Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2006 Aug 14;103(34):12837–12842. doi: 10.1073/pnas.0603642103

Estrogen receptor signaling protects against immune suppression by UV radiation exposure

Sitarina Widyarini 1,*, Diane Domanski 1, Nicole Painter 1, Vivienne E Reeve 1,†
PMCID: PMC1568934  PMID: 16908847

Abstract

The phytoestrogenic isoflavonoid equol is known to protect against solar-simulated UV radiation-induced inflammation, immunosuppression, and skin carcinogenesis. The mechanism may involve antioxidant actions, because equol not only is a radical scavenger but also enhances the induction of a relevant cutaneous antioxidant, metallothionein. However, this study in female hairless mice examined whether the estrogenicity of the isoflavonoid might be responsible. Protection by topically applied equol against photoimmune suppression was found to be strongly and dose-dependently inhibited by the estrogen receptor (ER) antagonist ICI 182,780. Furthermore, ICI 182,780 alone was found to significantly exacerbate immunosuppression resulting from solar-simulated UV radiation irradiation, suggesting a natural role for the ER in photoimmune protection. In support of this role, topical application of the physiological ligand 17-β-estradiol also provided dose-dependent photoimmune protection, inhibitable by ICI 182,780, that was attributed largely to the inactivation of the downstream actions of cis-urocanic acid, an important endogenous immunosuppressive photoproduct. Thus, a hitherto unrecognized function of the ER as a normal photoprotective immune regulator in the skin was revealed. The relationship between equol and cutaneous metallothionein suggests an association of the ER with this inducible antioxidant in constraining the photoimmune-suppressed state and therefore in the prevention of the facilitation of photocarcinogenesis by this immunological defect. This role for the ER may underlie important gender-specific differences in UV-responsiveness that would reflect different needs for environmental photoprotection in males and females.

Keywords: cis-urocanic acid; equol; hairless mouse; ICI 182,780; steroid hormone

Recent studies in mice with topically applied isoflavonoids derived from the red clover, Trifolium pratense, have shown that these phytochemicals may protect effectively against UV radiation-induced skin damage. Particular interest has centered on equol ([S]-4′7-dihydroxyisoflavane), an isoflavonoid metabolite produced from the dietary isoflavone daidzein by the gut microflora in mammals. Equol has been found to protect mice not only against UV radiation-induced cutaneous inflammation, observed as the sunburn reaction, but also against photoimmune suppression, which is evident as a defective contact hypersensitivity (CHS) reaction (1). Consistent with this observation, it was found that topical application to humans of an equol-related synthetic isoflavonoid, NV-07α, also protected against the UV radiation-induced suppression of the elicited Mantoux reaction (2). In mice it was shown that the immunoprotective mechanism of topically applied equol involved inactivation of the downstream actions of cis-urocanic acid, a major UV-induced immunosuppressive photoproduct produced in the skin (1). Because chronic photoimmune suppression is a prerequisite for the promotion of UV radiation-initiated tumors (3), and because topically applied equol was also demonstrated to have antiphotocarcinogenic properties in mice (4), the photoprotective effects of equol offer the opportunity to identify critical pathways linking the UV-induced immune defect with the development of skin cancer.

Inherent antioxidant properties of the isoflavones are implied by their polyphenolic molecular structures, which readily enable them to act as H atom donors, and have been thought to be responsible for their biological actions. It is well recognized that oxidative stress contributes to photobiological skin damage by UV radiation, both genetic and immunological, and that it may play a role in the development of UV radiation-induced skin cancers via either impairment (5). Oxidative radical scavenging by equol has been reported in numerous systems (6–8). Furthermore, both equol and NV-07α have the capacity to stimulate the solar-simulated UV radiation (SSUV) induction of metallothionein (MT) in the skin (9). In addition to its capacity for metal ion chelation, MT is one of the endogenous cutaneous antioxidants that have been identified as relevant for protection against oxidative photodamage (10). In MT-null gene knockout mice topical equol application failed to elicit its immunoprotective role against SSUV or cis-urocanic acid treatment (10). The responses to isoflavone treatment of a second UV-protective inducible endogenous antioxidant, heme oxygenase-1 (11), have not yet been assessed.

However, in addition to their antioxidant properties, many of the isoflavonoids are also recognized as phytoestrogens. Estrogenic activity of equol has been identified by others in cultured human cells (12, 13) and in healthy humans (14). Equol has been shown to bind to both estrogen receptor (ER) α and ER-β and to effectively activate both receptors, although with greater transactivation of ER-β (13, 15). Furthermore, antioxidant response elements associated with the ERs have been described (16), so that the antioxidant and estrogenic properties of equol might be functionally linked. Nevertheless, a dependence on estrogen or other steroid hormones of the immune dysfunction induced by SSUV radiation has not previously been recognized.

