I am testing two compounds against 3 cell lines to determine cell viability. Two lines grow in Eagle media, one in DMEM (+10 Percent bovine serum) which contain phenyl red indicator. I prepared 3 plates to let the compounds take effect for up to 96h. For each time point I treat the plate with MTT for 4 h, then remove the media, than add DMSO and incubate for 15 min. On the first plate it appeared to be a positive correlation between increasing concentration and decreasing cell viability. However on the second plate the trend is not as obvious. I would also assume that cell viability is reduced more than at the first time point which is not the case in a clear way or I can't see it from the data until mow. You find an image of the first plate here

enter image description here

On the plate columns 1-4 are cell line 1 (with increasing compound concentration), 5-8 cell line 2 and 9-12 is cell line 3. Rows A-C are triplicates of compound 1, rows D-F are triplicates of compound 2, and G+H and is just media as negative control.

For me the plate does not look good with bare eye: The first row is off completely and from column 1-4 the intensity should either decrease or at least stay steady.

I am assuming the problem could be related to multiple factors:

  1. During media removal always some crystals will be sucked in by the pipette. I tried to avoid this by centrifuging the crystals to create a more stiff layer but this did not work well. Also tilting the plates a little bit did not work nicely.

  2. When you're too careful not to suck any crystals in on the other hand media will remain in the well which will change concentration and also show some absorbance due to the indicator.

I am wondering how your experiences are with with these issues with the MTT assay? I personally find it to be a huge design flaw, that you need to remove the media, witch will lead to possible problems with the quantification.

The alternative works with 10 Percent SDS in 0.01 HCl overnight to dissolve the formazan. I am also wondering if you can share experiences on this modification as well. I am likely to do this for the 3rd plate to see if the overall structure of the data is more consistent.

Alternatively, should I switch to XTT, WSF-1, MSF etc.?


1 Answer 1


The easy answer is to use acidified SDS 10 Percent in 0,01 N HCl. I added this directly, to my media and did not have any problems with crystal removal anymore. The acid will remove the coloring of the indicator. The triplet accuracy increased enormously in comparison to the DMSO approach. I would not recommend the DMSO technique at all, because you might ruin all your work. At least if media removal is involved. But you need to incubate overnight as stated in various protocols if using SDS.Also additional mixing with a pipette or a plate shaker might help if you have a gradient of color in your wells. On the other hand maybe it would be possible to add acidified DMSO, or DMF or an alkohol in acidified Form without the need to remove media.

See this publication for an overview of MTT, MTS, XTT, WSF-1, WSF-8 and different advantages and cons of each variant: https://onlinelibrary.wiley.com/doi/10.1002/fft2.44


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .