# 20.109(S19):Confirm ligand binding using differential scanning fluorimetry assay (Day6)

## Contents

## Introduction

Interactions between low molecular weight ligands and proteins have been shown to increase the thermostability of proteins. This means that proteins bound to ligand are able to maintain tertiary structure, or resist denaturation, at higher temperatures than unbound proteins. Today we will use differential scanning fluorimetry (DSF) to examine the potential FKBP12 binders identified in our SMM screen.

DSF is a method used to identify low molecular weight ligands that bind and stabilize a protein of interest. In this assay, protein denaturation is measured via a fluorescent dye that has an affinity for hydrophobic regions. When the protein is folded the hydrophobic pockets are inaccessible to the dye and the fluorescent signal is quenched by water in the solution. As the protein unfolds, the dye interacts with the hydrophobic regions and emits a fluorescent signal that can be detected.

When a protein is bound to a ligand, the stability can be increased such that the temperature at which the protein denatures is increased. In the DSF assay, this is measured as a shift in the T_{m}, or melting temperature; which is defined as the temperature at which 50% of the protein is unfolded. This value represents the midpoint of the transition from structured (folded) to denatured (unfolded).

The ΔT_{m} is the difference between the T_{m} of the unbound protein sample, or protein sample without added ligand, and the bound protein sample, protein sample with added ligand. If the tested ligand binds the protein of interest, the ΔT_{m} can be observed as a shift in the plotted DSF data. For example, the data below show results of a pilot experiment completed in preparation for this module. In this graph the T_{m} of FKBP12 (blue curve) is ~50 °C. With the addition of rapamycin (red curve) the T_{m} is shifted to ~78 °C resulting in a ΔT_{m} of ~20 degrees. Data in this plot was obtained by Becky Leifer from the Koehler lab.

## Protocols

### Part 1: BE Communication Lab workshop

Our communication instructors, Dr. Sean Clarke and Dr. Prerna Bhargava, will join us today for a workshop on writing impactful abstracts and titles.

### Part 2: Prepare samples for DSF assay

As in the previous laboratory session, you will prepare master mixes for the conditions you will test. Because the master mixes for the DSF assay are more complicated, the below chart will assist you in completing the required calculations for each reaction. You will eventually make master mixes for each reaction, with enough volume to measure each in triplicate.

Reagent (stock concentration) | Final concentration of stock reagent in reaction | Volume of stock reagent in reaction |
---|---|---|

FKBP12 (191.6 μg/mL) | 1 μg/30 μL reaction | |

DMSO (3%) | 0.1% | |

rapamycin (150 μM) | 5 μM | |

ligand, [low] (90 μM) | 3 μM | |

ligand, [high] (900 μM) | 30 μM | |

dye (30X) | 1X | |

PBS (1X) | add for a total of 30 μL reaction | dependent upon master mix |

- Perform the necessary calculations to complete the above chart for a total reaction volume of 30 μL.
- Confirm your values with the teaching faculty before proceeding.

- Each team will setup triplicate reactions for 7 different conditions:
- Condition 1: no protein AND DMSO (internal control)
- Condition 2: FKBP12 AND DMSO (internal control)
- Condition 3: FKBP12 AND rapamycin (5 μM)
- Condition 4: FKBP12 AND ligand #1, (3 μM)
- Condition 5: FKBP12 AND ligand #1, (30 μM)
- Condition 6: FKBP12 AND ligand #2, (3 μM)
- Condition 7: FKBP12 AND ligand #2, (30 μM)

- Generate a chart, or list, that details what reagents will be in each master mix for Conditions #1 - #7 listed above.
- All reactions will contain dye.
- Only reactions without rapamycin or ligand will contain DMSO.
- Include the volume of each reagent (for a final volume of 3.25 the reaction volume, which is 30 μL) as each condition will be tested in triplicate.
- Again, confirm your values with the teaching faculty before proceeding.

- Obtain the appropriate aliquots from the front laboratory bench.
- Use the values calculated in Step #3 to prepare your master mixes in well-labeled 1.5 mL centrifuge tubes on ice.
- You will add all reagents
**except**FKBP12 protein, as the teaching faculty will add the protein to the samples immediately prior to measuring the fluorescence signal.

- You will add all reagents
- When you have prepared your master mixes, take them to the front laboratory bench.
- Be sure that all tubes are clearly labeled!

### Part 3: Examine binding shifts

You will receive two Excel sheets containing raw data from each well of a 384 well plate over the specified range of temperatures. The Excel sheet with "Melt Curve RFU Results" in the file name will contain raw fluorescence intensity data, while the other sheet with "Melt Curve Derivative Results" in its name will have the values for the first derivative of the melt curve. The teaching faculty will inform you which wells correspond to which conditions for your group.

