TR-FRET secondary assay for HTS discovery of chemical inhibitors of anti-apoptotic protein Bfl-1
Bfl-1, also known as A1 in mice is an anti-apoptotic and NF-kB-inducible member of the Bcl-2 protein family involved in regulation of apoptosis. Due to difficulties with accomplishing targeted gene ablation in mouse models, the endogenous functions of Bfl-1 are largely unknown. Chemical inhibitors of Bfl-1 can be used as research tools for neutralizing Bfl-1 in human and mouse cells. ..more
BioActive Compounds: 82
Sanford-Burnham Center for Chemical Genomics (SBCCG)
Sanford-Burnham Medical Research Institute (San Diego, CA)
NIH Molecular Libraries Screening Centers Network (MLSCN)
Bfl-1, also known as A1 in mice is an anti-apoptotic and NF-kB-inducible member of the Bcl-2 protein family involved in regulation of apoptosis. Due to difficulties with accomplishing targeted gene ablation in mouse models, the endogenous functions of Bfl-1 are largely unknown. Chemical inhibitors of Bfl-1 can be used as research tools for neutralizing Bfl-1 in human and mouse cells.
The current assay was developed at the Sanford-Burnham Center for Chemical Genomics (SBCCG), based on FITC-Bid BH3 peptide binding to GST-Bfl-1 in the presence of Terbium-labeled anti-GST antibody. The assay is aimed to support Bfl-1 chemical probe identification through the TR-FRET assay confirmation of the results obtained in the primary fluorescence polarization (FP) assay (PubChem AID 432) performed at the BCCG.
Bfl-1 screening was performed at the Sanford-Burnham Center for Chemical Genomics (SBCCG) as part of the Molecular Library Screening Center Network (MLSCN). XO1 submission, MH077632-01, Chemical Inhibitors of anti-apoptotic protein Bfl-1, Assay Provider Dr. John C. Read, Sanford-Burnham Medical Research Institute, San Diego, CA
Bfl-1 assay materials:
1)Bfl-1 protein and FITC-Bid peptide (FITC-Ahx-EDIIRNIARHLAQVGDSMDR ) were obtained from Prof. John Reed (Sanford-Burnham Medical Research Institute, San Diego, CA)
2)Tb-anti-GST Antibody (Invitrogen, PV4216)
3)Assay Buffer: 25 mM Bis-Tris, pH 7.0, 1 mM TCEP, 0.005% Tween 20.
4)Bfl-1 working solution contained 7.4 nM GST-Bfl-1 in assay buffer
5)FITC-Bid working solution contained 5.6 nM FITC-Bid in assay buffer
6)FITC-Bid/ Tb-Ab working solution contained 5.6 nM FITC-Bid and 2.5 nM Tb-Ab in assay buffer. Solution was prepared fresh and kept on ice prior to use.
Bfl-1 dose-response screening protocol:
1)Dose-response curves contained 10 concentrations of compounds obtained using 2-fold serial dilution. Compounds were serially diluted in 100% DMSO, and then diluted with water to 10% final DMSO concentration. 4 uL of compounds in 10% DMSO were transferred into columns 3-22 of Greiner 384-well white small-volume plates (784075). Each compound concentration was assayed in duplicate wells. Columns 1-2 and 23-24 contained 4 uL of 10% DMSO.
2)8 uL of assay buffer were added to columns 1-2, which were reserved for positive controls, using a WellMate bulk dispenser (Matrix).
3)8 uL of Bfl-1 working solution was added to columns 3-24 using WellMate bulk dispenser (Matrix). Columns 23-24 represent negative control wells.
4)Plates were incubated for 1h at +4oC.
5)8 uL of freshly prepared FITC-Bid/Tb-Ab working solution was added to the whole plate using WellMate bulk dispenser (Matrix).
6)Plates were incubated for 4h at room temperature protected from direct light.
7)Fluorescence was measured on an M5 plate reader, Molecular Devices (excitation: 340 nm, emission: 490 and 520 nm, cutoff: 475 and 515 nm, respectively) in Time Resolved (TR) mode with signal integrated for 1 ms after initial delay 0.1 ms and averaged from 5 readings. The TR-FRET signal was calculated as the ratio of TR-Fluorescence at 520 nm to TR-Fluorescence at 490 nm.
8)Data analysis was performed using a sigmoidal dose-response equation through non-linear regression.
Compounds interfering with an assay often appear as positives in screening, thus resulting in false positives. False positives, e.g., artifacts are frequently assay-specific. Therefore, reconfirmation of the positive compounds with an independent assay provides an easy way to separate the true positives hits from assay-specific hits.
The compounds identified as primary screening actives in the Bfl-1 fluorescence polarization assay proceed to the dose-response confirmation stage in both FP and TR-FRET assays performed in parallel from the same compound dilution plate. For the purposes of the TR-FRET assay, compounds that demonstrate IC50 values in the range of analyzed concentrations in the assay are defined as 'active' in the outcome column.
Compounds that failed dose-response confirmation are defined as 'inactive' in the outcome field.
To take the advantage of the results of the two assays available to us, a special activity ranking system was implemented. This system takes into account IC50 values in both assays to provide clear differentiation between the compounds that are active in a single vs. both assays. Details of the scoring system applied to the Bfl-1 project is described below.
Activity scoring rules developed at Sanford-Burnham Center for Chemical Genomics were devised to take into consideration compound efficacy, the screening stage of the data and the correlation of the results between primary and secondary assays. Details of the Scoring System will be published elsewhere.
Briefly, the outline of the scoring system utilized for the Bfl-1 TR-FRET assay is as follows:
1) First tier (0-40 range) is reserved for primary screening data and therefore is not applicable in this assay.
2) Second tier (41-80 range) is reserved for dose-response confirmation data
a. Inactive compounds of the confirmatory stage are assigned a score value equal 41.
b. The score is linearly correlated with a compound#s potency and, in addition, provides a measure of the likelihood that the compound is not an artifact based on the available information.
c. The Hill coefficient is taken as a measure of compound behavior in the assay via an additional scaling factor QC:
QC = 2.6*[exp(-0.5*nH^2) # exp(-1.5*nH^2)]
This empirical factor prorates the likelihood of a target- or a pathway-specific compound effect vs. its non-specific behavior in the assay. This factor is based on the expectation that a compound with a single mode of action that achieved an equilibrium in the assay would demonstrate the Hill coefficient value of 1. Compounds deviating from that behavior are penalized proportionally to the degree of their deviation.
d. The score is correlated with the compound potency in both the FP and the TR-FRET assays through the following function that takes into account the information obtained in both assays:
where pIC50 is a negative log(10) of the IC50 value expressed in mole/L concentration units, IM is calculated as described in PubChem AID 432, and indexes (FP) and (TR-FRET) refer to the primary and the secondary assays, respectively. This equation results in the Score values above 50 for compounds that demonstrate high potency and predictable behavior in both assays. Compounds that are inactive in either of the assays or whose concentration-dependent behavior is likely to be an artifact of their behavior in the assay will generally have lower score values.
3) Third tier (81-100 range) is reserved for resynthesized true positives and their analogues
Categorized Comment - additional comments and annotations
* Activity Concentration.
Data Table (Concise)