Gender related differences in the oxidative stress response to PCB exposure in an endangered goodeid fish (Girardinichthys viviparus)

Polychlorinated biphenyls (PCBs) are persistent xenobiotics within aquatic environments, which elicit diverse toxic effects such as induction of oxidative stress. Despite numerous earlier studies, no detailed information exists on the toxic response by different sexes in fish. The aim of this study was to determine sex-linked differences in oxidative stress response and antioxidant defenses in Girardinichthys viviparus, an endangered fish endemic to Mexico, when exposed to sub-lethal concentrations of waterborne PCBs. The biological markers evaluated were lipid peroxidation (LPOX), superoxide dismutase (SOD) and catalase (CAT) activity. Adult eight-month-old specimens born in the laboratory were exposed to (1/2) of the LC0 (0.92 mg PCBs/L) in semi-hard synthetic water and sacrificed on days 1, 2, 4, 8 and 16 for biomarker assays. Sex-linked differences were observed in the control fish with respect to all three factors assayed. PCBs elicited significant (p<0.01) time- and sex-dependent LPOX levels which were higher in the case of males. In PCB-treated G. viviparus, SOD activity was depressed in both sexes and appears to return to pre-exposure levels after 16 days in males only. In contrast, CAT was significantly induced (p<0.01) in both sexes. This enzyme may be responsible for balancing oxidative stress and antioxidant defenses under experimental conditions. PCBs at sub-lethal concentrations are hazardous to both sexes of G. viviparus since these compounds are able to induce liver LPOX and changes in the antioxidant defense activities. The relationship between these biomarkers and cytochrome P450 and CYP1A induction is also discussed.     

Endocrine disrupters with (anti)estrogenic and (anti)androgenic modes of action affecting reproductive biology of Xenopus laevis: I. Effects on sex steroid levels and biomarker expression

Adult Xenopus laevis were exposed in vivo to ethinylestradiol, tamoxifen, methyldihydrotestosterone and flutamide as (anti)estrogenic and (anti)androgenic compounds, respectively, for four weeks at a concentration of 10(-8) M and to Lambro river water, a polluted river from Italy. Effects of the treatments were analysed by mRNA expression of retinol-binding protein (RBP), transferrin (TF), transthyretin (TTR) and vitellogenin (VTG) in the liver of male and female X. laevis, to analyse the potential of these genes to detect endocrine disrupting compounds (EDC) with different modes of action. In addition, plasma VTG and sex steroid levels, estradiol-17beta (E(2)) and testosterone (T), were analysed. Sex steroids were depressed by ethinylestradiol in both sexes whereas tamoxifen increased E(2) in females. The induction of VTG protein plasma levels was more pronounced at the protein level compared to hepatic VTG mRNA expression in response to estrogenic treatment but VTG mRNA expression detected both, estrogenic and antiestrogenic EDC. The mRNA expression of TF was decreased by estrogenic and increased by antiestrogenic treatment while TTR mRNA expression was down-regulated and RBP mRNA up-regulated by estrogenic exposure. The other treatments did not affect the mRNA expression of the examined genes.     

Effects of nonylphenol and phytoestrogen-enriched diet on plasma vitellogenin, steroid hormone, hepatic cytochrome P450 1A, and glutathione-S-transferase values in goldfish (Carassius auratus)

The effects of nonylphenol (NP) on plasma vitellogenin (VTG) and steroid hormone values, as well as hepatic cytochrome P450 1A (CYP1A) and glutathione-S-transferase (GST) activities, were measured in goldfish (Carassius auratus) fed a diet with a low (formulated diet, FD) or high (commercial diet, CD) content of phytoestrogens, including genistein and daidzein. Male goldfish with secondary sexual characteristics were exposed to nominal NP concentrations of 0.1, 1.0, 10, and 100 microg/L in the water for 28 days while being fed either the FD or CD diet at 1.0% of body weight daily. Plasma VTG concentration in male goldfish exposed to 100 microg of NP/L and fed FD was significantly higher than that in the FD-fed control fish at seven, 21, and 28 days. However, fish of the CD-fed group exposed to 100 microg of NP/ L had significantly higher plasma VTG concentration than did fish of the CD-fed control group at 28 days only. Moreover, plasma VTG concentration in fish of the CD-fed control group was about 100-fold higher than that in fish of the FD-fed control group. Although the estrogenic effects of a phytoestrogen-enriched diet caused a decrease in testosterone and/or 11-ketotestosterone values in the CD-fed fish, there was no dose-response relationship between androgen and amount of NP to which the FD-fed fish were exposed. Nonylphenol does not have appreciable effects on hepatic CYP1A and GST activities in male goldfish at concentrations as low as 100 microg/L. These results suggest that NP has estrogenic activity in male goldfish at the nominal concentration of 100 microg/L, and that phytoestrogens, such as genistein and daidzein, in the CD inhibit an aspect(s) of steroid release and/or synthesis common to testosterone and 11-ketotestosterone. However, results of in vivo screening assays for endocrine-disrupting chemicals may be seriously affected by phytoestrogens in the diet, depending on content or potency of estrogenic activity; therefore, we recommend use in research of a standardized, open-formula diet in which estrogenic substances have been reduced to amounts that do not alter the results of studies that are influenced by exogenous estrogens.     

Gender related differences in the oxidative stress response to PCB exposure in an endangered goodeid fish (Girardinichthys viviparus)

Polychlorinated biphenyls (PCBs) are persistent xenobiotics within aquatic environments, which elicit diverse toxic effects such as induction of oxidative stress. Despite numerous earlier studies, no detailed information exists on the toxic response by different sexes in fish. The aim of this study was to determine sex-linked differences in oxidative stress response and antioxidant defenses in Girardinichthys viviparus, an endangered fish endemic to Mexico, when exposed to sub-lethal concentrations of waterborne PCBs. The biological markers evaluated were lipid peroxidation (LPOX), superoxide dismutase (SOD) and catalase (CAT) activity. Adult eight-month-old specimens born in the laboratory were exposed to ½ of the LC₀ (0.92 mg PCBs/L) in semi-hard synthetic water and sacrificed on days 1, 2, 4, 8 and 16 for biomarker assays. Sex-linked differences were observed in the control fish with respect to all three factors assayed. PCBs elicited significant (p < 0.01) time- and sex-dependent LPOX levels which were higher in the case of males. In PCB-treated G. viviparus, SOD activity was depressed in both sexes and appears to return to pre-exposure levels after 16 days in males only. In contrast, CAT was significantly induced (p < 0.01) in both sexes. This enzyme may be responsible for balancing oxidative stress and antioxidant defenses under experimental conditions. PCBs at sub-lethal concentrations are hazardous to both sexes of G. viviparus since these compounds are able to induce liver LPOX and changes in the antioxidant defense activities. The relationship between these biomarkers and cytochrome P450 and CYP1A induction is also discussed.     

The role of vitellogenin during gestation of Girardinichthys viviparus and Ameca splendens; two goodeid fish with matrotrophic viviparity

Goodeid fish have matrotrophic viviparity, and unlike lecitotrophic fish, yolk loss forces the female to provide the nutritional requirements for embryonic development. Vitellogenin (VTG) is the yolk precursor protein synthesized in the maternal liver, but there is only circumstantial evidence regarding VTG supply during the ontogenesis of bony fish with matrotrophic viviparity. Therefore, the goal of the present study was to identify and quantify VTG during gestation of the black fin goodeid Girardinichthys viviparus and the butterfly split-fin goodeid Ameca splendens. Females at different gonadic developmental stages were selected in order to evaluate VTG mRNA expression in the maternal liver using RT-PCR; VTG quantification in maternal muscle and liver, as well as in the embryos, was done using ELISA, and immunohistochemical detection of VTG was done in the black fin goodeid. The results suggest that VTG supplies nutrients during embryonic development of both species, which have different life histories. It is possible that the transition from lecitotrophy to matrotrophic viviparity in bony fish with intraluminal gestation involved adaptive transition strategies that included changes in the relationship between oocytes and follicular cells, as well as a gradual loss of VTG synthesis during embryonic development.     

Delayed gametogenesis and progesterone levels in soft-shell clams (Mya arenaria) in relation to in situ contamination to organotins and heavy metals in the St. Lawrence River (Canada)

Wild carp, Cyprinus carpio, were sampled in January and March 2000 in a section of the Anoia River (NE Spain) known to be polluted by estrogenic compounds. At each sampling time, three groups were distinguished: (1) apparently normal males; (2) apparently normal females; and (3) affected fish. The latter were characterized by the simultaneous development of male and female tissue in their gonads at a macroscopical level (six out of 31 fish sampled at this particular point), or testicular atrophy (three out of 31). Plasmatic and hepatic vitellogenin (VTG) levels and plasma testosterone (T) and estradiol (E2) were measured to observe the particular estrogenic response of the affected fish. Moreover, the response in the xenobiotic metabolizing capacity in liver was tested. This involved the analysis of mixed function oxygenase (MFO) system such as: total cytochrome P450 content, NAD(P)H cytochrome c reductases and the associated CYP1A1, EROD activity. Also, glutathione S-transferase (GST) and UDP-glucuronosyltransferase (UDPGT) as detoxifying enzymes were measured. Our results showed: (1) a highly variable VTG content in all fish groups; (2) an increase in sex hormones content in March for the female group; and (3) an enhanced xenobiotics metabolism in the affected fish group, measured as total cytochrome P450, EROD activity in the January survey and cytosolic GST in March. The observed increase in VTG, sex hormones and in most of the enzymatic activities from January to March that could also be attributed to higher water temperature.     

Effects of 2,2'4,4'5,5'-hexachlorobiphenyl (PCB 153) on plasma sex steroids and vitellogenin in rockfish (Sebastes schlegeli)

We examined the estrogenic effect of 2,2'4,4'5,5'-hexachlorobiphenyl (PCB 153) on the rockfish, Sebastes schlegeli. We measured levels of plasma estradiol-17beta (E(2)) and testosterone (T) by radioimmunoassay (RIA). Plasma concentrations of T and 17alpha-hydroxyprogesterone in female and male fish injected with PCB 153 using two dosages (0.16 mg/kg body weight. and 0.57 mg/kg) did not differ significantly between sexes or from sham-injected controls of the same sex. Plasma concentrations of E(2) in females injected with PCB 153 (both levels) increased at 12 and 24 h. Concentrations of plasma vitellogenin (VTG) in females increased 72 h after injection with PCB 153 and reached 0.38 and 1.25 mg/mL, respectively. No VTG was detected in males injected with the same dosage. These results suggest that PCB 153 may lead to the production VTG in female rockfish through a synergistic effect with E(2), resulting in indirect disruption of the aromatization process.     

Vitellogenin-immunohistochemistry in the liver and the testis of the medaka, Oryzias latipes, exposed to 17beta-estradiol and p-nonylphenol

Vitellogenin (VTG) produced in male fish has been used for a biomarker to study endocrine disrupters. However, the characteristics of VTG produced in male fish have not been studied well. In this study, we investigated the localization of VTG in the liver and the testis of male medaka (Oryzias latipes) treated with 17beta-estradiol (E2) and p-nonylphenol (NP). The male fish were exposed to 1 microg/L E2 and 500 microg/L NP for 1-12 days. Control groups were kept in water including only vehicle. The frozen sections of the liver and the testis were stained with immunohistochemical methods using an antiserum against medaka VTG as the first antibody. In the E2 and NP treated liver, the hepatocytes showed immunoreactivity. In particular, the cytoplasm close to the cell membrane surrounding the sinusoids was strongly immunopositive. In the testis of both treatments, the interstitial tissues and the cells (spermatocytes) in the seminiferous tubules were immunopositive. The concentration of VTG became gradually higher in both tissues with longer treatments. These results suggest that germ cells in the testis treated with E2 and NP are able to incorporate and accumulate VTG.     

Reproductive effects of prenatal exposure to nonylphenol on zebrafish (Danio rerio)

Mature female and male zebrafish were separated and exposed to nonylphenol (NP) at 0.1, 1, 10, 50, 100 and 500 microg/L, respectively, for 3 weeks. Gonadosomatic index (GSI) in both sexes and vitellogenin (VTG) induction in males was measured as the bioindicators for the impairment to the parents. The results indicated that 50 microg/L of NP was the non-observed effect concentration (NOEC) for GSI and VTG induction. Afterwards, the 50 microg/L NP exposed females and males, and the control females and males were cross-wise pair-bred in the control water for one week to examine the reproductive effects. The embryonic cathepsin D (CAT D) activity, eggshell thickness, fecundity, hatching rate and malformation (vertebral column flexure) rate of offspring were determined in the four pair-bred groups. While endpoints remained unchanged in the groups with exposed males, prenatal exposure of females to 50 microg/L of NP resulted in the impairment of reproduction in groups with exposed females including inhibition of CAT D activity (P < 0.05), decrease of eggshell thickness (by 23.6%) and elevation of malformation rate (P < 0.001). These results suggested NP could induce reproductive damage to zebrafish at NOEC for parents. The results also imply that alterations of CAT D activity and eggshell thickness may be more sensitive biomarkers to indicate the reproductive effects caused by endocrine disrupting chemicals.     

辛基酚胁迫对雄性泥鳅抗氧化酶及卵黄蛋白原的影响

为研究辛基酚(OP)对雄性泥鳅抗氧化酶活性及血清卵黄蛋白原(VTG)含量的影响,将雄性泥鳅分别暴露于4种不同质量浓度OP(0.12、0.19、0.32、0.52 mg/L)中持续7、14、21 d和28 d,采用试剂盒检测肝脏超氧化物歧化酶(SOD)与过氧化氢酶(CAT)的含量,采用碱不稳定性蛋白结合磷法检测血清VTG的含量。结果表明,0.12 mg/L OP胁迫14 d,肝脏SOD和CAT含量均无显著变化,但是随着胁迫剂量增大和时间延长,SOD和CAT含量降低极其显著,在0.52 mg/L OP胁迫28 d时降到最低水平;泥鳅在0.12 mg/L OP中暴露7 d时,血清VTG含量就有极其显著升高,且随着胁迫剂量增大和时间的延长,VTG含量呈升高趋势。提示OP胁迫对SOD和CAT活性有显著的抑制作用,并随胁迫剂量增大和时间延长而抑制加剧,造成氧化损伤;OP胁迫可诱导VTG合成,并随暴露剂量增大和时间延长而诱导增强,具有明显的雌激素效应,这可能与其氧化损伤有密切关系。     

Effect of land use on pollution status and risk of fish endocrine disruption in small farmland ponds

To study whether the intensity of agricultural activities affects pesticides loads in pond environment, a large number of Belgian farmland ponds were surveyed in spring 2004. Temporal distribution of pollutants was also investigated over restricted survey ponds sampled three times round year 2007. Sedentary pond Prussian carp juveniles were also captured to determine their brain aromatase activity (AA) and plasma vitellogenin (VTG) levels. Heavy metal distribution was also examined in various pond matrices. Amongst the pesticides analysed, only herbicides were detected. Contamination of pond water by atrazine was frequently observed during spring 2004, while isoproturon and glyphosate were detected round year 2007. Levels of herbicides were inversely related to the distance of ponds to crop field, and values peaked in April or October. Absence of endocrine disruptors in pond water was confirmed by lack of modulation in VTG and AA in male fish. Heavy metals were present in all the pond matrices, but overall contamination levels were low. The results demonstrated that Belgian ponds were mainly contaminated by herbicides and that pond sedentary fish were not affected by endocrine disruptors. They also demonstrated a marked effect of land-use intensity on herbicide pollution which can be mitigated by an adjustment of the buffer zones.     

2,2',6,6'-Tetrachlorobiphenyl is estrogenic in vitro and in vivo

Polychlorinated biphenyls (PCBs) are ubiquitous environmental contaminants whose effects on biological systems depend on the number of and the positions of the chlorine substitutions. In the present study we examined the estrogenicity of the fully ortho-substituted PCB, 2,2',6,6'-tetrachlorobiphenyl (2,2',6,6'-TeCB). This PCB was chosen as the prototypical ortho-substituted PCB to test the hypothesis that ortho-substitution of a PCB with no para- or meta-chlorine-substitutions results in enhanced estrogenic activity. The results indicate that 2,2',6,6'-TeCB is estrogenic both in vitro, in the MCF-7 cell focus assay, and in vivo, in the rat uterotropic assay. The estrogenic activity elicited by the addition of 5 microM 2,2',6,6'-TeCB to the medium of MCF-7 cultures was inhibited by the estrogen receptor (ER) antagonist, LY156758, suggesting that 2,2',6,6'-TeCB or a metabolite is acting through an ER-dependent mechanism. Results from competitive binding assays using recombinant human (rh) ER indicate that 2,2',6,6'-TeCB does not bind rhERalpha or rhERbeta. A metabolite of 2,2',6,6'-TeCB, 2,2',6,6'-tetrachloro-4-biphenylol (4-OH-2,6,2',6'-TCB), does bind rhERalpha and rhERbeta and is also 10-fold more estrogenic than 2,2',6,6'-TeCB in the MCF-7 focus assay; however, this metabolite is not detected in the medium of MCF-7 cultures exposed to 2,2',6,6'-TeCB. Taken together, the results suggest that the estrogenicity observed in human breast cancer cells and the rat uterus may be due to 1) an undetected metabolite of 2,2',6,6'-TeCB binding to the ER, 2) 2,2',6,6'-TeCB binding directly to a novel form of the ER, or 3) an unknown mechanism involving the ER.     

Inhibition of human breast cancer cell proliferation with estradiol metabolites is as effective as with tamoxifen.

The anti-estrogenic substance tamoxifen is effective in the adjuvant therapy applied in human breast cancer. Since it partly exhibits estrogenic activity and has serious side-effects, however, pure anti-estrogenic compounds are being sought. In our experimental study, we compared the anti-proliferative effect of estradiol and 13 endogenous estradiol metabolites on human breast cancer cells with the effect of tamoxifen. We used MCF-7 and MDA-MB 231, the well-established estrogen receptor-positive and -negative cell lines. 4-hydroxytamoxifen, the active metabolite of tamoxifen, estradiol and 13 estradiol metabolites were tested in concentrations ranging from 3.1 to 100 microM. Incubation time was 4 days and cell proliferation was measured by means of the ATP chemosensitivity test. 4-hydroxytamoxifen showed an IC50 value of 27 microM and 18 microM in MCF-7 and MDA-MB 231 cells, respectively. Estradiol and its metabolites were anti-proliferative in both cell lines. A few A-ring metabolites were more effective in inhibiting cell proliferation than D-ring metabolites and the parent substance 17beta-estradiol. 4-OHE1, 2-MeOE1 and 2-MeOE2 were as effective in both cell lines as tamoxifen. For the first time it has been demonstrated that endogenous estradiol metabolites are equally anti-proliferative as tamoxifen in the context of human breast cancer cells. Since some of these metabolites exhibit no estrogenic activity, they are likely to be valuable in clinical studies of chemoprevention and adjuvant therapy of breast cancer.     

Assessment of anti-estrogenic activity of tamoxifen in transgenic mice expressing an enhanced green fluorescent protein gene regulated by estrogen response element

Tamoxifen is an anti-estrogenic agent for the treatment of breast cancer, while exhibiting estrogenic activity in such tissues as the uterus. This study aimed to test whether these opposite properties of tamoxifen in the uterus can be evaluated separately in vivo. We employed two transgenic murine models named, respectively, the ERE-EGFP Ar+/+ mouse and ERE-EGFP Ar-/- mouse. Both types of mice possess an enhanced green fluorescent protein (EGFP) gene regulated by four copies of estrogen response elements (EREs), while the latter lacks a functional aromatase gene, which encodes an enzyme catalyzing conversion of androgens to estrogens. Tamoxifen clearly exhibited estrogenic activity in the uteri of ERE-EGFP Ar-/- mice, as it caused uterine wet weight gain and E2-target gene induction, as 17beta-estradiol (E2) did. However, tamoxifen did not enhance the EGFP expression in ERE-EGFP Ar-/- mice, although E2 induced it significantly. In ERE-EGFP Ar+/+ mice, tamoxifen suppressed the EGFP expression in a time- and dose-dependent manner. Thus, the present study demonstrated that estrogenic and anti-estrogenic activities of tamoxifen can be evaluated by using ERE-EGFP Ar-/- and ERE-EGFP Ar+/+ mice, respectively. Furthermore, these animal models are useful to select and evaluate estrogenic and anti-estrogenic activities of chemical compounds.     

Mercury and persistent organic pollutant concentrations in African fish eagles, marabou storks, and Nile tilapia in Uganda

The purpose of this research was to evaluate persistent organic pollutant (POP) and mercury concentrations in tissues of African fish eagles (Haliaeetus vocifer) and Nile tilapia (Oreochromis niloticus) from Lake Victoria near Entebbe and Lake Mburo, Uganda. Marabou stork (Leptoptilos crumeniferus) nestlings from urban Kampala (40 km from Entebbe) also were sampled for POPs and mercury. Total mercury was measured in the breast feathers of eight nestling and 10 adult African fish eagles from Lake Mburo, 10 nestling and five adult African fish eagles from Lake Victoria near Entebbe, and 20 nestling marabou storks from Kampala from June 2002 through January 2003. Mercury concentrations in all samples were below levels associated with adverse effects in similar species. Mercury concentrations were significantly higher in eagle adults and nestlings from Entebbe than in adults and nestlings from Lake Mburo (P< or =0.05). No significant differences (P> or =0.05) were found in mercury concentrations between sexes or between the entire fish eagle population sampled at Entebbe and marabou stork nestlings sampled at nearby Kampala. Plasma samples from the same birds were analyzed for 1,1,1-trichloro-2,2-bis(4-chlorophenyl)ethane, aldrin, hexachlorocyclohexane (alpha-HCH), dieldrin, endrin, heptachlor and their metabolites, as well as total polychlorinated biphenyls (PCBs). Nile tilapia whole-body cross sections collected from Lake Mburo (n=3) and Lake Victoria near Entebbe (n=8) also were analyzed for these POPs and mercury. No samples contained POPs or PCBs at the limits of detection except for 4,4'-1,1-dichloro-2,2-bis(4-chlorophenyl)ethylene in five adult eagle plasma samples (0.0026+/-0.0015 ppm wet weight) and five Nile tilapia samples (0.002+/-0.001 ppm wet weight) from Entebbe.     

Recovery of Historically Contaminated Watercourse Polluted by the Chemical Wood Industry: EROD Activity in Fish as Biomarker

Despite outstanding process alterations over decades, pulp- and paper-mill-contaminated sediments and continuing exposure by the effluents may still have effects on biota. In this study, ecotoxicological impacts in the boreal watercourse were analyzed by measuring ethoxyresorufin-O-deethylase (EROD) induction from wild fish populations and from experimentally exposed fish. In order to assess the role of sediment-borne chemicals, juvenile rainbow trout (Oncorhynchus mykiss) were exposed in the laboratory to the surface sediments of Lake Vatianjärvi and Southern Lake Saimaa, both watercourses impacted by the chemical wood industry for approximately a century. Hepatic EROD activity was also measured from roach (Rutilus rutilus) and perch (Perca fluviatilis) caught in Lake Vatianjärvi. Increased EROD activity was not observed in wild fish caught in Lake Vatianjärvi nor in rainbow trout exposed to the sediment of Lake Vatianjärvi, but it was observed in rainbow trout exposed to the sediment of Southern Lake Saimaa, probably explained by critically high concentrations of retene. The results of both the laboratory and field study indicate the absence of EROD-inducing compounds of the previously heavily contaminated Lake Vatianjärvi.     

Induction of Hsp22 (HspB8) by estrogen and the metalloestrogen cadmium in estrogen receptor-positive breast cancer cells

Estrogen (E2) plays a critical role in the etiology and progression of human breast cancer. The estrogenic response is complex and not completely understood, including in terms of the involved responsive genes. Here we show that Hsp22 (synonyms: HspB8, E2lG1, H11), a member of the small heat shock protein (sHSP) superfamily, was induced by E2 in estrogen receptor-positive MCF-7 breast cancer cells, resulting in an elevated Hsp22 protein level, whereas it was not induced in estrogen receptor-negative MDA-MB-231 cells. This induction was prevented by the pure anti-estrogen ICI182780 (faslodex, fulvestrant), whereas tamoxifen, a substance with mixed estrogenic and antiestrogenic properties, had no major inhibitory effect on this induction, nor did it induce Hsp22 on its own. Cadmium (Cd) is an environmental pollutant with estrogenic properties (metalloestrogen) that has been implicated in breast cancer. Treatment of MCF-7 cells with Cd also resulted in induction of Hsp22, and this induction was also inhibited by ICI182780. In live MCF-7 cells, Hsp22 interacted at the level of dimers with Hsp27, a related sHSP, as was shown by quantitative fluorescence resonance energy transfer measurements. In cytosolic extracts of MCF-7 cells, most of the E2- and Cd-induced Hsp22 was incorporated into high-molecular mass complexes. In part, Hsp22 and Hsp27 were components of distinct populations of these complexes. Finally, candidate elements in the Hsp22 promoter were identified by sequence analysis that could account for the induction of Hsp22 by E2 and Cd. Taken together, Hsp22 induction represents a new aspect of the estrogenic response with potential significance for the biology of estrogen receptor-positive breast cancer cells.