One basic way to determine the "melting temperature," or T_{m} of the protein is to determine temperature at the inflection point of the melting curve. This inflection point would occur at the maximum value of the first derivative. The BioRad CFX machine we use actually exports the negative of the first derivative in the Excel file, so we will find the minimum value in the first derivative Excel file, and take the corresponding temperature to be the T_{m} of FKBP12 in each condition.

- Open the Excel file corresponding to the first derivative data
- Column B should contain temperature information in Celsius.
- At a row on the bottom of column C, type in the following command: =INDEX($B$
*FirstRow*:$B$*LastRow*, MATCH(MIN(C*FirstRow*:C*LastRow*),C*FirstRow*:C*LastRow*,0)), where*FirstRow*corresponds to the row number of the first row containing data, and*LastRow*contains the row number of the last row containing data. - Press enter, and double check that the listed temperature occurs at the minimum value of the first derivative.
- Then, drag the bottom left corner of the cell across all relevant columns to apply the formula to those columns of interest.
- Plot the columns relevant to your data set by making a scatter plot ("straight marked scatter"), having the temperature (values in column B) on the x-axis, and the first derivative values on the y-axis.
- Double check by eye that the values you calculated to be the melting temperatures correspond to the minimum values on the curves. (See example plot in the introduction section of this wiki page)
- Next, you may also check to see what the melting curves look like in terms of raw fluorescence by plotting fluorescence intensity vs. temperature in the "Melt Curve RFU Results" file. Again, validate the results you found by eye to see if the T
_{m}s correspond to the inflection point of the raw fluorescence melt curves. - Check to see if the T
_{m}of the control protein shifted when its ligand was added. Quantify the shift. - Check to see if the T
_{m}of FKBP12 shifted when Rapamycin or other compounds were added. Quantify the shifts. - By varying the concentration of Rapamycin, you will be able to determine an apparent dissociation constant of Rapamycin and FKBP12. Here is the reference for finding the apparent dissociation constant that was mentioned in lecture: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4692391/
- You may use this MATLAB script (ApparentKd.m) to help fit the Tm vs. Rapamycin concentration curve to a single binding site model and find a value for the apparent K
_{D}. The function uses a nonlinear regression.- Based on the given article linked above, the single binding site model is as follows, where the fit parameters include Tm_min (minimum Tm at no ligand concentration), Tm_max (maximum Tm at infinite ligand concentration), and K
_{D}is the apparent K_{D}value. For our experiment, the concentration of FKBP12 (written as [FKBP12]) was 8.5 uM and the concentration of Rapamycin (written as [Rap]) was variable.

- Based on the given article linked above, the single binding site model is as follows, where the fit parameters include Tm_min (minimum Tm at no ligand concentration), Tm_max (maximum Tm at infinite ligand concentration), and K
- Create an array of Rapamycin concentrations by typing
*RapConc = [A, B, C, D, E, etc.]*at the MATLAB command prompt, where A, B, C are the various Rapamycin concentrations in units of uM. - The 10 concentrations of Rapamycin are: 20uM, 10uM, 5uM, 1uM, 0.1uM, 0.05uM, 0.01uM, 0.005uM, 0.001uM, and 0.0001uM
- Create an array of T
_{m}s by typing*Tm = [A2, B2, C2, D2, E2, et c.]*at the MATLAB command prompt, where A2, B2, C2 are T_{m}s corresponding to concentrations A, B, and C in the previous array. - Making sure the MATLAB function ApparentKd.m is in your current working directory, type in
*ApparentKd(RapConc, Tm)*at the command prompt and press enter - The function performs a nonlinear regression of your data with a single binding site model, and will return an apparent K
_{D}value in units of uM from the best fit. - If this regression does not work well, you may use alternative methods to estimate an apparent K
_{D}value- You can find the EC50 value by fitting with the following formula in this matlab function EC50.m
- Making sure the MATLAB function EC50.m is in your current working directory, type in
*EC50(RapConc, Tm)*at the command prompt and press enter - This file performs nonlinear regression using the following equation, where the fit parameters include Tm_min (minimum Tm at no ligand concentration, Tm_max (maximum Tm at infinite ligand concentration, and EC
_{50}is the EC_{50}or apparent K_{D}value, and a Hill coefficient which should not be meaningful in this context.

- You may use this MATLAB script (ApparentKd.m) to help fit the Tm vs. Rapamycin concentration curve to a single binding site model and find a value for the apparent K

## Reagents

- DSF dye (Thermo Fisher)
- ligands (Chembridge)

Next day: Complete data analysis