Biol Reprod
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hendry, W. J. I.
Right arrow Articles by Sheehan, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hendry, W. J. I.
Right arrow Articles by Sheehan, D. M.
Agricola
Right arrow Articles by Hendry, W. J. I.
Right arrow Articles by Sheehan, D. M.
Biology of Reproduction 61, 91-100 (1999)
©Copyright 1999 Society for the Study of Reproduction, Inc.

Articles

Differential Activity of Diethylstilbestrol versus Estradiol As Neonatal Endocrine Disruptors in the Female Hamster (Mesocricetus auratus) Reproductive Tract1

William J. III Hendry2,a, Brian L. DeBrota, Xinglong Zheng3,a, William S. Branhamb, and Daniel M. Sheehanb

a Department of Biological Sciences, Wichita State University, Wichita, Kansas 67260-0026 b Division of Reproductive and Developmental Toxicology, National Center for Toxicological Research, U.S. Food and Drug Administration, Jefferson, Arkansas 72079


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The synthetic estrogen diethylstilbestrol (DES) is a potent neonatal endocrine disruptor in the hamster. To test the specificity of this phenomenon, newborn animals were treated with 100 µg of either DES or the natural estrogen, estradiol-17ß (E2). Of the two, neonatal DES exposure caused greater morphological disruption throughout the female reproductive tract in prepubertal animals and in adults that either retained their ovaries or were ovariectomized and then given the same levels of chronic E2 stimulation. In the uterus, a characteristic histopathological profile, including enhancement of both hyperplastic and apoptotic activity, was initiated prepubertally and exclusively in the endometrial epithelial cell compartment from the neonatally DES-treated animals and then was promoted by E2 stimulation during adulthood. Interestingly, apoptotic activity was not detected in an area of endometrial epithelium that progressed to the neoplastic state in a DES-exposed animal. Lastly, chronic estrogen induction of lactoferrin was also restricted to the DES-exposed endometrium. We conclude that 1) DES is more active than E2 as a perinatal endocrine disruptor in the hamster and 2) this experimental system should be generally useful as a means to screen compounds for such activity and then probe their mechanism of action.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
From the 1940s to the 1960s, the synthetic estrogen diethylstilbestrol (DES) was often prescribed for the support of high-risk pregnancies [1]. The negative consequences of this practice began to emerge in 1971 when Herbst et al. [2] described the very early occurrence of a rare cancer, vaginal clear cell adenocarcinoma, in eight young women who had been exposed to DES in utero. Numerous clinical and experimental animal studies have since demonstrated that perinatal DES exposure results in fertility deficits plus teratogenesis and neoplasia throughout the male and female reproductive tract [3, 4]. Thus, DES is an original and potent member of the expanding list of chemicals known as "endocrine disruptors" [5]. There is now intense scientific and public interest in the concept that inadvertent and untimely exposure to such synthetic agents (xenoestrogens), or even to natural substances that possess estrogenic activity (phytoestrogens), may adversely affect the reproductive and general health of both human and wildlife populations [6, 7].

Our analyses of the endocrine-disruptive activity of perinatal DES exposure have been performed in the hamster because its gestation period is very predictable and short (16 days) compared to that (18 to 21 days) of other rodents. Thus we can treat neonates rather than fetuses but still target a very early stage of reproductive tract development. We also reasoned that treating neonates rather than fetuses would avoid uncertainties associated with maternal metabolism and placental transfer of the drug. After a single injection of DES on the day of birth, abnormalities developed throughout the female reproductive tract [8]. In our previous studies focusing on the uterus, we observed rapid and profound morphogenetic alterations during the prepubertal period [9] and a very high incidence of endometrial hyperplasia and neoplasia (adenocarcinoma) in adult animals [8]. Interestingly, the hyperplastic endometrial epithelium in such adult animals is also a site of massive cell death by apoptosis [10]. These lesions (endometrial hyperplasia, apoptosis, and adenocarcinoma) are not due merely to altered function of the hypothalamo-pituitary-ovarian axis, because they occurred in the DES-exposed group but not in the control group of animals when both groups were ovariectomized prior to puberty and then chronically stimulated with replacement levels of estradiol-17ß (E2) [8, 10]. To further investigate the mechanistic basis of this phenomenon, neonatal uteri from control or DES-treated donors were transplanted into the cheek pouches (another advantage of the hamster) of control or neonatally DES-treated adult hosts that were ovariectomized and stimulated with E2. The characteristic histopathological profile (endometrial hyperplasia/dysplasia and apoptosis) occurred only in uterine transplants derived from DES-exposed donors, and this outcome was not influenced by the neonatal treatment history of the host [10]. Thus, atypical estrogen responsiveness is an inherent property of the neonatally DES-exposed hamster uterus rather than an indirect effect mediated by DES acting peripherally within the host. However, this change in estrogen responsiveness does not appear to be due to any major alteration in the physicochemical or functional properties of the uterine estrogen receptor system [11]. We have also determined that neonatal DES treatment induces persistent and epithelial cell-specific imbalances in the estrogen-regulated uterine expression of certain oncogenes that have been implicated in the control of cell proliferation (c-jun, c-fos, c-myc) and apoptosis (bax, bcl-2, bcl-x) [12]. Together, these results support the hypothesis [10, 12] that the neonatal DES insult directly initiates a permanent developmental change and that subsequent exposure promotes the generation of uterine lesions that can ultimately progress to endometrial cancer.

An important unanswered question is whether estrogenicity alone determines the endocrine-disruptive activity of a given agent. The estrogenicity of DES is important to its activity as both a developmental toxicant and a carcinogen, since clinical and experimental data show that perinatal DES-induced teratogenesis and neoplasia are confined to estrogen target organs [3, 4]. However, few direct comparisons of perinatal endocrine-disruptive activity have been made between the synthetic agent (DES) and the primary natural estrogen (E2). Those performed in rodents have suggested that E2 is less disruptive than DES [13, 14] or have led to differing conclusions about whether the long-term consequences of perinatal DES exposure involves factors in addition to its estrogenicity [15, 16]. Complicating such studies is the fact that, in many perinatal rodents, E2 is much less potent than DES as an estrogen due to high serum concentrations of alpha-fetoprotein (AFP), which binds E2 but not DES with high affinity [17]. In contrast, the hamster is free of such complications since its AFP does not bind E2 [18, 19]. Thus the hamster is an excellent system for comparing the perinatal endocrine-disruptive activity of DES and E2. The comparison we have performed includes a comprehensive chronicle of the immediate (prepubertal) and delayed (adult) consequences of the two neonatal treatment regimens. Also, we employed prepubertal ovariectomy plus chronic E2 supplementation to 1) control for the possibility that function of the hypothalamo-pituitary-ovarian axis status in adult animals might be altered differently by the two neonatal treatment regimens and 2) test the relative ability of the two neonatal treatment regimens to directly and permanently disrupt estrogen responsiveness in the hamster uterus.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Animal Treatment

Animal treatments, ovariectomy, chronic E2 stimulation, anesthesia, and killing by decapitation were performed as described previously [8, 9, 11] and in an AAALAC-accredited facility according to the Guiding Principles for the Care and Use of Research Animals promulgated by the Society for the Study of Reproduction. Briefly, timed-pregnant Syrian golden hamsters (Mesocricetus auratus) from Charles River Breeding Laboratories (Wilmington, MA) or Harlan Sprague Dawley (Indianapolis, IN) were caged singly under a 14L:10D photoperiod and with food and water provided ad libitum. Within 6 h of birth (Day 0), litter size was adjusted to eight neonates per litter by eliminating males, and all litter mates received a single s.c. injection of 50 µl corn oil vehicle either alone (control) or containing 100 µg (~33 mg/kg BW) of either E2 or DES. This dose level is high but not unreasonable considering that DES ingestion by pregnant women was as much as 150 mg daily and 18.2 g total during pregnancy [1]. Ovariectomy and placement of E2 pellets were performed at 21 days of age under pentobarbital-induced anesthesia. The pellets, inserted s.c. between the shoulder blades, consisted of a plugged Silastic (Dow Corning, Midland, MI) tube (1.0-cm open lumen length; 1.57-mm i.d.; 2.41-mm o.d.) filled with crystalline E2. According to previous determinations [8, 10, 20], this procedure maintains serum E2 levels at ~200 pg/ml for at least 5 mo. At all the ages studied, at least three animals from each treatment group were anesthetized with CO2, weighed, killed, and immediately processed as described below.

Tissue Harvesting, Processing, and Histological Analysis

Prepubertal animals were eviscerated, and the lower torsos were immersed in ice-cold fixative for at least 24 h before reproductive tracts were excised, trimmed of adhering fat and mesentery under a dissecting microscope, weighed, and placed back in fixative. The fixative consisted of 4% paraformaldehyde in Dulbecco's PBS, pH 7.4. For the adult animals, reproductive tracts were immediately excised and then also placed in fixative for at least 24 h before trimming, weighing, and placement back in fixative. The harvested tracts were ultimately divided into cervix, uterine horns, and oviduct/ovary (when present) regions and embedded in paraffin; 4- to 5-µm cross sections were processed for light microscopy using standard hematoxylin and eosin staining. At least 4 sections cut from midregions of the uterine horns from all the animals per time point and treatment group were analyzed. The photomicrographs shown are representative of the histomorphological condition observed within that group of at least three animals. Comparisons of endometrial epithelial height (basal-to-apical cell dimensions) relied on manual measurements performed on the original 2.5" by 3.5" photomicrographs that correspond to those shown in the indicated figures.

Lactoferrin (LTF) Immunohistochemistry

Tissues harvested as described above from adult animals were fixed by immersion in Kryofix (EM Diagnostic Systems Inc., Gibbstown, NJ) at 4°C for 18 h and embedded in paraffin. After dewaxing, 4- to 5-µm uterine cross sections were quenched of endogenous peroxidase activity by incubation with 0.6% H2O2 in methanol at 37°C for 10 min and blocked with PBS + 1.5% goat serum at 37°C for 20 min. The rabbit anti-mouse LTF antiserum (from Dr. C. Teng; NIEHS, Research Triangle Park, NC) was diluted (1:1000) in PBS + 1.5% goat serum and incubated with the tissue sections at 4°C for 18 h. A kit based on the avidin:biotinylated enzyme complex method (Vectastain ABC; Vector Laboratories Inc., Burlingame, CA) was used to detect immunocomplexes. The diluent and washing media for all steps was PBS + 0.05% Tween 20. Incubations were performed in a moist chamber at 37°C with biotinylated anti-rabbit IgG (1:100) for 1 h and then with avidin:biotinylated peroxidase complex (1:100) for 30 min. Bound peroxidase was visualized with Sigma Fast (Sigma Chemical Co., St. Louis, MO) substrate solution (diaminobenzidine:H2O2:urea in 0.06 M Tris-HCl, pH 7.8) at 25°C for 2 min. Lastly, sections were counterstained with 0.2% methyl green in 0.1 M ammonium acetate (pH 4.6) for 10 min.

Statistics

For the data on animal weight, absolute tissue weight, and normalized tissue weight, a Statistical Analysis Systems statistical package [21] was used to calculate the mean, SE, and the significance levels of differences among the three neonatal treatment groups at all the time points. Factorial ANOVA was followed by Tukey's honestly significant differences test for multiple comparisons [22], and means were considered significantly different at p < 0.05.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Gross Reproductive Tract Development In Intact Animals

The objective of this study was to compare DES with E2 as perinatal endocrine disruptors in the female hamster. We began by studying reproductive tract development prior to the onset of puberty, which occurs normally at Week 4 of life in the hamster [23]. According to body weight measurements, growth of the prepubertal animals was not affected by neonatal treatment with either DES or E2 (not shown). The differential ability of the two neonatal treatment regimens to disrupt early female reproductive tract development was emphasized by normalizing reproductive tract tissue weights to body weights (Fig. 1A). In control animals, reproductive tract growth matched body growth during the first 3 wk of life. Both neonatal estrogen treatment regimens altered this balance by causing an early surge in tract growth. In E2-treated animals, the surge was relatively modest, peaking at Day 3 and then subsiding to control levels by Day 9. In DES-exposed animals, the early surge in tract weight was much more dramatic, peaking at Day 5 and then only partially declining so that relative tract weight remained double that in both the control and E2-treated groups. The differences in reproductive tract weight profiles shown in Figure 1A were due almost entirely to changes in mass of the uterine horns (not shown).



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 1. Effect of neonatal treatment with DES vs. E2 on normalized reproductive tract weight in prepubertal (A) and adult (B) female hamsters. Animals were injected on the day of birth with 50 µl of corn oil vehicle either alone (control, CON) or containing 100 µg of either E2 or DES. At the indicated ages, animals were weighed and reproductive tracts (cervix, uterine horns, oviducts, and ovaries) were removed and weighed. Reproductive tract results were expressed by normalizing tissue weight to the animal's body weight. Values represent the mean ± SE, n = 3 (error bars not shown where variability is so small as to be masked by the data point). Indicated are means significantly different (p < 0.05) from that of the control group (*) or from those of both the control and the E2-treated group (**) at that time point.

Although both neonatal estrogen treatment regimens slightly depressed animal growth after puberty (average body weight reductions compared to controls were 9% in the E2-treated group and 17% in the DES-treated group), this effect was insufficient to explain the dramatic treatment-dependent differences in reproductive tract weights that occurred (Fig. 1B). In control animals, the increase in normalized tract weight was relatively modest from 1 to 5 mo of age. In neonatally E2-treated animals, a slight additional increase occurred between 1 and 2 mo of age but then remained parallel to the control profile at the later ages. In contrast, normalized reproductive tract weights in the neonatally DES-treated animals averaged at least 5.7-fold greater than in the control animals and at least 2.7-fold greater than in the E2-treated animals at every age.

Uterine Histology In Intact Animals

The differential ability of the neonatal estrogen treatment regimens to disrupt endometrial histomorphology during early development is shown in Figure 2, A–C. The low-magnification photographs confirm the tissue mass trends demonstrated in Figure 1A. They show that uterine cross-sectional area was sometimes modestly increased in E2-exposed animals and always dramatically increased in the DES-exposed animals compared to controls. By Day 5 (Fig. 2A), the DES-exposed uteri exhibited endometrial epithelial cell pseudostratification and hypertrophy (almost double the height of that in the control and E2-exposed uteri). This was also the earliest age at which the endometrial epithelial cell compartment in DES-exposed uteri harbored cavities that contained degenerating cells with the histomorphology of apoptotic bodies [24]. By Day 9 (Fig. 2B), endometrial glands had developed only in the DES-exposed uteri, but endometrial epithelial cell height in those uteri had regressed to that seen in both the control and E2-exposed uteri. By Day 15, developing endometrial glands were present in all three groups (not shown). Also, DES-specific disruption of the endometrial epithelial compartment (cellular hypertrophy, pseudostratification, and cavities with apoptotic cells) intensified between Day 15 (not shown) and Day 21 (Fig. 2C).



View larger version (154K):
[in this window]
[in a new window]
  		FIG. 2. Effect of neonatal treatment with DES vs. E2 on uterine histology in prepubertal and adult hamsters. Animals were injected on the day of birth with 50 µl of corn oil vehicle either alone (control, CON) or containing 100 µg of either E2 or DES. Representative cross sections of uterine horns from animals killed on Day 5 (A), Day 9 (B), Day 21 (C), 1 mo (D), and 4 mo (E) of age were photographed at x40 (left) and x400 (right) (reproduced at 45%). Indicated in the high-magnification panels are the presence in neonatally DES-exposed uteri of pseudostratified endometrial epithelium (asterisks) with cavities that often contained apoptotic cells (arrows) and the presence in neonatally E2-exposed uteri of cystic glandular structures (stars).

Further DES-specific disruption of endometrial histomorphology during adulthood is shown in Figure 2, D and E. In uterine horns from 1-mo-old animals (Fig. 2D), histological analysis at low magnification confirmed that cross-sectional dimensions were increased to a much greater extent by neonatal treatment with DES than with E2. Even at low magnification, the endometrial epithelium in DES-exposed uteri was clearly hypertrophic and contained cavities. At higher magnification, the endometrial epithelial cells in the neonatally E2-exposed uteri were about 50% taller but were organized similarly to those in the control uteri. For neonatally DES-exposed uteri, disruption of the endometrial cell compartment was more dramatic in adult than in prepubertal animals (compare Fig. 2, D and E, with Fig. 2, A–C). In addition to being extremely hypertrophic (more than 3-fold taller than control), the epithelial cells exhibited a very complex pseudostratified organization indicative of a hyperplastic state. Additional distortion came from cavities that frequently harbored apoptotic cells. As the animals aged, cystic glandular structures developed occasionally in the endometrium of neonatally E2-treated animals (Fig. 2E). Even in such cases, the histomorphology of the epithelium lining the cysts resembled the epithelium lining the lumen of control uteri (low columnar cells with orderly and basally located nuclei) rather than that in neonatally DES-exposed uteri (hypertrophic/hyperplastic cells in a pseudostratified organization that is riddled with cavities containing apoptotic cells). In fact, the neonatally DES-exposed epithelium at this age contained so many cavities that it assumed a "foamy" appearance.

Adult Uterine Responses to Estrogen Stimulation

Like other rodents, hamsters exposed perinatally to estrogens enter a "persistent estrous" state characterized by anovulatory and cystic ovaries that continuously secrete high levels of E2 but little or no progesterone [8, 25]. If the severity of such an altered endocrine state was different after neonatal exposure to E2 compared to DES, it could explain the differences in uterine responses among the three groups of adult hamsters. To evaluate this possibility, animals in the three neonatal treatment groups were ovariectomized prior to puberty (Day 21) and then stimulated exogenously with E2. Under such conditions, overall animal growth from 1 mo to 5 mo of age was the same among the three neonatal treatment groups (not shown), and the general profiles were quite similar to that for the DES-exposed group of intact adult animals discussed above. Thus, differences in uterine weight profiles among the three treatment groups were quite similar whether expressed on an absolute basis (not shown) or normalized to body weight (Fig. 3). In the youngest animals (1 mo) that had been exposed to E2 for 9 days, normalized uterine weights were the same in control and neonatally E2-exposed animals but were about 2-fold higher in the DES-exposed animals. Thereafter, control uteri grew at a linear rate that generally kept pace with their body weight. Both neonatal estrogen treatment regimens increased the slope of the absolute (not shown) and the normalized uterine growth curves, but this effect was most pronounced in the neonatally DES-treated animals. Compared to control values, increased weight (both absolute and normalized average values) of the E2-exposed uteri barely exceeded 2-fold, while that of the neonatally DES-exposed uteri was approximately 3-fold higher.



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 3. Effect of neonatal treatment with DES vs. E2 on normalized uterine weights in ovariectomized adult hamsters given estrogen replacement. Animals were injected on the day of birth with 50 µl of corn oil vehicle either alone (control, CON) or containing 100 µg of either E2 or DES. On Day 21, they were ovariectomized and given a Silastic implant filled with crystalline E2. At the indicated ages, uteri were removed and weighed. Results were expressed by normalizing tissue weight to the animal's body weight. Values represent the mean ± SE, n = 3 or 4 (error bars not shown where variability is so small as to be masked by the data point). Indicated are means significantly different (p < 0.05) from that of the control group (*) or from those of both the control and the E2-treated group (**) at that time point.

Histological analysis revealed that a uniform level of E2 stimulation during adulthood induced very different patterns of endometrial disruption in the two neonatal estrogen treatment groups (Fig. 4). This was evident even in 1-mo-old animals exposed to E2 for only 9 days (Fig. 4A). For each treatment group, the endometrial epithelium was about 50–100% taller and also was more pseudostratified than in the same treatment group in the intact animals at this age (compare Fig. 4A with Fig. 2D). This indicates that the E2 pellets produced a generally higher estrogenic environment than in the intact adult animals. Figure 4A also demonstrates that the increased uterine tissue mass in both neonatal treatment groups at this early stage of E2 stimulation was primarily due to gains in the mesenchymal compartment (stroma). Lastly, Figure 4A demonstrates that the neonatally DES-exposed endometrial epithelium, even at this early stage of chronic E2 stimulation, was more severely disrupted than in the neonatally E2-exposed animals or at any stage in the intact adult animals (compare Fig. 4A with Fig. 2, D and E). At low magnification, the neonatally DES-exposed endometrial epithelium appeared very foamy. High magnification revealed that it also consisted of extremely hyperplastic cells in a very chaotic pseudostratified organization that was riddled with cavities containing apoptotic cells. One month later (Fig. 4B), low magnification showed that the endometrium in control and neonatally E2-exposed uteri contained many large cystic glands, whereas the complex folded epithelium in the neonatally DES-exposed endometrium retained a foamy appearance. High magnification revealed that the endometrial epithelial cells were only about 20% taller than control in the neonatally E2-exposed group, but more than 2-fold taller than control in the neonatally DES-exposed group, and still retained a severely disrupted cellular organization. As chronic E2 exposure continued (Fig. 4C), cystic glandular development began to develop in DES-exposed uteri (low magnification). While endometrial epithelial cells in the control and neonatally E2-exposed uteri were the same height, those in the neonatally DES-exposed uteri were almost 3-fold taller (high magnification). At this stage in the neonatally DES-exposed uterus, some areas of the endometrial epithelium assumed an adenomatous appearance (low magnification) that still retained the characteristic profile of disrupted cellular organization (high magnification). Also note in Figure 4 that the ratio of epithelial to stromal tissue in the DES-exposed uteri appeared to increase with the duration of E2 stimulation.



View larger version (134K):
[in this window]
[in a new window]
  		FIG. 4. Effect of neonatal treatment with DES vs. E2 on uterine histology in ovariectomized adult hamsters given estrogen replacement. Animals were injected on the day of birth with 50 µl of corn oil vehicle either alone (control, CON) or containing 100 µg of either E2 or DES. On Day 21, they were ovariectomized and given a Silastic implant filled with crystalline E2. Cross sections of uterine horns from animals at 1 mo (A), 2 mo (B), and 4 mo (C) of age were photographed at x40 (left) and x400 (right) (reproduced at 45%). Indicated are areas of hyperplastic endometrial epithelium with a chaotic pseudostratified organization (asterisks) plus numerous cavities often containing apoptotic cells (arrows), cystic glandular structures (stars), and areas of adenomatous epithelium (arrowheads).

A noteworthy observation was made in a DES-exposed animal at 5 mo of age. In a uterine cross section from this animal (Fig. 5), most of the endometrium (top panels) exhibited the characteristic disrupted state described above (hypertrophic/hyperplastic epithelial cells and cavities containing degenerating cells with the histopathology of apoptotic bodies). However, the section also contained the profile of a distinctive polypoid mass (bottom panels). It had the same histomorphology (numerous back-to-back glandular profiles consisting of atypical epithelial cells and little intervening stromal tissue) as other tumors that have been identified as endometrial adenocarcinomas in neonatally DES-treated hamsters [8]. Importantly, the tumor mass was devoid of the cavities and apoptotic bodies that lent a foamy appearance to the adjacent nontumor regions of the disrupted endometrium.



View larger version (150K):
[in this window]
[in a new window]
  		FIG. 5. Comparison of histopathology between nontumorous (upper panels) and tumorous (lower panels) endometrial regions of a neonatally DES-exposed uterus. The uterus was from a 5-mo-old animal that was injected on the day of birth with 100 µg DES, ovariectomized on Day 21, and also given a Silastic implant filled with crystalline E2. Uterine cross sections were photographed at x40 (left) and x400 (right) (reproduced at 47%). Apoptotic cells (arrows) were present in the hyperplastic endometrial epithelium but not in the adjacent tumor mass.

Immunohistochemical Analysis of LTF Expression

To extend our histomorphological observations with biochemical analyses, we chose the estrogen-regulated glycoprotein LTF. Shown in Figure 6 for the three neonatal treatment groups are typical immunohistochemical results from adult animals chronically stimulated with E2 in adulthood. In uteri from both the control and neonatally E2-treated animals, similar low levels of LTF-specific reaction product were localized within the endometrial epithelial compartment. Some apparently secreted immunoreactive material also appeared within the cystic glandular structures that developed in the uterus from the neonatally E2-treated animal. This phenomenon was occasionally observed in both the control and neonatally E2-treated groups (not shown). In contrast, the disrupted endometrial epithelium in neonatally DES-treated animals was always densely labeled with LTF-specific reaction product.



View larger version (98K):
[in this window]
[in a new window]
  		FIG. 6. Effect of neonatal treatment with DES vs. E2 on expression of LTF in the uterus of ovariectomized adult animals given replacement estrogen. Animals were injected on the day of birth with 50 µl of corn oil vehicle either alone (control, A) or containing 100 µg of either E2 (B) or DES (C). On Day 21, they were ovariectomized and given a Silastic implant filled with crystalline E2. Uteri were harvested from 5-mo-old animals, and cross sections were incubated with a polyclonal antibody against mouse LTF. Immunocomplexes were detected using the ABC/peroxidase method and visualized by reaction with a H2O2:diaminobenzidine substrate solution that deposits an insoluble, dark brown product. Sections were also counterstained with methyl green and then photographed at x40.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We found that DES is more active than E2 as a neonatal endocrine-disrupting agent in the female hamster. While the uterus is the focus of this report, the ovary, oviduct, and cervix also exhibited much greater disruption in neonatally DES-treated than in neonatally E2-treated animals (not shown). Preliminary descriptions of our ovarian, oviductal, and cervical observations have been reported [26, 27] and generally resemble those described in other rodents following perinatal DES treatment [13, 14].

In addition to the comparative data presented, this study provides a comprehensive histomorphological chronicle of 1) the appearance and progression of neonatal DES-induced disruption in the hamster uterus and 2) the way in which a directly induced and permanent form of altered estrogen responsiveness contributes to the disruption phenomenon. Thus it complements previous findings made at different developmental stages and under various experimental conditions [8–10]. Those aspects that differ from results of DES-related studies in other experimental systems are noted below.

In prepubertal hamsters, the pattern and degree of disrupted uterine development at the gross level was much greater in hamsters treated neonatally with DES than with E2. This result cannot be due to differences in the onset of ovarian steroidogenesis among the neonatal treatment groups because measurements of serum E2 levels during this period detected no significant difference between control and neonatally DES-treated animals, nor were these levels significantly altered in animals ovariectomized on Day 3 of life [8]. On the other hand, ovariectomy on Day 3 of life partially reduced the abnormal uterine growth that occurred between Days 15 and 21 in the DES-treated animals [8]. Thus, the prepubertal ovary appears to modulate the ability of neonatal DES exposure to disrupt early uterine development, but the mechanism remains to be determined.

In the endometrial epithelial cell compartment of prepubertal hamsters, neonatal E2 treatment also failed to induce the disruption pattern that was induced by neonatal DES treatment. The DES-specific disruption pattern, separate aspects of which were noted in previous studies [8–10], included a hypertrophic and hyperplastic response that resulted in cellular crowding and disorganization, the appearance of cavities that harbored degenerating (apoptotic) cells, and the precocious development of endometrial glands. The latter response contrasts with the situation in the rat, where endometrial gland development was inhibited by neonatal DES exposure [28].

As the ovary-intact animals reached adulthood, the difference in degree of uterine disruption at the gross and histological level continued to increase between the groups treated neonatally with DES or E2. Interestingly, this and previous studies [8–10] show that acute perinatal exposure to DES elicits a consistent hypertrophic/hyperplastic response in the hamster uterus, whereas it elicits a hypotrophic/hypoplastic response in the rat [28, 29] and either a hypotrophic/hypoplastic or a hypertrophic/hyperplastic response in the mouse [13, 30, 31].

When animals from both neonatal estrogen treatment groups were ovariectomized and exposed chronically during adulthood to the same level of E2, uterine disruption at the gross and histological level remained most severe in the DES-treated group. Thus the differences in uterine disruption are not due to different degrees of altered neuroendocrine function in the two treatment groups. The results also confirm previous evidence [10] that neonatal DES treatment permanently alters estrogen responsiveness in the hamster uterus via a direct mechanism. However, the increased estrogen responsiveness in the prenatally DES-treated hamster contrasts with evidence of reduced estrogen responsiveness in the prenatally DES-treated mouse [31, 32].

The scope and design of this comparative study provide some new insight into the morphogenesis of neonatal DES-induced uterine disruption in the hamster. For instance, the dysplastic changes in the endometrial epithelial cell compartment of neonatally DES-treated animals followed a pattern of progression with noteworthy architectural and cytological features. Architecturally, the ratio of epithelial to stromal tissue increased until, by 4 mo of age, the epithelium was often crowded into complex adenomatous structures with little intervening stroma. Evaluated according to the clinical description and staging of endometrial hyperplasia and carcinoma, this pathological situation in the hamster is reminiscent of the condition originally termed adenomatous hyperplasia [33] and now referred to as complex hyperplasia with cellular atypia [34]. We also captured an interesting example of the neoplastic outcome often promoted by E2 stimulation of the neonatally DES-exposed hamster endometrium [8]. It was characterized as an endometrial adenocarcinoma (distinct from complex hyperplasia) based on the greater degree of cytological atypia (larger nuclei of variable size and shape that have lost polarity, increased nuclear-to-cytoplasmic ratios, prominent nucleoli, and irregularly clumped chromatin) plus greater general architectural disruption (numerous areas of back-to-back glands with few or no intervening stromal cells) [34].

The tendency for chronic E2 stimulation to induce the formation of endometrial cysts in all three neonatal treatment groups is a phenomenon that we observe consistently in our experimental system. We assume that it somehow represents a distortion of the normal process of endometrial gland formation. However, the actual mechanism responsible for this architectural anomaly remains unknown.

The phenomenon of altered estrogen responsiveness in the neonatally DES-exposed hamster uterus includes increases in both cell growth and apoptosis in the endometrial epithelial cell compartment [10]; it appears to be driven by imbalances in the expression of certain oncogenes that have been implicated in the control of cell proliferation (c-jun, c-fos, c-myc) and apoptosis (bax, bcl-2, bcl-x) [12]. However, the tumor mass shown in this study was devoid of cavities with apoptotic cells even though adjacent regions of hyperplastic endometrial epithelium were riddled with such elements. Thus our experimental system may represent a situation in which apoptosis serves to eliminate mutated cells that develop as a result of abnormal proliferative activity and in which frank neoplasms erupt at sites where apoptotic activity is either lost or overwhelmed [35]. Two related processes may influence apoptosis in the endometrial epithelial compartment of neonatally DES-treated hamsters. One possibility is that inhibition of adhesion between epithelial cells induces them to undergo apoptosis [36]; the other is that disruption of interactions between epithelial cells and the basement membrane induces a form of apoptosis known as anoikis [37]. In fact, preliminary studies have provided 1) ultrastructural evidence that the basement membrane is disrupted under the severely dysplastic and apoptotic endometrial epithelium in neonatally DES-treated hamsters [38] and 2) biochemical evidence for an altered association between specific proteins (cadherins and ß-catenin) [39] that mediate adhesion between epithelial cells [40] and also appear to play a central role in some epithelial neoplasms [41].

Since neonatal DES treatment caused epithelial cell-specific imbalances in the estrogen-regulated expression of several genes in the hamster uterus [12], we tested whether neonatal E2 treatment could induce the same phenomenon. As a new target, we chose the glycoprotein product of the LTF gene. This iron-binding molecule is expressed in a wide variety of tissues, but it is a particularly sensitive and directly up-regulated marker of estrogen stimulation in the endometrial epithelium [42, 43]. Although the function of LTF is largely unknown, a recent study suggests that it is associated with malignant transformation of the human endometrium [44]. In the mouse, perinatal DES exposure permanently disrupts LTF expression [42, 45], apparently by a gene demethylation mechanism [46]. Our immunohistochemical analysis showed that neonatal treatment with DES but not with E2 resulted in up-regulated expression of the LTF protein throughout the endometrial epithelial cell compartment of E2-stimulated adult hamsters. This is strong biochemical evidence that DES is more potent or that it acts differently than E2 as a neonatal endocrine disruptor in the hamster uterus.

The plasma protein AFP is present at high concentration during fetal and neonatal development [47]. The fact that, in some rodents, AFP is a high-affinity and high-capacity binder of E2 but not DES has been suggested as an explanation for the striking transplacental and neonatal toxic effects of DES [17]. The hamster provides a good means to test this hypothesis because, like human AFP [48], hamster AFP does not bind E2 [18, 19]. Thus unequal availability of the two estrogens to target tissues because of differential binding to AFP in the neonatal hamster is unlikely to explain the observed differences in activity between the two estrogens as neonatal endocrine disruptors. On the other hand, albumin does bind DES preferentially in rats and hamsters [49, 50], which could influence the duration of exposure and thus duration of response. In fact, this phenomenon has been cited as a possible explanation [50] for our observation in the hamster uterus of a major discrepancy in the binding affinity of cytosol and nuclear forms of the estrogen receptor [11]. Another hypothetical scheme to explain the perinatal endocrine-disruptive action is based on the fact that DES can be metabolized to reactive intermediates that can form adducts with DNA [51] but retain considerable affinity for the estrogen receptor [52]. Such potentially mutagenic molecules could then be targeted by the estrogen receptor in target cells to the promoters of those regulatory genes that drive estrogen-dependent growth. These possibilities need to be probed in future studies.

In summary, the synthetic estrogen DES was more active than the natural estrogen E2 as a neonatal endocrine disruptor agent in the female hamster. The impact of this phenomenon was very dramatic in the uterus, where the neonatal DES insult initiated alterations in early uterine development that were strongly promoted by estrogen stimulation in adulthood. The DES-dependent uterine lesions were most striking in the endometrial epithelial cell compartment. They consisted of hyperplasia and apoptotic activity during the preneoplastic stage, whereas apoptosis appeared to be lost at sites of neoplastic progression. These morphological abnormalities were accompanied by altered expression of the estrogen-responsive gene LTF. More recently, we have determined that DES is also much more active than E2 as a neonatal endocrine disruptor in the reproductive organs of male hamsters [53]. On the basis of these and previous observations, the hamster appears to be a particularly valuable experimental system in which to screen compounds for potential endocrine-disruptive activity in both sexes and then probe their mechanism of action.



View larger version (175K):
[in this window]
[in a new window]
  		FIG. 2. Continued.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the excellent processing of histology samples by Pathology Associates International, Jefferson, AR. We also thank Dr. Christina T. Teng for the gift of the anti-LTF antiserum and Dr. Karen L. Brown for valuable guidance in performing statistical analyses.


   FOOTNOTES
 
1 This work was supported by the National Cancer Institute (CA60250), the Flossie West Memorial Trust Foundation, and the United States Food and Drug Administration. Back

2 Correspondence: William J. Hendry III, Department of Biological Sciences, Wichita State University, 1845 Fairmount, Wichita, KS 67260–0026. FAX: 316 978 3772; hendry{at}wsuhub.uc.twsu.edu Back

3 Current address: Howard Hughes Medical Institute, Division of Hematology and Oncology, Washington University School of Medicine, 660 South Euclid Avenue, Campus Box 8125, St. Louis, MO 63110–1093. Back

Accepted: February 9, 1999.

Received: October 21, 1998.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Herbst AL, Scully RE, Robboy SJ. Prenatal diethylstilbestrol exposure and human genital tract abnormalities. Natl Cancer Inst Monogr 1979; 51:25–35.
  2. Herbst AL, Ulfelder H, Poskanzer DC. Adenocarcinoma of the vagina: association of maternal stilbestrol therapy with tumor appearance in young women. N Engl J Med 1971; 284:878–881.
  3. Marselos M, Tomatis L. Diethylstilbestrol: I, Pharmacology, toxicology and carcinogenicity in humans. Eur J Cancer 1992; 28A:1182–1189.
  4. Marselos M, Tomatis L. Diethylstilbestrol: II, Pharmacology, toxicology and carcinogenicity in experimental animals. Eur J Cancer 1993; 29A:149–155.
  5. Colburn T, vom Saal FS, Soto AM. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 1993; 101:378–384.[Medline]
  6. Sharpe RM, Skakkebaek NE. Are estrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet 1993; 341:1392–1395.[CrossRef][Medline]
  7. Raloff J. Gender benders: are environmental "hormones" emasculating wildlife? Science News 1994; 145:24–27.
  8. Leavitt WW, Evans RW, Hendry III WJ. Etiology of DES-induced uterine tumors in the Syrian hamster. In: Leavitt WW (ed.), Hormones and Cancer. New York: Plenum Publishing Corporation; 1982: 63–86.
  9. Hendry III WJ, Leavitt WW. Altered morphogenesis of the immature hamster uterus following neonatal exposure to diethylstilbestrol. Differentiation 1993; 52:221–227.[CrossRef][Medline]
  10. Hendry III WJ, Zheng X, Leavitt WW, Branham WS, Sheehan DM. Endometrial hyperplasia and apoptosis following neonatal diethylstilbestrol exposure and subsequent estrogen stimulation in both host and transplanted hamster uteri. Cancer Res 1997; 57:1903–1908.[Abstract/Free Full Text]
  11. Hendry III WJ, Leavitt WW. Binding and retention of estrogen in the uterus of hamsters treated neonatally with diethylstilbestrol. J Steroid Biochem 1982; 17:479–487.[CrossRef][Medline]
  12. Zheng X, Hendry III WJ. Neonatal diethylstilbestrol treatment alters the estrogen-regulated expression of both cell proliferation and apoptosis-related protooncogenes (c-jun, c-fos, c-myc, bax, bcl-2 and bcl-x) in the hamster uterus. Cell Growth Differ 1997; 8:425–434.[Abstract]
  13. McLachlan JA, Newbold RR, Bullock BC. Long-term effects on the female mouse genital tract associated with prenatal exposure to diethylstilbestrol. Cancer Res 1980; 40:3988–3999.[Abstract/Free Full Text]
  14. Davies J, Russell M, Davenport GR. Effects of maternal administration of diethylstilbestrol and estradiol on the newborn guinea pig. Acta Anat 1985; 122:39–61.[Medline]
  15. Henry EC, Miller RK, Baggs RB. Direct injection of diethylstilbestrol (DES) and estradiol (E2) into fetal rats: a method for investigating their teratogenicity and distribution metabolism. Teratology 1982; 25:47A–48A.
  16. Henry EC, Miller RK, Baggs RB. Direct fetal injections of diethylstilbestrol and 17ß-estradiol: a method for investigating their teratogenicity. Teratology 1984; 29:297–304.[CrossRef][Medline]
  17. Sheehan DM, Young M. Diethylstilbestrol and estradiol binding to serum albumin and pregnancy plasma of rat and human. Endocrinology 1979; 104:1442–1446.[Abstract]
  18. Savu L, Nunez EA, Jayle MF. High affinity testosterone, corticosterone and progesterone binding activities in pregnant hamster serum. Endocrinology 1977; 101:369–377.[Medline]
  19. Schatz R, Laugier C, Soto AM, Sonnenschein C. Alpha-fetoprotein serum levels and the development of estrogen-sensitive cell multiplication in the hamster uterus. Biol Reprod 1983; 28:1148–1154.[Abstract]
  20. Hendry III WJ, Branham WS, Sheehan DM. The hamster cheek pouch as a convenient ectopic site for studies of uterine morphogenesis and endocrine responsiveness. Differentiation 1992; 51:49–54.[CrossRef][Medline]
  21. SAS/STAT User's Guide, Version 6. Cary, NC: Statistical Analysis System Institute, Inc.; 1989.
  22. Sokal RR, Rohlf FJ. Biometry. San Francisco, CA: W.H. Freeman and Co.; 1981.
  23. Kent GC Jr. Chapter 8: Physiology of reproduction. In: Hoffman RA, Robinson PF, Magalhaes H (ed.), The Golden Hamster: Its Biology and Use in Medical Research. Ames, IA: The Iowa State University Press; 1968: 119–138.
  24. Sandow BA, West NB, Norman RL, Brenner RM. Hormonal control of apoptosis in hamster uterine luminal epithelium. Am J Anat 1979; 156:15–36.[CrossRef][Medline]
  25. Saunders FJ. Effects of sex steroids and related compounds on pregnancy and on the development of the young. Physiol Rev 1968; 48:601–643.[Free Full Text]
  26. Hendry III WJ, Branham WS, Sheehan DM. Differential potency of a synthetic (diethylstilbestrol) and a natural (estradiol) estrogen as neonatal endocrine disruptors of the hamster uterine cervix. Biol Reprod 1998; 58(suppl 1):abstract 49.
  27. May JV, Rueda BR, Hendry III WJ. Physiological and morphological ramifications of neonatal diethylstilbestrol (DES) exposure in the hamster ovary. Biol Reprod 1998; 58(suppl 1):abstract 57.
  28. Medlock KL, Sheehan DM, Nelson CJ, Branham WS. Effects of postnatal DES treatment on uterine growth, development, and estrogen receptor levels. J Steroid Biochem 1988; 29:527–532.[CrossRef][Medline]
  29. Medlock KL, Branham WS, Sheehan DM. Long-term effects of postnatal exposure to diethylstilbestrol on uterine estrogen receptor and growth. J Steroid Biochem Mol Biol 1992; 42:23–28.[CrossRef][Medline]
  30. Newbold RR, McLachlan JA. Vaginal adenosis and adenocarcinoma in mice exposed prenatally or neonatally to diethylstilbestrol. Cancer Res 1982; 42:2003–2011.[Abstract/Free Full Text]
  31. Ostrander PL, Mills KT, Bern HA. Long-term responses of the mouse uterus to neonatal diethylstilbestrol treatment and to later sex hormone exposure. J Natl Cancer Inst 1985; 74:121–135.
  32. Maier dB, Newbold RR, McLachlan JA. Prenatal diethylstilbestrol exposure alters murine uterine responses to prepubertal estrogen stimulation. Endocrinology 1985; 116:1878–1886.[Abstract]
  33. Ferenczy A. Recent advances in endometrial neoplasia. Exp Mol Pathol 1979; 31:226–235.[CrossRef][Medline]
  34. Gordon MD, Ireland K. Pathology of hyperplasia and carcinoma of the endometrium. Semin Oncol 1994; 21:64–70.
  35. Wyllie AH. Apoptosis and the regulation of cell numbers in normal and neoplastic tissues: an overview. Cancer Metastasis Rev 1992; 11:95–103.[CrossRef][Medline]
  36. Bates RC, Buret A, van Helden DF, Horton MA, Burns GF. Apoptosis induced by inhibition of intercellular contact. J Cell Biol 1994; 125:403–415.[Abstract/Free Full Text]
  37. Frisch SM, Francis H. Disruption of epithelial cell-matrix interactions induces apoptosis. J Cell Biol 1994; 124:619–626.[Abstract/Free Full Text]
  38. Hendry III WJ, Walker NJ. Histomorphological effects of neonatal diethylstilbestrol exposure in the hamster uterus. In: Program of the 79th annual meeting of the Endocrine Society; 1997; Minneapolis, MN. Abstract P1–543.
  39. Hendry III WJ, Simpson K-L, Schmidt J. Neonatal diethylstilbestrol exposure and cell adhesion aspects in the hamster endometrium. Biol Reprod 1997; 56(suppl 1):abstract 416.
  40. Schmidt JW, Piepenhagen PA, Nelson WJ. Modulation of epithelial morphogenesis and cell fate by cell-to-cell signals and regulated cell adhesion. Semin Cell Biol 1993; 4:161–173.[Medline]
  41. Peifer M. ß-Catenin as oncogene: the smoking gun. Science 1997; 275:1752–1753.[Free Full Text]
  42. Newbold RR, Hanson RB, Jefferson WN. Ontogeny of lactoferrin in the developing mouse uterus: a marker of early hormone response. Biol Reprod 1997; 56:1147–1157.[Abstract]
  43. Tourville DR, Ogra SS, Lippes J, Tomasi JB. The human reproductive tract: immunohistochemical localization of IgA, IgG, IgM, secretory "piece" and lactoferrin. Am J Obstet Gynecol 1970; 108:1102–1108.[Medline]
  44. Walmer DK, Padin CJ, Wrona MA, Healy BE, Bentley RC, Tsao M-S, Kohler MF, McLachlan JA, Gray KD. Malignant transformation of the human endometrium is associated with overexpression of lactoferrin messenger RNA and protein. Cancer Res 1995; 55:1168–1175.[Abstract/Free Full Text]
  45. Nelson KG, Sakai Y, Eitzman B, Steed T, McLachlan J. Exposure to diethylstilbestrol during a critical developmental period of the mouse reproductive tract leads to persistent induction of two estrogen-regulated genes. Cell Growth Differ 1994; 5:595–606.[Abstract]
  46. Li S, Washburn KA, Moore R, Uno T, Teng C, Newbold RR, Mclachlan JA, Negishi M. Developmental exposure to diethylstilbestrol elicits demethylation of estrogen-responsive lactoferrin gene in the mouse uterus. Cancer Res 1997; 57:4356–4359.[Abstract/Free Full Text]
  47. Mizejewski GJ. The phylogeny of alpha-fetoprotein in vertebrates: survey of biochemical and physiological data. Crit Rev Eukaryotic Gene Expr 1995; 5:281–316.[Medline]
  48. Nunez E, Vallette G, Benassayag C, Jayle M-F. Comparative study on the binding of estrogens by human and rat serum proteins in development. Biochem Biophys Res Commun 1974; 57:126–133.[CrossRef][Medline]
  49. Okulicz WC, Johnson LD. The relative binding affinity of diethylstilbestrol to uterine nuclear estrogen receptor: effect of serum and serum albumin. Proc Soc Exp Biol Med 1987; 185:478–483.[Abstract]
  50. Okulicz WC, Leavitt WW. Binding and biologic activity of diethylstilbestrol in the hamster: influence of a serum component on estrogen receptor binding and estrogenic activity. J Steroid Biochem 1988; 31:371–375.[CrossRef][Medline]
  51. Metzler M. Metabolic activation of xenobiotic stilbene estrogens. Fed Proc 1987; 46:1855–1857.[Medline]
  52. Korach KS, Metzler M, McLachlan JA. Estrogenic activity in vivo and in vitro of some diethylstilbestrol metabolites and analogs. Proc Natl Acad Sci USA 1978; 75:468–471.[Abstract/Free Full Text]
  53. Khan SA, Ball RB, Hendry III WJ. Effects of neonatal administration of diethylstilbestrol in male hamsters: disruption of reproductive function in adults after apparently normal pubertal development. Biol Reprod 1998; 58:137–142.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Biol. Reprod.Home page
W. J. Hendry III, W. S. Branham, and D. M. Sheehan
Diethylstilbestrol Versus Estradiol as Neonatal Disruptors of the Hamster (Mesocricetus auratus) Cervix
Biol Reprod, May 1, 2004; 70(5): 1306 - 1316.
[Abstract] [Full Text] [PDF]

Home page
 Biol. Reprod.Home page
C. T. Teng, C. Beard, and W. Gladwell
Differential Expression and Estrogen Response of Lactoferrin Gene in the Female Reproductive Tract of Mouse, Rat, and Hamster
Biol Reprod, November 1, 2002; 67(5): 1439 - 1449.
[Abstract] [Full Text] [PDF]

Home page
 Exp. Biol. Med.Home page
W. J. Hendry III, D. M. Sheehan, S. A. Khan, and J. V. May
Developing a Laboratory Animal Model for Perinatal Endocrine Disruption: The Hamster Chronicles
Experimental Biology and Medicine, October 1, 2002; 227(9): 709 - 723.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hendry, W. J. I.
Right arrow Articles by Sheehan, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hendry, W. J. I.
Right arrow Articles by Sheehan, D. M.
Agricola
Right arrow Articles by Hendry, W. J. I.
Right arrow Articles by Sheehan, D. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS