Registration Dossier

Toxicological information

Additional toxicological data

Currently viewing:

Administrative data

Endpoint:
additional toxicological information
Type of information:
experimental study
Adequacy of study:
key study
Study period:
no information
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
other: Study well documented, meets generally accepted scientific principles, acceptable for assessment.
Cross-referenceopen allclose all
Reason / purpose:
reference to same study
Reason / purpose:
reference to other study

Data source

Reference
Reference Type:
publication
Title:
Estrogen-like activity of metals in MCF-7 breast cancer cells
Author:
Martin MB, Reiter R, Pham T, Avellanet YR, Camara J, Lahm M, Pentecost E, Pratap K, Gilmore BA, Divekar S, Dagata RS, Bull JL and Stoica A
Year:
2003
Bibliographic source:
Endocrinol. 144(6):2425-2436

Materials and methods

Type of study / information:
Type: other: estrogen-like activity
Principles of method if other than guideline:
The ability of metals to activate estrogen receptor-alpha (ERalpha) was measured in the human breast cancer cell line, MCF-7.
GLP compliance:
not specified

Test material

Reference
Name:
Unnamed
Type:
Constituent
Details on test material:
- Name of test material (as cited in study report): Copper, cobalt, nickel, lead, mercury, tin, and chromium or with the metal anion vanadate

