This in vitro Estrogen Receptor (ERα) saturation and competitive binding assay uses a ligand binding domain (LBD) of the human ERα (hrERα). This protein construct was produced by the Chemicals Evaluation Research Institute (CERI), Japan, and exists as a glutathione-S-transferase (GST) fusion protein, and is expressed in E. coli. The CERI protocol underwent an international multi-laboratory validation study (2) which has demonstrated its relevance and reliability for the intended purpose of the assay.
This assay is a screening procedure for identifying substances that can bind to the hrERα. It is used to determine the ability of a test chemical to compete with 17β-estradiol for binding to hrERα-LBD. Quantitative assay results may include the IC50 (a measure of the concentration of test chemical needed to displace half of the [3H]-17β-estradiol from the hrERα) and the relative binding affinities of test chemicals for the hrERα compared to 17β-estradiol. For chemical screening purposes, acceptable qualitative assay results may include classifications of test chemicals as either hrERα binders, non-binders, or equivocal based upon criteria described for the binding curves.
The assay uses a radioactive ligand that requires a radioactive materials license for the laboratory. All procedures with radioisotopes and hazardous chemicals should follow the regulations and procedures as described by national legislation.
The “ GENERAL INTRODUCTION” and “hrER BINDING ASSAY COMPONENTS” should be read before using this assay for regulatory purposes. Definitions and abbreviations used in this TG are described in Appendix 1.
The hrERα binding assay measures the ability of a radiolabelled ligand ([3H]17β-estradiol) to bind with the ER in the presence of increasing concentrations of a test chemical (i.e. competitor). Test chemicals that possess a high affinity for the ER compete with the radiolabelled ligand at a lower concentration as compared with those chemicals with lower affinity for the receptor.
This assay consists of two major components: a saturation binding experiment to characterise receptor-ligand interaction parameters, followed by a competitive binding experiment that characterises the competition between a test chemical and a radiolabelled ligand for binding to the ER.
The purpose of the saturation binding experiment is to characterise a particular batch of receptors for binding affinity and number in preparation for the competitive binding experiment. The saturation binding experiment measures, under equilibrium conditions, the affinity of a fixed concentration of the estrogen receptor for its natural ligand (represented by the dissociation constant, Kd), and the concentration of active receptor sites (Bmax).
The competitive binding experiment measures the affinity of a substance to compete with [3H]17β-estradiol for binding to the ER. The affinity is quantified by the concentration of test chemical that, at equilibrium, inhibits 50 % of the specific binding of the [3H]17β-estradiol (termed the “inhibitory concentration 50 %” or IC50). This can also be evaluated using the relative binding affinity (RBA, relative to the IC50 of estradiol measured separately in the same run). The competitive binding experiment measures the binding of [3H]17β-estradiol at a fixed concentration in the presence of a wide range (eight orders of magnitude) of test chemical concentrations. The data are then fit, where possible, to a form of the Hill equation (Hill, 1910) that describes the displacement of the radioligand by a one-site competitive binder. The extent of displacement of the radiolabelled estradiol at equilibrium is used to characterise the test chemical as a binder, non-binder, or generating an equivocal response.
Prior to routinely conducting the saturation and competitive binding assays, each new batch of hrERα should be shown to be performing correctly in the laboratory in which it will be used. A two-step process should be used to demonstrate performance. These steps are the following:
Conduct a saturation [3H]-17β-estradiol binding assay to demonstrate hrERα specificity and saturation. Nonlinear regression analysis of these data (e.g. BioSoft; McPherson, 1985; Motulsky, 1995) and the subsequent Scatchard plot should document hrERα binding affinity of the [3H]-17β-estradiol (Kd) and the number of receptors (Bmax) for a particular batch of hrERα.
Conduct a competitive binding assay using the control substances (reference estrogen (17β-estradiol), a weak binder (e.g. norethynodrel or norethindrone), and a non-binder (octyltriethoxysilane, OTES). Each laboratory should establish an historical database to document the consistency of IC50 and the relevant values for the reference estrogen and weak binder among experiments and different batches of hrERα. In addition, the parameters of the competitive binding curves for the control substances should be within the limits of the 95 % confidence interval (see Table 1) that were developed using data from laboratories that participated in the validation study for this assay (2).
[Mean ± Standard Deviation (SD) with (sample size (n) were calculated using curve fit estimates (4-parameter Hill equation) for control runs conducted in four laboratories during the validation study (see Annex N of reference 2).] [The 95 % confidence are provided as a guide for acceptability criteria.] [Testing of norethindrone was optional for Subtask 4 during validation study (see Reference 2, see Subtask 4). Thus, the mean ± SD (n) were calculated using curve fit estimates (4-parameter Hill equation) for control runs conducted in two laboratories. The range for the IC50 will be dependent upon the Kd of the receptor preparation and concentration of radiolabelled ligand used within each laboratory. Appropriate adjustment for the range of the IC50 based upon the conditions used to conduct the assay will be acceptable.]
Substance | Parameter | Mean | Standard Deviation(n) | 95 % Confidence Intervals
Lower Limit | Upper Limit
17β-estradiol | Top | 104,74 | 13,12 (70) | 101,6 | 107,9
Bottom | 0,85 | 2,41 (70) | 0,28 | 1,43
HillSlope | –1,22 | 0,20 (70) | –1,27 | –1,17
LogIC50 | –8,93 | 0,23 (70) | –8,98 | –8,87
Norethynodrel | Top | 101,31 | 10,55 (68) | 98,76 | 103,90
Bottom | 2,39 | 5,01 (68) | 1,18 | 3,60
HillSlope | –1,04 | 0,21 (68) | –1,09 | –0,99
LogIC50 | –6,19 | 0,40 (68) | –6,29 | –6,10
Norethindrone | Top | 92,27 | 7,79 (23) | 88,90 | 95,63
Bottom | 16,52 | 10,59 (23) | 11,94 | 21,10
Hill Slope | –1,18 | 0,32 (23) | –1,31 | –1,04
LogIC50 | –6,01 | 0,54 (23) | –6,25 | –5,78
See paragraphs 17 and 18 and Table 2 in “ hrER BINDING ASSAY COMPONENTS” of this test method. Each assay (saturation and competitive binding) should consist of three independent runs (i.e. with fresh dilutions of receptor, chemicals, and reagents) on different days, and each run should contain three replicates.
The concentration of active receptor varies slightly by batch and storage conditions. For this reason, the concentration of active receptor as received from the supplier should be determined. This will yield the appropriate concentration of active receptor at the time of the run.
Under conditions corresponding to competitive binding (i.e. 0,5 nM [3H]-estradiol), nominal concentrations of 0,1, 0,2, 0,4 and 0,6 nM receptor should be incubated in the absence (total binding) and presence (non-specific binding) of 1 μM unlabelled estradiol. Specific binding, calculated as the difference of total and non-specific binding, is plotted against the nominal receptor concentration. The concentration of receptor that gives specific binding values corresponding to 40 % of added radiolabel is related to the corresponding receptor concentration, and this receptor concentration should be used for saturation and competitive binding experiments. Frequently, a final hrER concentration of 0,2 nM will comply with this condition.
If the 40 % criterion repeatedly cannot be met, the experimental set up should be checked for potential errors. Failure to achieve the 40 % criterion may indicate that there is very little active receptor in the recombinant batch, and the use of another receptor batch should then be considered.
Eight increasing concentrations of [3H]17β-estradiol should be evaluated in triplicate, under the following three conditions (see Table 2):
In the absence of unlabelled 17β-estradiol and presence of ER. This is the determination of total binding by measure of the radioactivity in the wells that have only [3H]17β-estradiol.
In the presence of a 2000- fold excess concentration of unlabelled 17β-estradiol over labelled 17β-estradiol and presence of ER. The intent of this condition is to saturate the active binding sites with unlabelled 17β-estradiol, and by measuring the radioactivity in the wells, determine the non-specific binding. Any remaining hot estradiol that can bind to the receptor is considered to be binding at a non-specific site as the cold estradiol should be at such a high concentration that it is bound to all of the available specific sites on the receptor.
In the absence of unlabelled 17β-estradiol and absence of ER (determination of total radioactivity)
A 40 nM solution of [3H]-17β-estradiol should be prepared from a 1 μM stock solution of [3H]-17β-estradiol in DMSO, by adding DMSO (to prepare 200 nM) and assay buffer at room temperature (to prepare 40 nM). Using this 40 nM solution, the series of [3H]-17β-estradiol dilutions prepared, ranging from 0,313 nM to 40 nM with assay buffer at room temperature (as represented in lane 12 of Table 2). The final assay concentrations, ranging from 0,0313 to 4,0 nM, will be obtained by adding 10 μl of these solutions to the respective assay wells of a 96-well microtiter plate (see Tables 2 and 3). Preparation of assay buffer, calculation of the original [3H]-17β-estradiol stock solution based on its specific activity, preparation of dilutions and determination of the concentrations are described in depth in the CERI protocol (2).
Dilutions of unlabelled 17β-estradiol solutions should be prepared from a 1 nM 17β-estradiol stock solution by adding assay buffer to achieve eight increasing concentrations initially ranging from 0,625 μM to 80 μM. The final assay concentrations, ranging from 0,0625 to 8 μM, will be obtained by adding 10 μl of these solutions to the respective assay wells of a 96-well microtiter plate dedicated to the measurement of non-specific binding (see Tables 2 and 3). Preparation of unlabelled 17β-estradiol dilutions is described in depth in the CERI protocol (2).
The concentration of receptor that gives 40±10 % specific binding should be used (see paragraphs 12-13). The hrERα solution should be prepared with ice-cold assay buffer immediately prior to use, i.e. after all wells for total binding, non-specific binding and hot ligand alone have been prepared.
The 96-well microtiter plates are prepared as illustrated in Table 2, with 3 replicates per [3H]-17β-estradiol concentration. Volume assignment of [3H]-17β-estradiol, unlabelled 17β-estradiol, buffer and receptor are provided in Table 3.
TB total binding,
NSB non-specific binding
[3H] E2 [3H]17β-estradiol
E2 unlabelled 17β-estradiol
[The indicated concentrations here are the final concentrations in each well] [The dilutions of unlabelled E2 and [3H]E2 can be prepared in a different plate.]
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12
For measurement of TB | For measurement of NSB | For determination of hot ligand alone | | Unlabelled E2 dilutions for plate column 4-6 | [3H]E2 dilutions for plate column 1-9
A | 0,0313 nM [3H] E2+ ER | 0,0313 nM [3H] E2+ 0,0625 μM E2+ ER | 0,0313 nM | | 0,625 μM | 0,313 nM
B | 0,0625 nM [3H] E2+ ER | 0,0625 nM [3H] E2+ 0,125 μM E2+ ER | 0,0625 nM | | 1,25 μM | 0,625 nM
C | 0,125 nM [3H] E2+ ER | 0,125 nM [3H] E2+ 0,25 μM E2+ ER | 0,125 nM | | 2,5 μM | 1,25 nM
D | 0,250 nM [3H] E2+ ER | 0,250 nM [3H] E2+ 0,5 μM E2+ ER | 0,250 nM | | 5 μM | 2,5 nM
E | 0,50 nM [ H] E2+ ER | 0,50 nM [3H] E2+ 1 μM E2+ ER | 0,50 nM | | 10 μM | 5 nM
F | 1,00 nM [3H] E2+ ER | 1,00 nM [3H] E2+ 2 μM E2+ ER | 1,00 nM | | 20 μM | 10 nM
G | 2,00 nM [3H] E2+ ER | 2,00 nM [3H] E2+ 4 μM E2+ ER | 2,00 nM | | 40 μM | 20 nM
H | 4,00 nM [3H] E2+ ER | 4,00 nM [3H] E2+ 8 μM E2+ ER | 4,00 nM | | 80 μM | 40 nM
[If an LSC for microplates is used for measuring dpms, the preparation of hot ligand alone in the same assay plate of TB and NSB wells is not appropriate. The hot ligand alone should be prepared in a different plate.] [If centrifugation is used to separate DCC, the 50 μl of supernatant should be measured by LSC in order to avoid contamination of DCC.]
Lane Number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
Preparation Steps | TB Wells | NSB Wells | Hot Ligand Alone
Volume of components for reaction wells above and order to add | Buffer | 60 μl | 50 μl | 90 μl
unlabelled E2 from lane 11 in Table2 | — | 10 μl | —
[3H]E2 from lane12 in Table2 | 10 μl | 10 μl | 10 μl
hrERα | 30 μl | 30 μl | —
Total reaction volume | 100 μl | 100 μl | 100 μl
Incubation | FOLLOWING 2 HOUR INCUBATION REACTION | Quantification of the radioactivity just after the preparation. No incubation
Treatment with 0,4 % DCC | Yes | Yes | No
Volume of 0,4 % DCC | 100 μl | 100 μl | —
Filtration | Yes | Yes | No
MEASURING THE DPMS
Quantification volume added to scintillation cocktail | 100 μl | 100 μl | 50 μl
Assay microtiter plates for the determination of total binding and non-specific binding should be incubated at room temperature (22oC to 28oC) for two hours.
Following the two hour incubation period, [3H]-17β-Estradiol bound to hrERα should be separated from free [3H]-17β-Estradiol by adding 100μl an ice cold 0,4 % DCC suspension to the wells. The plates should then be placed on ice for 10 minutes and the reaction mixture and DCC suspension should be filtered, by transfer to a mictotiter plate filter, to remove DCC. A 100 μl of the filtrate should then be added to scintillation fluid in LSC vials for determination of disintegration per minute (dpms) per vial by liquid scintillation counting.
Alternatively, if a microplate filter is not available, removal of DCC can be obtained by centrifugation. A 50 μl of supernatant containing the hrERα-bound [3H]-17β-estradiol should then be taken with extreme care, to avoid any contamination of the wells by touching DCC, and should be used for scintillation counting.
The hot ligand alone condition is used for determining the disintegration per minute (dpm) of [3H]-17β-estradiol added to the assay wells. The radioactivity should be quantified just after preparation. These wells should not be incubated and should not be treated with DCC suspension but their content should be delivered directly into the scintillation fluid. These measures demonstrate how much [3H]-17β-estradiol in dpms was added to each set of wells for the total binding and non-specific binding.
The competitive binding assay measures the binding of a single concentration of [3H]-17β- estradiol in the presence of increasing concentrations of a test chemical. Three concurrent replicates should be used at each concentration within one run. In addition, three non-concurrent runs should be performed for each chemical tested. The assay should be set up in one or more 96-well microtiter plates.
When performing the assay, concurrent solvent and controls (i.e. reference estrogen, weak binder, and non-binder) should be included in each experiment. Full concentration curves for the reference estrogen and controls (i.e. weak binder and non-binder) should be used in one plate during each run. All other plates should contain (i) a high- (maximum displacement i.e. approximately full displacement of radiolabelled ligand) and medium- (approximately, the IC50) concentration of E2 and weak binder in triplicate; (ii) solvent control and non-specific binding, each in triplicate. Procedures for the preparation of assay buffer, [3H]-17β-estradiol, hrERα and test chemical solutions are described in depth in the CERI protocol (2).
The solvent control indicates that the solvent does not interact with the test system and also measures total binding (TB). DMSO is the preferred solvent. Alternatively, if the highest concentration of the test chemical is not soluble in DMSO, ethanol may be used. The concentration of DMSO in the final assay wells should be 2,05 % and could be increased up to 2,5 % in case of lack of solubility of the test chemical. Concentrations of DMSO above 2,5 % should not be used because of interference of higher solvent concentrations with the assay. For test chemicals that are not soluble in DMSO, but are soluble in ethanol, a maximum of 2 % ethanol may be used in the assay without interference.
The buffer control (BC) should contain neither solvent nor test chemical, but all of the other components of the assay. The results of the buffer control are compared to the solvent control to verify that the solvent used does not affect the assay system.
17β-estradiol (CAS 50-28-2) is the endogenous ligand and binds with high affinity to the ER, alpha subtype. A standard curve using unlabelled 17β-estradiol should be prepared for each hrERα competitive binding assay, to allow for an assessment of variability when conducting the assay over time within the same laboratory. Eight solutions of unlabelled 17β-estradiol should be prepared in DMSO and assay buffer, with final concentrations in the assay wells to be used for the standard curve spaced as follows: 10–6, 10–7, 10–8, 10–8.5, 10–9, 10–9.5, 10–10, 10–11 M. The highest concentration of unlabelled 17β-estradiol (1 μM) should serve as the non-specific binding indicator. This concentration is distinguished by the label “NSB” in Table 4 even though it is also part of the standard curve.
A weak binder (norethynodrel (CAS68-23-5), or alternate, norethindrone (CAS 68-22-4)) should be included to demonstrate the sensitivity of each experiment and to allow an assessment of variability when conducting the assay over time. Eight solutions of the weak binder should be prepared in DMSO and assay buffer, with final concentrations in the assay wells as follows: 10–4.5, 10–5.5, 10–6, 10–6.5, 10–7, 10–7.5, 10–8 and 10–9 M.
Octytriethoxysilane (OTES, CAS 2943-75-1) should be used as the negative control (non-binder). It provides assurance that the assay as run, will detect test chemicals that do not bind to the hrERα. Eight solutions of the non-binder should be prepared in DMSO and assay buffer, with final concentrations in the assay wells as follows: 10–3,10–4, 10–5, 10–6, 10–7, 10–8, 10–9, 10–10 M. Di-n-butyl phthalate (DBP, CAS 84-72-2) can be used as an alternative non-binder, but only tested up to 10–4M. The maximum solubility of DBP in the assay has been demonstrated to be 10–4M.
The amount of receptor that gives specific binding of 40±10 % should be used (see paragraphs 12-13 of Appendix 3). The hrERα solution should be prepared by dilution of the functional hrERα into ice cold assay buffer, immediately prior to use.
The final concentration of [3H]-17β-estradiol in the assay wells should be of 0,5 nM.
In the first instance, it is necessary to conduct a solubility test to determine the limit of solubility for each test chemical and to identify the appropriate concentration range to use when conducting the test protocol. The limit of solubility of each test chemical is to be initially determined in the solvent and then further confirmed under assay conditions. The final concentration tested in the assay should not exceed 1mM. Range finder testing includes a solvent control along with at least 8 log serial dilutions, starting at maximum acceptable concentration (e.g. 1 mM or lower, based upon the limit of solubility), and the presence of cloudiness or precipitate noted (see also paragraph 35 of Appendix 3). Once the concentration range for testing has been determine, a test chemical should be tested using 8 log concentrations spaced appropriately as defined in the preceding range finding test. Concentrations tested in the second and third experiments should be further adjusted as appropriate to better characterise the concentration response curve, if necessary.
Dilutions of the test chemical should be prepared in the appropriate solvent (see paragraph 25 of Appendix 3). If the highest concentration of the test chemical is not soluble in either DMSO or ethanol, and adding more solvent would cause the solvent concentration in the final tube to be greater than the acceptable limit, the highest concentration may be reduced to the next lower concentration. In this case, an additional concentration may be added at the low end of the concentration series. Other concentrations in the series should remain unchanged.
The test chemical solutions should be closely monitored when added to the assay well, as the test chemical may precipitate upon addition to the assay well. The data for all wells that contain precipitate should be excluded from curve-fitting, and the reason for exclusion of the data noted.
If there is prior existing information from other sources that provide a log(IC50) of a test chemical, it may be appropriate to geometrically space the dilutions more closely around the expected log(IC50) (i.e. 0,5 log units). The final results should show enough sufficient spread of concentrations on either side of the log(IC50), including the “top” and “bottom”, such that the binding curve can be adequately characterised.