The Fluorescence Polarisation (FP) assay described here works through the binding of two partners of differing size, a small fluorescent peptide (< 1500 Da) to a larger target protein (> 10,000 Da). When the target protein binds to the peptide, a large complex forms, rotating slowly in solution. The rotation is not fast enough to disrupt the light, so it remains polarised (Figure 1). When a small molecule (or hit compound) displaces the fluorescent peptide, it falls into solution and begins to spin rapidly disrupting the light and forcing it to become depolarised (Figure 1).
Step 1. Titrate the fluorescent peptide alone
Titrate the fluorescent peptide to determine the concentration at which the polarisation signal remains stable and has low variability. At low concentration, there may be higher variability as the fluorescent signal nears the limits of sensitivity of the plate reader. Choose the lowest concentration of peptide at which the signal remains stable, shows low variability and gives FI at least 10-fold higher than buffer only control wells.
Note: it is prudent to label your peptide with a Red (TAMRA, BODIPY) rather than a Green (Fluorescein) dye to reduce the number of compounds that will cause interference within the 405 nm range.
Graph 1. Example of a fluorescent peptide titration
Step 2. Titrate the target protein at the chosen fixed concentration of the fluorescent peptide
This step measures the binding affinity between the fluorescent peptide and target protein. A plot of milli-polarisations (mP, see glossary for the equation) versus protein concentration will produce a hyperbola curve that begins at the baseline polarisation for the free probe (approx. 30 – 50 mP) and rises to plateau at a maximal polarisation value that corresponds to 100% bound peptide (approx. 100 – 800 mP). If the Kd of the peptide is unknown several binding curves over a range of protein concentrations (1nM to 1µM) should be measured and evaluated.
Graph 2. Example of Kd calculated for fluorescent peptide binding to target protein.
For the next steps in development choose a protein concentration that yields an acceptable assay window (approx. 100 – 800 mP) at which 50% (Kd) to 80% of the peptide is bound.
Step 3. Displacement of the fluorescent peptide using an identical unlabelled peptide
This experiment will demonstrate that the unlabelled peptide can fully displace the fluorescent peptide from the target protein and whether labelling has dramatically increased or decreased the affinity of the peptide. Following addition of the unlabelled peptide the mP values should decrease to the value observed with the free fluorescent probe alone (Graph 3). Use a fixed concentration of the target protein and fluorescent peptide, chosen in steps 1 & 2. Then mix with your unlabelled peptide over a range of concentrations. The top concentration of unlabelled peptide should be at least 100X the estimated Kd and is usually approx. 100 µM. Equilibrium is usually reached within the first 5 minutes if not within the first few minutes for most binding assays so long incubation times are not normally required. Perform this experiment according to the protocol you will use for the final screen. A pre-incubation (~ 30 – 60 min) with your compound/peptide and target protein followed by incubation with the fluorescent peptide (~ 30 – 60 min) is standard.
Graph 3. Example of IC50 calculated for unlabelled peptide competitively displacing the labelled peptide from the target protein.
Step 4. DMSO tolerance & Z Prime Assessment
You have successfully shown your assay is capable to detecting compounds that bind to the same site as your fluorescent peptide, now it is time to assess the robustness and reproducibility.
As with all HTS assays it is essential to assess tolerance to DMSO and ensure a robust S:B and low variability across your plate. In general, FP binding assays are very insensitive to higher than average concentrations of DMSO. Titrate a range of concentrations of DMSO from 0 – 10 % at your chosen concentrations of target protein and fluorescent peptide to determine any impact on Z’ (Graph 4).
Graph 4. Z’ remains stable up to 10% DMSO.
Next perform a full plate assessment of S:B using the method you will use to dispense reagents during your screen, either using manual or automated additions. Alternate columns of a 384 well plate contain either max (target protein plus fluorescent peptide) or min (fluorescent peptide only) controls.
Figure 2. Example Plate Map for a full 384 well plate assessment
Table 1. Example of Max & Min mP, SD, S:B and Z’ calculated from a full 384 well plate assessment for an FP assay
Step 5. Data Analysis
You have successfully developed an FP binding assay! Time to go to the pub to celebrate. However, look closely at your data, there may be some further questions that need answering.
Is your assay practical to run? Is it sensitive enough to detect hit compounds?
An interesting paper by Huang, X. 2003, contrary to popular belief, explains that the higher the affinity (lower Kd) between your binding partners the greater the range of resolvable potency. They argue that using tight binding peptides has no impact on the ability to detect inhibitors of low or intermediate affinity but instead allows for improved detection of very potent inhibitors. In short, the lowest resolvable inhibitor potency (lowest Kd achievable) is the Kd value of the fluorescent peptide.
Figure 3. IC50 and Ki decreases with increases in peptide affinity. Taken from Huang, X. 2003. Square = 10 nM Kd, Triangles = 100 nM Kd, Inverted Triangle = 1000 nM Kd, Diamonds = 10,000 nM Kd.
If the affinity between your protein and peptide is low (µM range) it may not be practical to use your assay for a HTS. A large amount of target protein is required and the assay would be insensitive for the reasons discussed above.
To determine the IC50 use a sigmoidal dose-response (4-parameter, variable slope equation). To determine the Kd and Ki you can use Cheng-Prusoff equation (Figure 5), but is this the most appropriate equation?
The use of Cheng-Prusoff results in an overestimation of the Ki from the IC50 due to the fact FP assays are set up with a high amount of fluorescent peptide bound (50-80%). As a large amount of the fluorescent peptide is bound, you cannot substitute the [Lf] (free ligand/peptide concentration) for total ligand/peptide concentration. The Kenakin equation (Figure 6) takes into account the proportion of peptide bound allowing for accurate determination of Kd. Another possibility is the use of biophysical techniques to determine Kd following primary hit identification.
Can you remove any false positives? If you have used a green fluorophore, such as FITC, fluorescent compounds can interfere at these wavelengths resulting in false positives. Luckily, there is a quick and easy way to eliminate these. Calculate the total fluorescence (Figure 7) by taking into account the fluorescent intensity observed from both the Perpendicular and Parallel channels. If the total fluorescence is > 3 fold that observed in the control wells (protein & peptide plus vehicle only) this is indicative of compound fluorescence.
Congratulations, you are now the proud owner of an FP binding assay for HTS! Check out the references for further information on this topic.
HTS = High Throughput Screening
FP = Fluorescence Polarisation
Ligand = peptide or small compound
FI = Fluorescence Intensity
S:B = Signal (Target Protein plus Peptide) to Background (Peptide only)
mP = Millipolarisation
Figure 4. Equation to calculate milli-polarization. S = Parallel emission P = Perpendicular emission G = G Factor. Taken from Assay Guidance Manual, Eli Lilly & Company and the National Center for Advancing Translational Sciences.
Figure 5. Cheng Prusoff Equation.
Figure 6. Kenakin Equation.
Total Fluorescence Intensity = Parallel FI + 2*Perpendicular FI
Figure 7. Equation for total fluorescence intensity
Blog written by Jess Booth
Auld, R et al., 2012. Practical Use of Fluorescence Polarization in Competitive Receptor Binding Assays. Assay Guidance Manual. Eli Lilly & Company and the National Center for Advancing Translational Sciences https://www.ncbi.nlm.nih.gov/books/NBK91992/
Matthew D Hall et al., Fluorescence polarization assays in high-throughput screening and drug discovery: a review. Methods Appl. Fluoresc. 2016, 4.
Huang X. Fluorescence polarization competition assay: The range of resolvable inhibitor potency is limited by the affinity of the fluorescent ligand. J. Biomol. Screening 2003; 8:34–38.
Kenakin, TP (1993) in Pharmacologic analysis of drug/receptor interaction, 2nd ed., New York:Raven p. 483.
Roehrl, M et al., Discovery of Small-Molecule Inhibitors of the NFAT-Calcineurin Interaction by Competitive High-Throughput Fluorescence Polarization Screening. 2004. Biochemistry 2004; 43 (51):16067–16075.
Turconi S, Shea K, Ashman S, Fantom K, Earnshaw DL, Bingham RP, Haupts UM, Brown MJB, Pope A. Real experiences of uHTS: A prototypic 1536-well fluorescence anisotropy-based uHTS screen and application of well-level quality control procedures. J. Biolmol. Screening 2001;6:275–290.