This study examines whether the mechanism underlying the photoimmune protection by equol in the Skh:hr-1 hairless mouse involves the isoflavonoid’s phytoestrogenicity. Estrogen pathways were modified in the mouse by using surgical ovariectomy, and more powerfully by specific pharmacological ER blockade, and the effects on moderate SSUV exposure of the skin were observed. Equol treatment was by topical application, and systemic changes in immune function were measured by the CHS reaction, a sensitive assay of T cell-mediated immunity relevant to both mice and humans.

Results

Ovariectomy and ER Blockade Increased SSUV-Induced Suppression of CHS and Reduced the Photoimmune Protection by Equol (10 μM).

Surgically ovariectomized hairless mice were produced in an attempt to reduce natural estrogen availability. This procedure did not affect the ability of the mice to produce a normal CHS reaction (Fig. 1). However, ovariectomy resulted in a small but significant (P < 0.05) exacerbation of SSUV-induced suppression of CHS, compared with control mice, from 64.0% to 72.7% (Fig. 1). In ovariectomized mice the protection by 10 μM topical equol lotion against irradiation with three times the minimum edematous dose (3 × 1MEdD) of SSUV was also slightly reduced (P < 0.05) compared with control mice, i.e., from 36.9% immune suppression in control mice (UV+Equol) to 46.7% in ovariectomized mice. This observation suggested that normal circulating estrogen may contribute to, rather than compete with, the photoimmune protection by equol, but such a cooperative effect was not strong.

Fig. 1.

Fig. 1.

Open in a new tab

CHS reactions in groups of four to five normal (control) or ovariectomized mice irradiated with 3 × 1MEdD SSUV (UV) and treated with topical 10 μM equol or with 10 nmol/week ICI 182,780. Matched symbols indicate treatments that were significantly different.

Therefore, the estrogenic pathways in both normal and ovariectomized mice were further inhibited by treating the mice with the ER antagonist ICI 182,780 (10 nmol/week). No immunologically significant interaction between equol and ICI 182,780 was found in nonirradiated normal or ovariectomized mice (Fig. 1). In normal mice the ER antagonist partially reversed the photoimmune protection by equol (increased the suppression from 36.9% to 48.4%; P < 0.01), but in the ovariectomized mice equol photoimmune protection was totally abrogated and CHS remained as suppressed (69.9%) as in SSUV-irradiated mice without equol treatment (72.7% suppression). This observation implicated not only the circulating estrogen levels, but also the ER itself as an important mechanistic target for equol’s photoimmune-protective capacity. Because ER blockade was more effective in negating the photoimmune protection by equol in the ovariectomized mice, it also suggested that there could be complex competitive receptor binding locally in the skin between the natural ligand estrogen, the receptor antagonist ICI 182,780, and the phytoestrogenic isoflavonoid equol that might result in systemic immunological modulation.

Equol (20 μM) Protection Against 3 × 1MEdD SSUV Was Inhibited by ICI 182,780 (10, 100, or 1,000 nmol/Week) Dose-Dependently.

Neither the topical treatment with 20 μM equol alone nor the combination of 20 μM equol and ICI 182,780 between 10 and 1,000 nmol/week was immunologically modulating in unirradiated mice (Fig. 2), consistent with observations (Fig. 1) with lower concentrations of each agent. The 20 μM equol treatment was shown to totally protect against the SSUV-induced 65.1% suppression of CHS, in agreement with previous data (1). However, this photoimmune-protective effect of equol was strongly inhibited by ICI 182,780 treatment administered topically at 10 nmol/week. This observation clarified the preliminary observations above, confirming a role for the ER in the photoimmune-protective mechanism of topical equol, and implied that the isoflavonoid was able to bind to the ER to exert such immunomodulation.

Fig. 2.

Fig. 2.

Open in a new tab

CHS reactions in groups of five mice irradiated with 3 × 1MEdD SSUV (UV) and treated with topical 20 μM equol with 10, 100, or 1,000 nmol/week ICI 182,780. Matched symbols indicate treatments that were significantly different.

Interestingly, there was even greater photoimmune suppression observed, despite concurrent equol treatments, by 100 and 1,000 nmol/week ICI 182,780 (Fig. 2). This dose-dependent exacerbation of the immune suppression by SSUV, which was statistically significant at 1,000 nmol/week ICI 182,780 (P < 0.05), suggested that ER blockade, in addition to preventing equol binding, was inhibiting some form of endogenous ER-dependent protective mechanism.

ER Blockade by ICI 182,780 (10, 100, or 1,000 nmol/Week) Exacerbated SSUV (3 × 1MEdD or 3 × 0.5MEdD) Suppression of CHS Dose-Dependently.

