Hello Maedeh,

In MTT assay, the higher is the signal, the more viable cell you have. So the viability curve is more intuitive in this case.

I would say, the first thing to do is to choose MTT incubation time where your assay is still in the linear range, i.e., the signal measured is proportional to the number of viable cells. You can easily test it by diluting the cell suspension and incubating your non-treated cells for as long as you would like to grow these during your following experiments. This needs to be optimized for different treatment times (e.g., if you want to incubate cells with toxins for 24 h or for 72 h) as well as cell lines.

For normalization, you will also need a 100% viability control (e.g., non-treated cells, or cells treated with DMSO or whatever you use for toxin dilutions – at the same volume % as in case of most concentrated toxin treatments) and a 0% viability control (usually, an empty well without cells). For normalization, you subtract the average signal obtained for 0% viability controls from the signal measured for all other data points, and divide by average difference of signals for 100% and 0% viability control. There is also an automated normalization feature available in Prism, but it’s good if you understand how this calculation is done.

Next, you make an XY graph with decimal logarithms of molar concentrations of toxins plotted on the X scale and the normalized signal plotted on the Y scale. For the 0% viability control, you can just assign a very large X value (because it models the situation for extremely high concentration of toxin), and for the 100% viability control, you can just assign a very small X value (because it models the situation for extremely low concentration of toxin). In the easiest case, you can fit the data to a dose-response curve (three parameters is the easiest one, then you do not have to worry about the slope value). Make sure to check that the upper plateau value that the algorithm finds is close to 100 and the lower plateau value is close to 0 (if not, you can fix those as constants, but it indicates your data is off).

Good luck!

Darja

I agree with Darja. In MTT assays, the measured optical density (OD) will be directly related to the amount of viable cells that can metabolize the MTT salt. So %viability makes more sense in that case. If it were an assay in which you are measuring cell death, than you could use %cytotoxicity.

To all the above mentioned answers, I humbly add, the Dose-Response setting on Prism can be used on either processed %cell survival, or raw OD readings, fitted to various non-linear curves. More information can be found here, (on Prism): https://www.graphpad.com/guides/prism/latest/curve-fitting/dose-response_curves_-_stimulation_top.htm