). However, the
regulation of BRCA1 is complex and does not appear to involve direct
estrogen action in the mouse. Furthermore, the exact function of BRCA1
is controversial, but there is a consensus developing that it is
important in some aspects of DNA repair (25). It would seem unlikely,
therefore, despite its expression in the uterus (26), that BRCA1 is an
effector of acute responses to E2 such as growth. Instead,
it seems more likely that the QTL regulate the expression of BRCA1 in
response to E2. RAR, expressed in the uterus and
regulated during the estrous cycle (27), is known to interact in its
ligand-bound state with estrogen response elements (ERE) half sites and
possibly to determine the occupancy of ERs at these sites (28). Indeed,
retinoic acid can inhibit estrogen-induced uterine cell proliferation
at least, in immature rats (29). Thus, genetically regulated
alterations in the level of RAR within the cell could conceivably
lead to differential responses to estrogens. The key now will be to
refine the map positions and determine if these genes, or others,
modulate estrogen responsiveness and to molecularly clone the sequences
responsible for the quantitative effects to determine the genetics
basis of their actions and interactions.
Mouse genomics is proceeding at a pace with the chromosomal-marker density increasing daily and, along with sequencing efforts, many gene loci are becoming known. The refinement of these techniques is providing means of mapping many modifier genes. Classically, modifier genes were shown to influence the penetrance of mouse mutations such as the t-locus (30). Recently, such genes have been thrown into the limelight because of dramatic influences of mouse strain on the phenotypes of induced-null-mutations. A compelling recent example of this effect is the variations in phenotype of the EGF receptor (EGFR) null. These can range from periimplantation to perinatal lethality, according to strain background (31). As mentioned above, EGF cross-talks to the ER and, the dramatic variations in cell responsiveness to EGF according to genetic background shown by the EGFR nullizygous mouse, strongly supports the idea that modulation of growth factor signaling in the uterus could alter responsiveness to E2.
QTL have also recently been mapped for several other growth traits. Of interest to the present discussion is the mapping of ten QTL affecting peak bone mineral density in female mice (32). Estrogens are known to be major regulators of bone density, and their loss in postmenopausal women is a major determinant of osteoporosis that results from decreased bone density (33). The estrogen effects are thought to be enacted at least in part, through the regulation of IL-6 expression in the bone (34). Fascinatingly, Il-6 and Est2 maps to the same chromosomal region. Could Est2 be coincident or interact with one of the loci affecting bone density mapped by Klein et al. (32)?
Are there other pathological conditions caused or ameliorated by estrogens whose severity or incidence could be affected by QTL? In a Fischer 344 rat model, chronic estrogen stimulation results in excessive proliferation and benign tumor formation in the pituitary. Because this occurs in Fischer 344 but not other strains such as Sprague Dawley, it allowed the identification by Wendell and Gorski (35), using similar techniques to those described by Roper et al., of five QTL affecting tumor growth. None of these rat QTL were syntenic with Est2–4 mapped in the current study, although Est1 regulating the E2-induced eosinophil recruitment was syntenic to the QTL mapped to rat chromosome 5. In humans, estrogens are the major risk factor in the development of breast and endometrial cancer (36, 37). Could one of the Est loci, mapped by either group in rodents, modulate susceptibility to these cancers in humans through altering responsiveness to E2? The localization of Brca1 to Est3locus, makes this suggestion even more intriguing.
Estrogen responsiveness is clearly a complex genetic trait, and the loci involved in such traits are only now becoming amenable to gene mapping techniques (38). Analysis of these loci and their interactions are going to affect profoundly the understanding of estrogen responsiveness and will have an impact on human health. For example, there has been considerable debate over the protective effects of plant phytoestrogens on bone loss, or in the prevention of cancer, or cardiovascular disease in human females. Consequently, a regular alternative-medicine industry has grown up around products such as soya that are enriched in such low potency estrogens (39). Indeed, estrogens in red wine might explain the "French paradox" of how a high fat diet is correlated with low cardiovascular disease risk (40)! Similarly, the increase in breast cancer frequency or the acceleration of puberty in females in the West has been suggested to be due to compounds such as PCBs that mimic estrogens and that persist in the environment as a result of excessive pesticide use (41, 42). Extrapolation of the rodent data to humans would suggest that some individuals are high, whereas others are low responders to estrogens. Maybe a low potency estrogen might attain full efficacy in a high-responder woman. If this were the case, some of the environmental estrogens might turn out to be sufficiently potent to be carcinogenic in some individuals while protective in others. Such genetic variations could provide a partial explanation as to reason why only a cohort of women develop breast cancer.
A Holy Grail of the pharmaceutical industry is to find estrogens that are effective in one tissue but not others so that preventative treatments can be employed for bone loss, or in the prevention of cardiovascular disease, without increasing the risk of endometrial or breast cancer. In light of the current paper mapping QTL for estrogen responsiveness and the observation that the newly mapped QTL in the rat alter estrogen responsiveness in the pituitary gland but not in the uterus, an alternative therapeutic approach suggests itself that could lead to a new generation of pharmaceuticals. These would, rather than modulating ER responses, alter estrogenic activity by regulating, in a tissue-specific manner, the pathways that affect estrogen responsiveness.
Received November 25, 1998.
|
References |
|---|
|
TopIntroductionReferences |
|---|
are produced in the
moue uterus during the estrous cycle and are induced by estrogen and
progesterone. Dev Biol 151:297–305[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  N. Geary, L. Asarian, K. S. Korach, D. W. Pfaff, and S. Ogawa Deficits in E2-Dependent Control of Feeding, Weight Gain, and Cholecystokinin Satiation in ER-{alpha} Null Mice Endocrinology, November 1, 2001; 142(11): 4751 - 4757. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Spearow, P. Doemeny, R. Sera, R. Leffler, and M. Barkley Genetic Variation in Susceptibility to Endocrine Disruption by Estrogen in Mice Science, August 20, 1999; 285(5431): 1259 - 1261. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