To demonstrate that the ER normally plays a role in regulating photoimmune suppression, mice were treated topically with increasing concentrations of ICI 182,780 (Fig. 3). After irradiation with 3 × 1MEdD of SSUV, topical ICI 182,780 treatment with 10, 100, or 1,000 nmol/week resulted in a dose-dependent and statistically significant progressive exacerbation of the SSUV-suppressed CHS response from 65.1% to 72.8% (P < 0.05), 73.6% (P < 0.05), and 87.7% (P < 0.01) suppression, respectively. If mice were irradiated with a reduced SSUV dose (3 × 0.5MEdD) that alone was not significantly immunosuppressive (9.5% suppression), ICI 182,780 treatment with 10 or 100 nmol/week resulted in the development of a highly significant (P < 0.001) degree of photoimmune suppression (28% and 30.7% suppression, respectively), whereas 1.0 nmol/week ICI 182,780 had no effect. This finding provided direct evidence for an ameliorating role for actions of the ER in determining the severity of a photoimmune suppressive response. Therefore, the possible effects of exogenous treatment with a natural estrogen, 17-β-estradiol, were examined.

Fig. 3.

Fig. 3.

Open in a new tab

CHS reactions in groups of five mice irradiated with 3 × 1MEdD SSUV (UV) and treated with 10, 100, or 1,000 nmol/week ICI 182,780. Mice were also irradiated with the lower dose of 3 × 0.5MEdD and treated with 1.0, 10.0 or 100 nmol/week ICI 182,780. Matched symbols indicate treatments that were significantly different.

Topical 17-β-Estradiol (1, 10, or 100 nmol/Week) Protected Dose-Dependently Against 3 × 1MEdD of SSUV or Exogenous cis-Urocanic Acid Application.

Mice were immunosuppressed with either 3 × 1MEdD of SSUV or topically applied cis-urocanic acid. Treatment with topical estradiol/acetone at 1, 10, or 100 nmol/week was not immunomodulatory alone but was shown dose-dependently to inhibit the SSUV-induced immunosuppression (Fig. 4). At the lowest concentration, 1 nmol/week, estradiol appeared to be slightly immunoprotective against both SSUV and cis-urocanic acid, but this did not reach statistical significance. However, the higher concentrations of estradiol significantly reversed the suppression of CHS. The cis-urocanic acid-induced immunosuppression was more effectively prevented (P < 0.001) by estradiol than the SSUV-induced immunosuppression (P < 0.01), and 100 nmol/week estradiol restored the CHS response (5.9% suppression; not significantly different from “Nil” treatment) to normal. By comparison, in SSUV-exposed mice 100 nmol/week estradiol reduced the suppression of CHS from 42.9% to 23.2%. Nevertheless, the protection provided by topical estradiol was notably less than protection by 20 μM equol, equivalent to 20 nmol/week per mouse (Fig. 2), suggesting that biweekly application of estradiol in acetone solution was less bioavailable than week–daily application of the phytoestrogen in a lotion vehicle.

Fig. 4.

Fig. 4.

Open in a new tab

CHS reactions in groups of five mice immunosuppressed with either 3 × 1MEdD SSUV (UV) or cis-urocanic acid application (CIS) and treated with 1, 10, or 100 nmol/week topical estradiol/acetone. Matched symbols indicate treatments that were significantly different.

ER Blockade by ICI 182,780 (10 nmol/Week) Inhibited the Estradiol (10 nmol/Week) Protection Against SSUV.

To confirm that the immunoprotective action of exogenous estradiol treatment was mediated by means of the ER, mice irradiated with 3 × 1MEdD of SSUV received topical estradiol/lotion, followed or not by topical ICI 182,780 (Fig. 5). Unirradiated mice treated with estradiol or the combination of estradiol and ICI 182,780 displayed normal CHS responses, similar to the previous observations with equol and ICI 182,780 (Fig. 2). The SSUV irradiation resulted in 66.0% suppression of CHS, and this was markedly reduced to 13.8% suppression by topical estradiol/lotion. The more effective inhibition of photoimmune suppression here by 10 nmol/week estradiol, compared with Fig. 4, suggests that administration in a lotion base resulted in increased epicutaneous availability, and the protection appeared to be superior to that provided by 10 μM equol (10 nmol/week) (Fig. 1), and similar to 20 μM equol (20 nmol/week) (Fig. 2). However, the concurrent topical treatment with the ER antagonist markedly inhibited the estradiol immune protection, and the mice remained strongly photoimmune-suppressed (46.3% suppression). This finding is consistent with the exogenous estradiol acting via the ER and is strong support for a hitherto-unrecognized role of the ER and its endogenous ligand estradiol, in protection against the photoimmune-suppressed state.

Fig. 5.

Fig. 5.

Open in a new tab

CHS reactions in groups of five mice immunosuppressed with 3 × 1MEdD SSUV (UV) and treated topically with 10 nmol/week estradiol/lotion alone or combined with 10 nmol/week ICI 182,780. Matched symbols indicate treatments that were significantly different.

Discussion

These studies in the hairless mouse have revealed that estrogen-dependent pathways play a role in the protection of the skin against UV-induced damage. The phytoestrogen equol has proven to be a very useful tool in revealing a hitherto-unrecognized pathway by which photodamage to the immune system might normally be ameliorated endogenously by means of the ER. Thus, we have demonstrated a dose-dependent antagonism of the photoimmune-protective effect of equol by the blockade of the ER with ICI 182,780. When the phytoestrogen was replaced by exogenous estradiol treatment, a dose-dependent photoimmune protection against both SSUV and its immunosuppressive cutaneous mediator, cis-urocanic acid, was established. Because cis-urocanic acid-induced immune suppression was more effectively inhibited by ER blockade than SSUV-induced immune suppression, it appeared that this immunosuppressive mediator was the target, rather than other immunosuppressive factors known to contribute to photoimmune suppression such as histamine, prostaglandins, or various cytokines (3). The downstream effects of cis-urocanic acid that culminate in photoimmune suppression have not been clarified, but our data suggest that this mediator might function to inhibit ER signaling or its responses. The subsequently demonstrated inhibition by ICI 182,780 of the photoimmune-protective property of the natural estrogen 17-β-estradiol confirmed that the ER-associated pathway played an important role. Moreover, ER blockade by ICI 182,780 alone was found to exacerbate photoimmune suppression in a dose-dependent manner, adding support to the physiological relevance of the ER in this function.

It remains unknown whether this action of the ER is mediated by antioxidant response elements. However, cutaneous MT might have such a role, being a recognized antioxidant and a photoimmune protectant in both mice and humans, as well as a readily inducible factor by topical equol treatment in SSUV-irradiated normal-haired mouse skin (9, 10). There is now evidence from cultured human tumor cell lines that the MT gene is regulated by the ER (17) and that the zinc-finger domain of the ER is supplied with zinc by MT (18), demonstrating an intimate intracellular relationship between MT and estrogenic activity. In addition, two other phytoestrogens, genistein and apigenin, have, like equol and NV-07α, also been shown to stimulate MT expression, and this was preferentially mediated by the ER-β in cultured human cells (19, 20). The receptor specificity is interesting because ER-β distribution differs from that of the classical ER-α and is expressed in several nonclassical estrogen target tissues including the skin of the mouse (21) and humans (22). It will be of interest to identify the ER subtype in the skin that is involved in the photoimmune regulatory pathway.

In humans there is good evidence for a natural gender difference in immune responsiveness, reviewed recently (23), that may be highly relevant. Physiological levels of estradiol are immunostimulatory, whereas high levels such as occur during pregnancy are immunosuppressive (24–26). In contrast, testosterone at all concentrations is immunosuppressive, and males have a less vigorous cellular and humoral immune response than females (27). Sex steroid receptors occur in many tissues including immune cell populations such as lymphocytes, monocytes, and macrophages, suggesting that estrogen could directly affect cellular immunity and cytokine production. The cytokine IL-6 seems to play a key role in estrogen regulation of immune function (28–30), and this cytokine is also active in up-regulating the transcription of both cutaneous antioxidants, MT and heme oxygenase-1 (31, 32), and is released after UV irradiation of human skin (33), supporting the immunological link among immune mediators, estrogen, and protection from UV-induced oxidative stress.

Meanwhile, it remains unknown whether these induced antioxidants can also protect the epidermal DNA from oxidative photolesions known to contribute to photocarcinogenesis and whether the established anticarcinogenic effect of equol in mice (4) can be related to these processes. The phytoestrogen genistein has been shown to have antitumor promotional effects in vivo and in vitro attributed to the inhibition of the production of oxidative radical species (34–36). Because both genistein and estradiol, but not the nonsteroidal oleanolic acid or protocatechuic acid, were reported to protect DNA from the induction of 8-hydroxy-deoxyguanosine by UVC radiation in vitro (36), an ER-dependent mechanism was implicated. In support are very recent studies in the rat reporting that the relevant DNA repair enzyme, 8-oxoguanine glycosylase, was significantly decreased in neuronal tissue after ovariectomy (37). Further studies of this pathway in the skin are needed.

Because the ER appears to play a major regulatory role against photodamage in the skin, it is consistent that there are epidemiological data, although scant, indicating a relationship between estrogen and skin cancer. Two studies are compelling, providing persuasive evidence that females are at less risk of developing nonmelanoma skin cancer than males (38, 39). For melanoma the evidence is controversial. Increased risk of melanoma and worse prognosis has been reported associated with hormone replacement therapy and oral contraceptives and during pregnancy, when estrogen levels are elevated ≈100-fold (40, 41), but women had a better prognosis than men. In contrast, pregnancy or exogenous estrogen therapies had no effect in other studies (42). Estrogen involvement in melanoma regulation is thus unclear at present.

From the data in this study we conclude that the ER and estrogenic pathways have a natural function in immune protection against cutaneous photodamage that may influence the development of skin cancer in humans chronically overexposed to sunlight. Strategies like topical equol treatment, which protect against transient photoimmune suppression measured after acute UV exposure as in this study, have been shown in mice to reduce photocarcinogenesis after chronic irradiation (4, 5). On the other hand, it has been suggested that transient photoimmune suppression is a normal physiological reaction intended to prevent development of autoimmunity to cutaneous “photoantigens” induced by UV exposure (43). Nevertheless, we suggest that phytoestrogens like equol may have a useful role as potential supplementary ingredients in topical sun-protective cosmetic products for humans particularly susceptible to nonmelanoma skin cancer. It is recognized that several UV-absorbing sunscreen ingredients have demonstrable estrogenic properties in cultured cells and rodents (44, 45), but whether they disrupt normal endocrinology in humans is not yet clear (46, 47). There may be a need for gender-specific photoprotective therapy in males, who can be anticipated to be deficient in this newly identified protective pathway. Of greater consequence, this pathway of immunoregulation in the skin might be helpful for patients on long-term pharmacological immunosuppression regimes, such as organ transplant recipients, who are exceedingly susceptible to serious skin cancer development that can often be alleviated only by termination of their suppressive drug therapy.

Materials and Methods

Mice.

Inbred female albino hairless Skh:hr-1 mice aged 6–10 weeks were produced from the Veterinary Science breeding colony and maintained at 25°C under gold lighting (GEC F40 GO) that does not emit UV radiation, under conventional animal house conditions. From at least 1 week before experiments they were fed isoflavone-free rodent pellets (Gordon’s Specialty Stockfeeds, Yanderra, NSW, Australia) and tap water ad libitum. All procedures were approved by the University of Sydney Animal Ethics Committee and conform to the current Animal Welfare Act of New South Wales.

SSUV Radiation.

SSUV radiation (290–400 nm) was provided by a planar bank of fluorescent tubes composed of six UVA tubes (40-W F40T 10/BL; Hitachi, Tokyo, Japan) flanking one UVB tube (TL 40W/12RS; Philips, Eindhoven, The Netherlands), and the radiation was filtered through 0.125-mm cellulose acetate sheeting to remove nonenvironmental wavelengths of <290 nm. The spectral output from this source has been previously characterized (48) and emits 3.45 × 10−3 W/cm2 UVA (320–400 nm) and 1.94 × 10−4 W/cm2 UVB (290–320 nm) when measured with an Optronics OL754 integrating spectroradiometer (Optronics, Orlando, FL). Mice were exposed to 1MEdD on days 1–3, this having been previously established as the lowest exposure resulting in a statistically significant (P < 0.05) 24-h increase in skin fold thickness, when measured with a spring micrometer (Interapid, Geneva, Switzerland), comprising 22.49 kJ/m2 UVA and 1.26 kJ/m2 UVB.

Topical Treatment.

Equol (4′,7-dihydroxyisoflavane; Sigma Chemical, St. Louis, MO) was dissolved in DMSO and diluted to the required concentration in a base lotion consisting of an innocuous oil-in-water cosmetic emulsion as previously described (1), so that the final lotion contained 10–20 μM equol and 0.5% vol/vol DMSO. The control lotion contained only 0.5% DMSO. Volumes of 0.2 ml were applied to the mouse dorsum and spread with a gloved finger. Lotion application was week–daily for 1 week before commencing exposure to SSUV or treatment with cis-urocanic acid, then immediately after irradiation or after cis-urocanic acid application each day, then continuing daily until contact sensitization, providing 10–20 nmol/week equol per mouse.

Published doses for topical administration to mice of estrogen or the ER antagonist ICI 182,780 (Tocris Bioscience, Ellisville, MO) were not available and were devised from data of others describing biweekly s.c. administration (49). The ICI 182,780 was dissolved in acetone, and 17-β-estradiol (Sigma Chemical) was dissolved in acetone or the oil-in-water lotion vehicle as for equol. Aliquots of 0.1 ml were applied to the dorsal skin twice weekly, from 1 week before exposure to SSUV or cis-urocanic acid, then continuing twice weekly until contact sensitization, providing between 10 and 1,000 nmol/week ICI 182,780 per mouse or between 1 and 100 nmol/week estradiol per mouse. To avoid possible UV-screening effects of the steroid, these were applied on non-UV irradiation days or immediately after the UV exposures.

cis-urocanic acid was prepared from trans-urocanic acid (Sigma Chemical) as the photostationary mixture of ≈50% of each of the trans and cis isomers by SSUV irradiation of a thin-layer solution in DMSO followed by dilution in base lotion as previously described (50). Volumes of 0.1 ml containing 0.2% wt/vol photoisomerized urocanic acid (here referred to as “cis-urocanic acid”) with 5% DMSO were applied to the dorsal skin on three consecutive days. The control lotion contained only 5% DMSO.

Induction of CHS.

The CHS response was induced in treatment groups of five mice by sensitization with 0.1 ml of freshly prepared 2% wt/vol oxazolone (Sigma Chemical) in ethanol applied to the unirradiated abdominal skin on days 8 and 9 after the first SSUV or cis-urocanic acid treatment and thus indicates the systemic alterations in this immune reaction. The mice were then challenged by the application of 5 μl of fresh 2% oxazolone/ethanol to each surface of the pinnae. Ear thickness was measured before and repeatedly between 16 and 24 h after challenge to ascertain the peak swelling time point in the unirradiated control mice. The group average ear swelling was obtained at this time point for each treatment group, and the percentage suppression of the normal unirradiated CHS response was calculated. Statistical significance of the differences between treatments was obtained by using Student’s paired t test. No attempt was made to define the stage of the estrus cycle in the mice during the studies, which can account for the variability in the control CHS reactions between experiments (average ear swelling varied between 29.0 and 39.2 mm × 0.01) and in the degree of photoimmune suppression induced by 3 × 1MEdD of SSUV (varied between 42.9% and 72.7%). The small within-group variability in each experiment indicates estrus synchrony in grouped female mice that were within pheromonal contact with adjacently housed male mice (51).

Ovariectomy.

Mice were anesthetized with tribromoethanol injected i.p., and the ovaries were accessed via the dorsal approach. Each ovary was removed following a suture between the fallopian tube and the uterus, and the abdominal wall and skin incisions were closed with nylon sutures. The nonovariectomized mice were anesthetized and sham-operated. The mice were left to recover for 4 weeks before further studies were performed.

Acknowledgments

We are grateful to the University of Sydney Sesqui Research Grants Scheme for financial support of this project. S.W. was the recipient of a joint Indonesian–Australian AusAID scholarship for her Ph.D. studies.

Abbreviations

CHS

contact hypersensitivity

ER

estrogen receptor

MT

metallothionein

SSUV

solar-simulated UV radiation

3 × 1MEdD

three times the minimum edematous dose

3 × 0.5MEdD

three times half the minimum edematous dose.

Footnotes

Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

References

  • 1.Widyarini S., Spinks N., Husband A. J., Reeve V. E. Photochem. Photobiol. 2001;74:464–470. doi: 10.1562/0031-8655(2001)074<0465:icfrct>2.0.co;2. [DOI] [PubMed] [Google Scholar]
  • 2.Friedmann A. C., Halliday G. M., Barnetson R. S., Reeve V. E., Walker C., Patterson C. R. S., Damian D. L. Photochem. Photobiol. 2004;80:416–421. doi: 10.1562/0031-8655(2004)080<0416:TTINRS>2.0.CO;2. [DOI] [PubMed] [Google Scholar]
  • 3.Ullrich S. E. Mutat. Res. 2005;571:185–205. doi: 10.1016/j.mrfmmm.2004.06.059. [DOI] [PubMed] [Google Scholar]
  • 4.Widyarini S., Husband A. J., Reeve V. E. Photochem. Photobiol. 2005;81:32–37. doi: 10.1562/2004-06-02-RA-183. [DOI] [PubMed] [Google Scholar]
  • 5.Afaq F., Adhami V. M., Mukhtar H. Mutat. Res. 2005;571:53–173. doi: 10.1016/j.mrfmmm.2004.07.019. [DOI] [PubMed] [Google Scholar]
  • 6.Arora A., Nair M. G., Strasburg G. M. Arch. Biochem. Biophys. 1998;356:133–141. doi: 10.1006/abbi.1998.0783. [DOI] [PubMed] [Google Scholar]
  • 7.Mitchell J. H., Gardner P. T., McPhail D. B., Morris P. C., Collins A. R., Duthie G. G. Arch. Biochem. Biophys. 1998;360:142–148. doi: 10.1006/abbi.1998.0951. [DOI] [PubMed] [Google Scholar]
  • 8.Sierens J., Hartley J. A., Campbell M. J., Leathem A. J., Woodside J. V. Teratogen. Carcinogen. Mutagen. 2002;22:227–234. doi: 10.1002/tcm.10015. [DOI] [PubMed] [Google Scholar]
  • 9.Widyarini S., Allanson M., Gallagher N., Pedley J., Boyle G., Parsons P. G., Whiteman D. C., Walker C., Reeve V. E. J. Invest. Dermatol. 2006;126:198–204. doi: 10.1038/sj.jid.5700013. [DOI] [PubMed] [Google Scholar]
  • 10.Reeve V. E., Nishimura N., Bosnic M., Michalska A. E., Choo K. H. A. Immunology. 2000;100:399–404. doi: 10.1046/j.1365-2567.2000.00026.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Reeve V. E., Tyrrell R. M. Proc. Natl. Acad. Sci. USA. 1999;96:9317–9321. doi: 10.1073/pnas.96.16.9317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Matsumura A., Ghosh A., Pope G. S., Darbre P. D. J. Steroid Biochem. Mol. Biol. 2005;94:431–443. doi: 10.1016/j.jsbmb.2004.12.041. [DOI] [PubMed] [Google Scholar]
  • 13.Mueller S. O., Simon S., Chae K., Metzler M., Korach K. S. Toxicol. Sci. 2004;80:14–25. doi: 10.1093/toxsci/kfh147. [DOI] [PubMed] [Google Scholar]
  • 14.Setchell K. D., Clerici C., Lephart E. D., Cole S. J., Heenan C., Castellani D., Wolfe B. E., Nechemias-Zimmer L., Brown M. N., Lund T. D., et al. Am. J. Clin. Nutr. 2005;81:1072–1079. doi: 10.1093/ajcn/81.5.1072. [DOI] [PubMed] [Google Scholar]
  • 15.Harris D. M., Besselink E., Henning S. M., Go V. L., Heber D. Exp. Biol. Med. 2005;230:558–568. doi: 10.1177/153537020523000807. [DOI] [PubMed] [Google Scholar]
  • 16.Falkenstein E., Tillmann H.-C., Christ M., Feuring M., Wehling M. Pharmacol. Rev. 2000;52:513–555. [PubMed] [Google Scholar]
  • 17.Harris H., Henderson R., Bhat R., Komm B. Endocrinology. 2001;142:645–652. doi: 10.1210/endo.142.2.7952. [DOI] [PubMed] [Google Scholar]
  • 18.Cano-Gauci D. F., Sarkar B. FEBS Lett. 1996;386:1–4. doi: 10.1016/0014-5793(96)00356-0. [DOI] [PubMed] [Google Scholar]
  • 19.Hua P., Tsai W. J., Kuo S. M. Biochim. Biophys. Acta. 2003;1627:63–70. doi: 10.1016/s0167-4781(03)00075-7. [DOI] [PubMed] [Google Scholar]
  • 20.Miller C. P., Collini M. D., Harris H. A. Bioorg. Med. Chem. Lett. 2003;13:2399–2403. doi: 10.1016/s0960-894x(03)00394-9. [DOI] [PubMed] [Google Scholar]
  • 21.Ohnemus U., Uenalan M., Conrad F., Handjiski B., Mecklenburg L., Nakamura M., Inzunza J., Gustafsson J. A., Paus R. Endocrinology. 2006;146:1214–1225. doi: 10.1210/en.2004-1219. [DOI] [PubMed] [Google Scholar]
  • 22.Pelletier G., Ren L. Histol. Histopathol. 2004;19:629–636. doi: 10.14670/HH-19.629. [DOI] [PubMed] [Google Scholar]
  • 23.Kovacs E. J., Messingham K. N., Gregory M. S. Mol. Cell. Endocrinol. 2002;193:129–135. doi: 10.1016/s0303-7207(02)00106-5. [DOI] [PubMed] [Google Scholar]
  • 24.Thong Y. H., Steele R. W., Vincent M. M., Hensen S. A., Bellanti J. A. N. Engl. J. Med. 1973;28:604–606. doi: 10.1056/NEJM197309202891203. [DOI] [PubMed] [Google Scholar]
  • 25.Finn R., Hill C. A., Govan A. J., Ralfs I. G., Gurney F. G. Br. J. Med. 1972;3:150–152. doi: 10.1136/bmj.3.5819.150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Purtilo D. T., Hallgren H. M., Yunis E. J. Lancet. 1972;754:769–771. doi: 10.1016/s0140-6736(72)90522-3. [DOI] [PubMed] [Google Scholar]
  • 27.Ansar-Ahmed S., Penhale W. J., Tala N. Am. J. Pathol. 1985;121:531–551. [PMC free article] [PubMed] [Google Scholar]
  • 28.Girasole G., Jilka R. L., Passeri G., Boswell S., Boder G., Williams D. C., Manolagas S. C. J. Clin. Invest. 1992;89:883–891. doi: 10.1172/JCI115668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Gregory M. S., Faunce D. E., Diffner L. A., Kovacs E. J. J. Leukocyte Biol. 2000;67:319–326. [PubMed] [Google Scholar]
  • 30.Jilka R. L., Hangoc G., Girasole G., Passeri G., Williams D. C., Abrams J. S., Boyce B., Broxmeyer H., Manolagas S. C. Science. 1992;257:88–91. doi: 10.1126/science.1621100. [DOI] [PubMed] [Google Scholar]
  • 31.Nishimura N., Reeve V. E., Nishimura H., Satoh M., Tohyama C. J. Invest. Dermatol. 2000;114:343–348. doi: 10.1046/j.1523-1747.2000.00862.x. [DOI] [PubMed] [Google Scholar]
  • 32.Tyrrell R. M. BioEssays. 1996;18:139–148. doi: 10.1002/bies.950180210. [DOI] [PubMed] [Google Scholar]
  • 33.Urbanski A., Schwarz T., Neuner P., Krutmann J., Kirnbauer R., Kock A., Luger T. A. J. Invest. Dermatol. 1990;94:808–811. doi: 10.1111/1523-1747.ep12874666. [DOI] [PubMed] [Google Scholar]
  • 34.Giles D., Wei H. Nutr. Cancer. 1997;29:77–82. doi: 10.1080/01635589709514605. [DOI] [PubMed] [Google Scholar]
  • 35.Wei H., Cai Q., Rahn R. Carcinogenesis. 1996;17:73–77. doi: 10.1093/carcin/17.1.73. [DOI] [PubMed] [Google Scholar]
  • 36.Wei H., Wei L., Frenkel K., Bowen R., Barnes S. Nutr. Cancer. 1993;20:1–12. doi: 10.1080/01635589309514265. [DOI] [PubMed] [Google Scholar]
  • 37.Araneda S., Pelloux S., Radicella J. P., Angulo J., Kitahama K., Gysling K., Forray M. I. Neuroscience. 2005;136:135–146. doi: 10.1016/j.neuroscience.2005.06.080. [DOI] [PubMed] [Google Scholar]
  • 38.Preston D. S., Stern R. S. N. Engl. J. Med. 1992;327:1649–1662. doi: 10.1056/NEJM199212033272307. [DOI] [PubMed] [Google Scholar]
  • 39.Foote J. A., Harris R. B., Giuliano A. R., Roe D. J., Moon T. E., Cartmel B., Alberts D. S. Int. J. Cancer. 2001;95:7–11. doi: 10.1002/1097-0215(20010120)95:1<7::aid-ijc1001>3.0.co;2-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Holly E. A., Cress R. D., Ahn D. K. Cancer Epidemiol. Biomarkers Prev. 1994;3:661–668. [PubMed] [Google Scholar]
  • 41.Mor Z., Caspi E. Clin. Dermatol. 1997;15:147–154. doi: 10.1016/s0738-081x(96)00117-4. [DOI] [PubMed] [Google Scholar]
  • 42.Grin C. M., Driscoll M. S., Grant-Kels J. M. Semin. Cutan. Med. Surg. 1998;17:167–171. doi: 10.1016/s1085-5629(98)80010-2. [DOI] [PubMed] [Google Scholar]
  • 43.van de Pas C. B., Kelly D. A., Seed P. T., Young A. R., Hawk J. L., Walker S. L. J. Invest. Dermatol. 2004;122:295–299. doi: 10.1046/j.0022-202X.2004.22201.x. [DOI] [PubMed] [Google Scholar]
  • 44.Heneweer M., Muusse M., van den Berg M., Sanderson J. T. Toxicol. Appl. Pharmacol. 2005;208:170–177. doi: 10.1016/j.taap.2005.02.006. [DOI] [PubMed] [Google Scholar]
  • 45.Morohoshi K., Yamamoto H., Kamata R., Shiraishi F., Koda T., Morita M. Toxicol. In Vitro. 2005;19:457–469. doi: 10.1016/j.tiv.2005.01.004. [DOI] [PubMed] [Google Scholar]
  • 46.Janjua N. R., Mogensen B., Andersson A. M., Petersen J. H., Henriksen M., Skakkebaek N. E., Wulf H. C. J. Invest. Dermatol. 2004;123:57–61. doi: 10.1111/j.0022-202X.2004.22725.x. [DOI] [PubMed] [Google Scholar]
  • 47.Waring R. H., Harris R. M. Mol. Cell. Endocrinol. 2005;244:2–9. doi: 10.1016/j.mce.2005.02.007. [DOI] [PubMed] [Google Scholar]
  • 48.Reeve V. E., Greenoak G. E., Gallagher C. H., Canfield P. J., Wilkinson F. J. Aust. J. Exp. Biol. Med. Sci. 1985;63:655–665. doi: 10.1038/icb.1985.69. [DOI] [PubMed] [Google Scholar]
  • 49.Dhaher Y. Y., Greenstein B. D., Khamashta M. A., Hughes G. R. Autoimmunology. 2001;33:237–243. [PubMed] [Google Scholar]
  • 50.Reeve V. E., Bosnic M., Rozinova E. Immunology. 1993;78:99–104. [PMC free article] [PubMed] [Google Scholar]
  • 51.Whittingham D. G., Wood M. J. In: The Mouse in Biomedical Research. Foster H. L., Small J. D., Fox J. G., editors. Vol. III. New York: Academic; 1983. pp. 137–164. [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES