MTT problem, please help - (Oct/08/2010 )

Hi all

I have been working with Caco-2 cells for a while and use them to do some MTT assay. I have some questions. Could anyone please help me?

1. How many cells are enough to do MTT assay? I tried 10,000 cells/well in a 96-well plate and incubated for 24hrs, but I didn't get much formazen.
2. Is there any better way to remove media and MTT solution without touching any formazen? I tried centrifuge before I took out the media, but I still couldn't remove the media without touching any formazen by using a pipette.

Can you please post your protocol? This will make it easier for others to help you.

Here is my protocol.

1. Incubate Caco-2 cells in a T75 flask until reach 80-100% confluence.
2. Dilute the cells to 100,000 cells/ml and add 100 ul/well to a 96-well plate, then incubate 24 hrs at 5% CO2, 37C
3. After 24 hrs, remove all the medium and add designed compounds, incubate for 4 hrs
4. Remove all the compounds and add MTT solution (5mg/ml), 20 ul/well, then 80 ul of Medium (DMEM with 10 % FBS, 1% NEA and !% penicillin)
5. Wait for another 4 hrs of incubation
6 After 4hrs, centrifuge the plate and remove the supernatant, then add 50 ul DMSO to dissolve the formazan
7. Measure the plate at 570 nm

I use DMEM, SDS and PBS as blank, negative and positive controls.

I have tried SDS/DMF solution to dissolve formazan and this can save the step of no.6, but I had problem to remove the colour from MTT solution mixed with DMEM, which affected my result significantly.

The best option is to simply add more cells. I would try 200uL/well of you 100,000 cells/ml or 20,000 cells per well. There are several different steps you could alter including the length of treatment, length of incubation, time to dissolve formazan crystals. I also recommend that you incubate the plates at 37 degrees - in the CO2 incubator is fine. This simply speeds the dissolution of the formazan crystals in the DMSO. What are you trying to measure with the MTT? Do you expect the effect to occur within the 4 hrs of treatment? These are questions you should ask yourself and that I'm curious about, but there are cases where you treatment may be sufficient to see the effect your desire it just depends. If you are measuring proliferation specifically, I would say treat the cells for at least 24hrs. If you try the larger number of cells and it doesn't seem like the formazan crystals are dissolving, then leave the plate out over night and measure at 570nm the following morning. I hope this helps.

Just a note to say that 100,000 cells/well in a 96 well plate is massively over-confluent so not realistic in terms of normal cell growth. It also won't work for kill curves as most drugs work on cells that are less than 70% confluent.

Thank so much for everyone

I am doing drugs screening. My compounds are peptide-based. I think 4hr treatment is sufficient to examine their toxicity. My senior strongly suggests me to incubate the cells at 50,000 cells/well for 3-5 days. He concerns that 24 hrs incubation of cell culture are not enough for cell adhesion to the plate surface. However, I wonder high cell number might affect sensitivity if the cells are over confluent . What I am trying to do is to modify this method to be more accurate and reduce error bar in the data, including limit lost of formazan, figuring out which cell number is sufficient for this method (it depends on different cell lines). Next time, I would like to try phenol red free medium to see if this can limit the interference of colour from medium.