Results and discussion

Any other information on results incl. tables

Effect of metals on the growth of MCF-7 cells:To determine whether metals mimic the effects of estradiol on cell proliferation, the ability of the metals to promote anchorage-dependent growth of MCF-7 cells was determined. Cells were treated with 10-9 M estradiol and/or 10-6 M metal salt in the presence or absence of 500 nm ICI 182,780, and the number of cells was counted at different times. The results are presented in Fig. 1. Copper, cobalt, nickel, lead, mercury, tin, chromium (II), and vanadate significantly stimulated the growth of MCF-7 cells compared with that of cells grown in estrogen-depleted medium. A 2- to 5-fold increase in cell number was observed after 6 d of treatment with the metals compared with a 9-fold increase in cell growth after estradiol treatment. As expected, the antiestrogen blocked estradiol-stimulated growth. The antiestrogen also blocked metal-stimulated proliferation, suggesting that the growthpromoting effects of the metals are mediated by the ER. When cells were treated with estradiol and copper or cobalt,there was an additive effect on cell growth (P < 0.05). Effect of metal treatment on the concentration of ERalpha protein To determine whether the metals demonstrate estrogenlike effects on the expression of ERalpha, an enzyme immunoassay was employed to measure the concentration of ERalpha protein. MCF-7 cells were treated for 24 h with 10_6 M metals in the presence and absence of 10-9 M estradiol (Fig. 2). In agreement with previous observations (13), treatment with estradiol resulted in a 62% decrease in ERalpha protein (13). Similar to estradiol treatment, the concentration of ERalpha decreased by approximately 40-60% from 415 nmol/mg protein in control cells to 166-249 nmol/mg protein in cells treatment with copper, cobalt, nickel, lead, mercury, tin, chromium, or vanadate. However, treatment with chromium (III) or zinc (II) did not alter the concentration of ERalpha protein, suggesting that some, but not all, metals mimic the effects of estradiol. Treatment of cells with estradiol and copper, cobalt, lead, or vanadate had an additive effect on ERalpha expression (P < 0.05). However, no additional effect on ERalpha was observed when cells were treated with estradiol and nickel, mercury, tin, or chromium (II). Effect of metals on the steady state amount of ERalpha mRNA To determine whether the reduction in ERalpha protein parallels a reduction in the steady state amount of ERalpha mRNA, an RNase protection assay was performed. MCF-7 cells were treated with 10-6M metal in the presence and absence of 10-9M estradiol, and the effects of treatment on the steady state amount of total ERalpha mRNA were measured. In this experiment the amount of ERalpha mRNA was quantified by phosphorimaging and normalized to the amount of the acidic ribosomal phosphoproteinmRNA, and the data are presented in Fig. 2 as a percentage of the control ratio of the ERalpha signal to the ribosomal protein signal. Treatment with 10_9 m estradiol resulted in a 70% decrease inERalpha mRNA.Treatment ofMCF-7cells with copper, cobalt, nickel, lead, tin, chromium, or vanadate resulted in a 40-60% decrease in ERalpha mRNA. The decrease in ERalpha mRNA correlated with the magnitude of the effect of the metals on the amount of ERalpha protein. Treatment with a combination of estradiol and copper, cobalt, or lead, but not with nickel, mercury, tin, or vanadate, resulted in an additive effect on ERalpha mRNA (P < 0.005). Effect of metal treatment on the concentration of progesterone receptor protein and mRNA To determine whether the metals also mimic the effects of estradiol on the expression of progesterone receptor protein, an enzyme immunoassay was performed. MCF-7 cells were treated with 10-6 m metal salt and/or 10-9 m estradiol for 24 h, and the concentration of progesterone receptor protein was measured (Fig. 3A). Treatment with estradiol resulted in a 6-fold increase in progesterone receptor over control values. In response to treatment with copper, cobalt nickel, lead, mercury, chromium, or vanadate, the progesterone receptor concentration increased 1.8- to 4-fold compared with control receptor protein. To determine whether the effects were mediated by ERalpha the ability of the antiestrogen ICI-182,780 (5x10_7 M) to block the effects of the metals was tested. As expected, the antiestrogen blocked the effect of estradiol. The antiestrogen also blocked the effects of the metals, suggesting that the effects of these compounds are mediated by ERalpha. Treatment of cells with both estradiol and metal had an additive effect on progesterone receptor induction (P<.05). To determine whether the metals induced progesterone receptor mRNA, an RNase protection assay was employed. Similar to the results obtained for progesterone receptor protein, the metals increased progesterone receptor mRNA (Fig. 3A). Metal induction of progesterone receptor mRNA was also inhibited by the antiestrogen and was additive with estradiol. Chromium (III) and zinc had no effect on the expression of progesterone receptor. Effect of metals on the steady state amount of pS2 mRNA To determine whether metals regulate the expression of other estrogen-responsive genes, MCF-7 cells were treated with either 10_6 m metal salt or 10_9 m estradiol for 24 h, and the amount of pS2 mRNA was measured by an RNase protection assay. In this study estradiol induced a 2-fold increase in pS2 mRNA. Similar to estradiol, all of the metals, including tin, induced pS2mRNAby 1.6- to 2.8-fold over control values (Fig. 3B). As in the case of progesterone receptor, the effects of metals on the expression of pS2 mRNA were blocked by 5 _ 10_7 m ICI-182,780, suggesting that the effects of metals are mediated by ER_. Activation of ER_ by metal To measure the ability of the metals to activate endogenously and exogenously expressed ER_, transient transfection assays were employed. In the first assay, the ability of metals to activate endogenous ER_ was tested by transiently transfecting MCF-7 cells with an estrogen response elementluciferase reporter construct (Fig. 4A). In the second assay, the ability of metals to activate exogenously expressed ER_ was measured by transiently cotransfecting COS-1 cells with a wild-type ER_ expression vector and an estrogen response element-CAT reporter construct (Fig. 4B). The transfected cells were treated with either 10_9 m estradiol or 10_6 m metal, and reporter activity was measured in the presence or absence of antiestrogen. Treatment with estradiol stimulated luciferase and CAT activities by approximately 7- to 13-fold, respectively, whereas treatment with the metals produced a 1.5- to 3-fold increase in luciferase activity and a 2- to 4-fold increase in CAT activity. The metal-induced increase in reporter activity was blocked by the antiestrogen, demonstrating that the metals activate endogenously as well as an exogenously expressed receptor. To measure the estrogenic potency of the metals, the 50% effective concentration (EC50) of each metal was determined from dose-response curves performed in MCF-7 cells transiently transfected with the luciferase reporter construct. Transfected cells were treated with concentrations of estradiol or metal from 10_12-10_5 m. The EC50, defined as the concentration that produces a 50% increase in luciferase activity, was determined for each metal (Table 1). The relative potency of the metals was then determined as the ratio of the EC50 of estradiol to the EC50 of the metal. The EC50 values for the metals ranged from 2 _ 10_9 m for nickel to 1 _ 10_8 m for copper compared with the EC50 value for estradiol of 2 _ 10_9 m with the relative potencies of the metals varying from 0.2-1.0. Interaction of metals with the hormone-binding domain of ER_ To identify the region of ER_ involved in activation by the metals, a chimeric receptor containing the hormone-binding domain of ER_ was employed. The chimeric receptor, GALER, contains the DNA-binding domain of the yeast transcription factor GAL4 fused to the hormone-binding domain of ER_. Stimulation of transcription by GAL-ER chimera from a GAL4-responsive CAT reporter gene is dependent upon estradiol (22). When the chimeric receptor GAL-ER and the GAL4-CAT reporter construct were transiently cotransfected into COS-1 cells, and the cells were treated with estradiol, there was a 4.4-fold increase in CAT activity (Fig. 4B). Treatment with 10_6 m metal resulted in a 2- to 5.8-fold increase in CAT activity, suggesting that the metals activate ER_ through the hormone-binding domain. To determine whether activation by metals requires a functional AF-2 domain, an ER_ mutant that has an intact AF-1 domain, but is mutated in the AF-2 domain, was used. The ER_ mutant, referred to as ERTAF-1, is mutated at residues 538, 542, and 545 (24).Asecond mutant containing the AF-2 mutations and also lacking the AF-1 domain (ER-null) was used as a control. The AF-2 mutants were transiently transfected with an estrogen response element-CAT reporter gene into COS-1 cells, and CAT activity was determined after treatment with estradiol or metal. Similar to estradiol, treatment of the AF-1 mutants with 10_6 m metal resulted in no increase in CAT activity (data not shown), suggesting that a functional AF-2 domain was required for activation of ER_ by the metals. To establish whether the interaction of ER_ with metals altered its interaction with coactivators, MCF-7 and CHO cells were transiently cotransfected with a splice variant of AIB1 that lacks exon 3 (AIB1 _3). This isoform of AIB1 is more effective than full-length AIB1 in promoting the transcriptional activity of ER_ in both MCF-7 and CHO cells (25). As expected, the presence of the coactivator enhanced the transcriptional activity of ER_ in MCF-7 cells upon treatment with estradiol. There was a 6-fold increase in luciferase activity in the absence of AIB1 _3 and an 18-fold increase in the presence of the coactivator (Fig. 4A). However, AIB1 _3 did not enhance the transcriptional activity of ER_ upon treatment with metals. In the presence of AIB1 _3, metal treatment did not produce a significant increase in luciferase activity compared with treatment with metals in the absence of the coactivator. In contrast to the results observed in MCF-7 cells, the coactivator produced a significant enhancement of metalinduced activation of ER_ in CHO cells. When CHO cells were transiently cotransfected with ER_ and an estrogenresponsive luciferase reporter gene in the presence and absence of AIB1 _3, the coactivator produced a 2- to 4-fold enhancement of the metal response, suggesting that metals do not interfere with the interaction of ER_ and coactivators. Activation of ER_ mutants by metals Metals are capable of forming high affinity complexes directly or indirectly with many different amino acids, including cysteines. The hormone-binding domain of ER_ contains four cysteines at positions C381, C417, C447, and C530. To test the roles of these cysteines in the interaction with the cysteine mutants C381A, C417A, C447A, and C530A were transiently cotransfected with an estrogen-responsive CAT construct into COS-1 cells, and the cells were treated with 10_9 m estradiol or 10_6 m metal. The amount of CAT activity was measured, expressed as the percent conversion, and normalized to the amount of _-galactosidase activity (Fig. 4D). After treatment with heavy metals, there was an approximately 2- to 4-fold increase in CAT activity with mutants C417A and C530A. In contrast to the effects observed with these mutants, the metals failed to activate the mutants C381A and C447A, suggesting that cysteines C381 and C447 may be involved in activation of ER_ by the metals. To demonstrate that the mutation of cysteine to alanine did not interfere with the activity of the receptor, the transiently transfected cells were treated with 10_9mestradiol. After hormone treatment, there was an approximately 10- to 17-fold increase in CAT activity with all mutants. These results corroborate previous studies employing these mutants that demonstrate that mutation of cysteines in the hormone-binding domain to alanine does not alter the ability of estradiol to trans-activate the receptor (16). In addition to cysteines, metals are capable of interacting with histidines. Histidine H524 has previously been shown to interact with estradiol (26) as well as cadmium, arsenite, and selenite (4-6). Mutation of histidine H524 to alanine resulted in the complete loss of activation of ER_ by the metals, but did not alter the ability of estradiol to trans-activate the receptor (Fig. 4E). To identify other possible interaction sites within the hormone-binding domain of ER_, aspartic acid D538 and glutamic acid E523 were mutated to asparagine and glutamine, respectively. With the exception of vanadate, the metals failed to activate the receptor, suggesting that aspartic acid D538 and glutamic acid E523 interact with the divalent metals. Vanadate activated E523Q and D538N, resulting in an approximately 4-fold increase in CAT activity. As vanadate has a negative charge, it would be expected to interact with positively charged amino acids rather than negatively charged amino acids. Possible candidate amino acids include lysines K529 and K531 and asparagine N532 that were mutated to glutamic acid and aspartic acid, respectively (17-19). Vanadate did not activate mutants K529Q K531Q N532D, K529Q K531Q, or N532D (Fig. 4F), suggesting that asparagine N532 and at least one of the lysines, K529 and/or K531, may also play a role in the interaction of vanadate with ER_. With the exception of H524, estradiol treatment of all mutants resulted in an approximately 10- to 17-fold increase in CAT activity, suggesting that mutation of these amino acids did not interfere with the activity of ER_. Estradiol treatment of H524A resulted in 7-fold induction of CAT activity, in agreement with previously published results (18). Effect of metals on the binding of estradiol to ER_ To determine whether metals blocked estradiol binding to ER_, the effects of these compounds on hormone binding were measured using a single dose ligand binding assay. Human recombinant ER_ was treated with 10_6 m metal salt for 1 h at 4 C. The ability of ER_ to bind hormone was then assayed by incubating the extract with 10 nM [3H]estradiol in the presence or absence of a 200-fold molar excess of diethylstilbestrol for 2 h at 37 C. As shown in Fig. 5, all of the heavy metals tested, with the exception of Cr (III) and Zn (II), blocked the binding of estradiol to the receptor. To further characterize the binding of the metals, a multiple dose ligand binding assay was performed. Cytosolic extracts from MCF-7 cells were treated with 10_6 m metal for 1 h at 4 C. Various concentrations of [3H]estradiol (10_12-10_7m) were added in the presence or absence of a 200-fold molar excess of diethylstilbestrol. The apparent binding affinity and binding capacity of the ER were determined according to the method of Scatchard (23) (Table 2). In the absence of metal, estradiol bound to the receptor with an equilibrium dissociation constant of 3.06_2.88_10_10m(n_3). In the presence of metal, the dissociation constant of estradiol was unchanged, but the number of binding sites decreased. These results suggest that the metals compete with estradiol for binding to ER_ in a noncompetitive manner. Metal binding to ER_ and ER_ mutants To characterize the binding of the metals to ER_, Scatchard analysis was performed using radioactive cobalt and nickel and human recombinant ER_. Recombinant ER_ was incubated with various concentrations of 57Co or 63Ni (10_12-10_7 m) in the presence and absence of a 200-fold excess of diethylstilbestrol, and the results are shown in Table 3. The metals bound to ER_ with an affinity similar to that of estradiol for the receptor. The dissociation constants for cobalt and nickel were 3 _ 10_9 and 2 _ 10_9 m, respectively. The binding of cobalt and nickel to ER_ in MCF-7 cells was also measured for comparison with the binding to recombinant receptor. Similar to recombinant receptor, the dissociation constants in whole cell extracts for cobalt and nickel were 9.5 _ 10_9 and 7 _ 10_9 m, respectively. These results demonstrate that there was no significant difference in the binding affinity of cobalt, nickel, or estradiol to ER_. To determine whether the ER_ mutants were capable of binding cobalt and nickel, the mutants were expressed in COS-1 cells, and the specific binding of 57Co and 63Ni was determined using a whole cell binding assay (Table 3). All of the mutants were capable of binding estradiol with high affinity, but at a lower affinity than the wild-type receptor. Radioactive cobalt, on the other hand, bound with high affinity to the ER_ mutants C447A, E523A, D538N, and N532D, but not to the C381A and H524A mutants. Radioactive nickel bound only to the N523D mutant. There was no detectable binding of nickel to the C381A, C447A, E523A, H524A, and D538N mutants. The inability of some of these mutants to bind metals suggests that these amino acids play a role in the binding as well as the activation of ER

Applicant's summary and conclusion

Conclusions:
The study was not conducted with Cadmium metal. However, this mechanistic study conducted with divalent metals (copper, cobalt, nickel, lead, mercury, tin, and chromium or with the metal anion vanadate) suggested that divalent metals and metal anions activate estrogen-receptor, ERalpha through the formation of a complex within the hormone-binding domain of the receptor.
Executive summary:

A mechanistic study was conducted to measure the ability of divalent metals to activate estrogen receptor-alpha (ERalpha) in the human breast cancer cell line, MCF-7.

 

Similar to estradiol, treatment of cells with the divalent metals copper, cobalt, nickel, lead, mercury, tin, and chromium or with the metal anion vanadate stimulated cell proliferation; by d 6, there was a 2- to 5-fold increase in cell number. The metals also decreased the concentration of ERalpha protein and mRNA by 40-60% and induced expression of the estrogen-regulated genes progesterone receptor and pS2 by1.6- to 4-fold. Furthermore, there was a 2- to 4-fold increase in chloramphenicol acetyltransferase activity after treatment with the metals in COS-1 cells transiently cotransfected with the wild-type receptor and an estrogen-responsive chloramphenicol acetyltransferase reporter gene. The ability of the metals to alter gene expression was blocked by an antiestrogen, suggesting that the activity of these compounds is mediated by ERalpha. In binding assays the metals blocked the binding of estradiol to the receptor without altering the apparent binding affinity of the hormone (K(d) = 10(-10) M). Scatchard analysis employing either recombinant ERalpha or extracts from MCF-7 cells demonstrated that (57)Co and (63)Ni bind to ERalpha with equilibrium dissociation constants of 3 and 9.5 x 10(-9) and 2 and 7 x 10(-9) M, respectively.

 

The ability of the metals to activate a chimeric receptor containing the hormone-binding domain of ERalpha suggests that their effects are mediated through the hormone-binding domain. Mutational analysis identified amino acids C381, C447, E523, H524, N532, and D538 as potential interaction sites, suggesting that divalent metals and metal anions activate ERalpha through the formation of a complex within the hormone-binding domain of the receptor.