Comprehensive Analysis of Drug Response using the FLICK Assay

This research focuses on understanding how anti-cancer drugs function. The primary goal is to investigate how these drugs activate cell death. Methods for evaluating drug sensitivity quantify the number of cells, but not indicate whether the cells are dying, the extent of cell death, or which death mechanisms are activated. This protocol allows for precise calculations of drug-induced growth in death rates, which cannot be achieved by studying only population size, as is common with other assays. FLICK scores both live and dead cells with equal precision within the same assay, with no limitations to the types of cells or forms of cell death that can be studied.

[Instructor] To begin, plate the desired number of cells in each well of a 96 well optical bottom black walled plate. Add 100 microliters of complete medium and allow the cells to adhere to the plate overnight. Add 10 microliters of 1.5% Triton X solution to each well containing the plated cells without mixing. Observe cell morphologies under a light microscope using a 10X objective and inspect cells every hour until cell bodies are no longer visible. For each concentration to be tested, plate 40,000 cells in 180 microliters of cell culture medium in triplicate along the leftmost column of a 96 well optical bottom black walled plate. Add 90 microliters of medium to the remaining wells. Using a multi-channel pipette, transfer 90 microliters from the leftmost column into the adjacent column to the right to create a one to two serial dilution. Pipette up and down 15 times to mix. For the second to last column, remove 90 microliters so that all wells contain 90 microliters of medium with a varied number of cells. Add 10 microliters of 10X DNA dye to each well of the 96 well plate. Then add 10 microliters of 1.5% Triton X solution to permeabilize the cells. To measure the fluorescence intensities, load the plate into a fluorescence plate reader. Adjust excitation and emission wavelengths and digital gain depending on the plate reader used. And measure fluorescence intensities across the cell titration plate. Quantify fluorescence over a range of acquisition settings. Subtract the average signal of the column containing no cells to remove the background signal from each measurement. Plot the cell number against the fluorescent signal and perform linear regression to determine the linearity of each DNA dye concentration for each acquisition setting. Choose the DNA dye concentration and acquisition settings that offer the best combination of linearity and dynamic range. After counting the cells, mix the counted cell suspension with the appropriate media volume using a serological pipette. Transfer this suspension to a V-bottom reagent reservoir. Using a multichannel pipette, add 90 microliters of the cell suspension to each well of the 96 well plates. And mix the cell suspension regularly by repeated pipetting to maintain the desired cell concentration. Using a 10X concentration of the selected DNA dye in complete growth media, prepare a 10 X concentration of each drug to be tested. And serially dilute using a multi-channel pipette. Mix 15 times between each well. Now, add 10 microliters of the drug and DNA dye mixture to the plates containing cells. Acquire fluorescence readings for all drug treated plates every three to four hours after drug addition. At the final desired time point, acquire a fluorescence reading. Immediately after that, add 10 microliters of 1.5% Triton X solution to lise the cells. Then acquire fluorescence readings following cell permeabilization. Calculate the average fluorescence values from the T zero control plate using the 50% trimmed mean. Using curve fitting and an exponential growth function, calculate the population growth kinetics for all wells. Based on the growth parameters, calculate the number of total cells at every measured time point in the assay for every well. Now, subtract the dead cell measurement from the total cell counts to determine the number of live cells at each measured time point. Calculate the lethal fraction, or LF, at each time point by dividing the dead cell fluorescence signal by the total cell signal for each time point. Fit a lag exponential death, or LED, equation to the lethal fraction time course data. For drug doses not causing significantly lethality, use a linear model with zero slope. Determine significant levels of lethality based on the noise in the assay, or the LF observed for un drugged conditions. From the LED equation, extract four parameters using non-linear regression. The initial lethal fraction, or LFI, lethal fraction plateau, or LFP, maximum death rate, or DR. And the death onset time, or DO. Compute the fractional viability, or FV, at the assay endpoint for each drug at each dose. Subtract the endpoint lethal fraction from one, or divide the number of live cells by the total cells. Determine the average number of live cells for each well at the start of the assay by calculating the difference between the post permeabilization T zero reading and the T zero reading for each well. This value is referred to as X zero. Now, for each drug treated well, compute the normalized growth rate inhibition value, or GR, using the given equation. Perform curve fitting using a four parameter logistic regression. Next, determine the relationship between growth and death rates and the population size using a simulation based on a birth/death model. Use the average number of live cells from the T zero control plate, the length of the assay. And user-defined ranges for plausible proliferation and death rates as inputs for this simulation. To determine plausible growth rates, use the untreated growth rate in cell population doublings per hour as the highest possible rate and zero as the lowest. Divide this range into 500 equally spaced segments. Determine the GR values for each simulated proliferation and death rate pair. Then determine the LF values. Compute the pairwise distance between each calculated GR and LF pair and each theoretical GR and LF pair in the lookup table. Identify the theoretical pair with the minimum distance to the experimentally observed GR and LF pair as the true drug-induced proliferation and death rates. This figure illustrates how the FLICK assay and GRADE analysis quantify and distinguish between drug-induced growth inhibition and cell death in U2 OS cells. Belinostat at one micromolar showed a GR value of zero, indicating cyto stasis, but also caused approximately 50% cell death. GRADE analysis showed that one micromolar belinostat caused both growth inhibition and cell death with distinct contributions to overall population size. The apoptotic inhibitor Z-VAD reduced belinostat induced lethality by 18%, indicating apoptosis involvement. Conventional analysis showed all drugs reduced viability, but GR analysis revealed only belinostat and Camptothecin caused cell death at high doses. GRADE analysis showed Belinostat induced both growth inhibition and death at all doses. Camptothecin was biphasic, causing only growth arrest at low doses and death at high doses. And palbociclib caused growth arrest only.