See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/51073437

Cell sensitivity assays: The MTT assay

Article in Methods in molecular biology (Clifton, N.J.) · March 2011

DOI: 10.1007/978-1-61779-080-5_20 · Source: PubMed

CITATIONS READS

886 30,496

3 authors, including:

Johan Van Meerloo Jacqueline Cloos

Amsterdam University Medical Center University Medical Centers Amsterdam, Cancer Center Amsterdam
31 PUBLICATIONS 1,544 CITATIONS 347 PUBLICATIONS 7,375 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

IN2016 View project

RNA profiling in liquid biopsies to predict MTX efficacy in Rheumatoid Arthritis and Acute Lymphoblastic Leukemia. View project

All content following this page was uploaded by Johan Van Meerloo on 07 July 2015.

The user has requested enhancement of the downloaded file.

 Chapter 20

Cell Sensitivity Assays: The MTT Assay

Johan van Meerloo, Gertjan J.L. Kaspers, and Jacqueline Cloos

Abstract
The MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assay is based on the
conversion of MTT into formazan crystals by living cells, which determines mitochondrial activity. Since
for most cell populations the total mitochondrial activity is related to the number of viable cells, this assay
is broadly used to measure the in vitro cytotoxic effects of drugs on cell lines or primary patient cells.
In this chapter the protocol of the assay is described including important considerations relevant for each
step of the assay as well as its limitations and possible applications.

Key words: MTT, 3-[4,5-Dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide, Viability assay,

IC50, LC50, Drug sensitivity assay, Cytotoxicity assay

1. Introduction

The general purpose of the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5

diphenyl tetrazolium bromide) assay is to measure viable cells in
relatively high throughput (96-well plates) without the need for
elaborate cell counting. Therefore the most common use is to
determine cytotoxicity of several drugs at different concentra-
tions. The principle of the MTT assay is that for most viable cells
mitochondrial activity is constant and thereby an increase or
decrease in the number of viable cells is linearly related to mito-
chondrial activity. The mitochondrial activity of the cells is
reflected by the conversion of the tetrazolium salt MTT into
formazan crystals, which can be solubilised for homogenous
measurement. Thus, any increase or decrease in viable cell
number can be detected by measuring formazan concentration
reflected in optical density (OD) using a plate reader at 540 and
720 nm. For drug sensitivity measurements the OD values of

Ian A. Cree (ed.), Cancer Cell Culture: Methods and Protocols, Second Edition, Methods in Molecular Biology, vol. 731,
DOI 10.1007/978-1-61779-080-5_20, © Springer Science+Business Media, LLC 2011

237
238 van Meerloo, Kaspers, and Cloos

wells with cells incubated with drugs are compared to the OD

of wells with cells not exposed to drugs.
The MTT assay is suitable for the measurement of drug
sensitivity in established cell lines as well as primary cells. For
dividing cells (usually cell lines) the decrease in cell number
reflects cell growth inhibition and the drug sensitivity is then
usually specified as the concentration of the drug that is required
to achieve 50% growth inhibition as compared to the growth
of the untreated control (50% inhibitory concentration, IC50).
For primary (nondividing) cells, drug sensitivity is measured as
enhanced cell kill of treated cells as compared to the loss of
cells already commonly seen in untreated cells (50% lethal con-
centration, LC50). For some cell types such as fresh acute myeloid
leukemia (AML) cells the control median cell survival is 100%,
which is no problem as long as results for treated cells are
compared to the controls.
For primary pediatric acute lymphoblastic leukemia (ALL)
cells, the in vitro MTT assay has been extensively applied to
predict drug sensitivity in vivo. In these studies, a relatively good
correlation was observed between in vitro sensitivity and clinical
outcome (1–4). Despite these and other positive results, the MTT
assay is not commonly used as predictive test for the clinic, but
rather as an in vitro tool to determine potential antitumor activity
of new drugs and/or combinations of drugs (5). In addition,
the MTT assay is a suitable tool to study resistance to drugs (6, 7)
(see Table 1 for applications).
In order to set up the MTT assay for cells and/or drugs that
have not been tested before, several considerations are of crucial
importance and will be discussed in this chapter.

Table 1
Possible applications for MTT assay

Application References
Study of cross-resistance between related and (2, 3, 5–7)
unrelated drugs
Measuring drug sensitivity ex vivo to predict clinical (1–4, 7, 8)
outcome
Sensitivity testing of new drugs (4–6)
Drug screening on cell lines and/or patient samples (3–5, 7, 8)
Testing drug combinations on cell lines and/or patient (1, 2)
material
 Cell Sensitivity Assays: The MTT Assay 239

2. Materials

2.1. C
 ells and Controls Class 2B biocabinet suitable for drug experiments. Incubator
with 5% CO2 at 37°C. Microplates – 96 well (Greiner Bio-one,
Alphen a/d Rijn, The Netherlands). For suspension cells, we use
round bottom wells or flat bottom wells, while for attached
cells only flat bottoms can be used (see Subheading 3.1 for the
different loading volume per plate) (see Note 1).

2.2. Solutions Dissolve 500 mg MTT powder (Sigma, St. Louis, USA) in
and Solvents 10 mL PBS. Stir with a magnetic stirrer for approximately 1 h
in the dark. Filter sterilize the solution with a 0.22 mm filter
2.2.1. MTT Solution
(Millipore, Carrigtwohill, Ireland) and store in 10-mL aliquots
at −20°C.
Warning: MTT is toxic and harmful. MTT is light sensitive,
hence protect from light.

2.2.2. Acidified Isopropanol To dissolve the formazan crystals, different solutions can be used
such as methanol, ethanol, and DMSO. Our lab has the best
experience with acidified isopropanol. To make this solution, add
50 mL 2 M HCl to 2.5 L isopropanol. Store the solution at least
a month at room temperature before use. When the isopropanol
is not acidified correctly the suspension will become cloudy.

2.3. E quipment 1. Plateshaker (IKA schuttler MST4, Janke & Kunkerel, Staufen,
Germany).
2. Pipettes 0.001–1 mL, single channel and 0.01–0.3, multi­
channel.
3. Class 2B hood.
4. Benchtop centrifuge.
5. Microplate reader (Anthos-Elisa-reader 2001, Labtec,
Heerhugowaard, The Netherlands).
6. O2 incubator.

3. Methods

3.1. P
 late Setup Generally cells are plated in triplicates to minimize the variability
of the results. The volume of cells depends on the type of plate
used. For round bottom and flat bottom plates, 80 mL and 120 mL
are used, respectively. Each plate should contain control wells
(without drugs) and blank wells (without cells). For some drugs
that also show absorbance at the given wavelenghts, an additional
control is required of wells with medium (without cells) including
240 van Meerloo, Kaspers, and Cloos

the range of drug used. The number of plates needed depends on

the specific experiment. A common MTT assay experiment
requires a testing plate for the OD of the drug, a testing plate to
determine the growth curve for the starting amount of cells
seeded per well, and a broad dilution range to determine the dilu-
tion range for the experiments. However, for a standardized drug
sensitivity assay of patient material, a single plate often suffices.
For cell lines, it is recommended to include a day 0 plate in order
to accurately determine the extent of growth (in the course of the
experiment for the control cells). The outer wells are not used
for the experiment due to evaporation and are filled with phos-
phate buffered saline (PBS) to keep the evaporation of the plate
to the minimum, some drugs can also have influence on neigh-
bouring wells; this has to be investigated before starting the
experiment (see Note 2).

3.2. Drug Incubation There are two methods of drug exposure (1); if the drug is
unstable or cannot be stored in medium at −20°C, the plates have
to be prepared fresh by making the drug dilution stocks and
adding the total amount of drugs in 20 mL for round bottom
plates and 30 mL of drug solution to flat bottom wells to each
well just before adding the cells. This renders the total volume of
cell and drug suspension to 100 mL for round bottom and 150 mL
for flat bottom 96-well plates. (2) For testing a stable drug, plates
can be prepared with the 30 mL drug concentrations and can be
stored in −20°C for later use. When the cells are added to the
wells containing the drug suspension, this will mix the drugs
through the cells (see Note 2).

3.3. Culture Period For primary leukemic cells, the plates are incubated for 4 days to
determine the optimal effect for most standard drugs. For cell lines,
the plates have to be incubated during log phase, which is usually
72–96 h in an incubator with 5% CO2 at 37°C (see Note 1).

3.4. MTT Incubation After the appropriate incubation time, add 1:10 volume of MTT
solution (5 mg/mL); e.g. 10 mL for round bottom and 15 mL for
flat bottom 96-well plates. Unused MTT can be frozen and
reused. Shake plates for 5 min on a plateshaker by slowly increas-
ing the shaking speed to a maximum of 900 shakes/min. Then
incubate the plate for another 4–6 h at 37°C in a CO2 incubator,
depending on the cell type.

3.5. Dissolution 150 mL of acidified isopropanol is added to each well and

of Formazan Crystals resuspended until all crystals have been dissolved. Mix each well
thoroughly using a multichannel pipette. Between each row,
rinse tips of multichannel pipette with isopropanol and discard
the used isopropanol. Blow out tips thoroughly before mixing
the next rows. Start with the control wells, before mixing the
rows with drugs.
 Cell Sensitivity Assays: The MTT Assay 241

3.6. O
 D Measurement The OD is measured at 540 and 720 nm in order to get a more
exact measurement by correcting for background noise. The
720 nm OD background will be subtracted from the 540 nm OD
total signal. The measured data are copied into an excel sheet and
with the use of the following formula, the percentage of living
cells can be determined: The average OD of the blank control
wells (without cells and if the drug has no specific OD without
drug as well) is subtracted from the average OD of the control
wells (cells but no drugs) and the wells containing the drugs.
Some drugs may interfere with the OD measurement and then the
average OD of the wells containing drugs, but without cells is
subtracted from the average OD of the control wells. The leukemic
cell survival is calculated by: (OD treated well [−blank])/(mean OD
control well [−blank]) × 100. The LC50 (the drug concentration
which results in 50% leukemic cell survival) can be calculated. For
more reliable results the experiment should be done in triplicate
and in duplicate in case of primary cells and when a range of drug
concentrations is being tested.

3.7. E xample Results After adding acidified isopropanol and dissolving the formazan
crystals thoroughly the plate has to rest for 10 min before
measuring. In Fig. 1, a plate is shown after dissolving the
formazan crystals with acidified isopropanol showing the blank
culture medium wells, untreated cell control wells, and three
drugs for which the cells are either sensitive (C), intermediate
sensitive (E), and resistant (D) in these dose ranges. To be able

Fig. 1. A 96-well plate after formazan crystals are dissolved in acidified isopropanol.
(a) Blanks control wells, (b) untreated cell control wells, (c) cell line with drug C with a
dose–response curve from 100% cell death to no response on cell growth, (d) cell line
with drug D with a dose–response curve showing no growth inhibition, (e) cell line with
drug E with a dose–response curve with dose-dependent modest growth inhibition at
high drug concentrations. Outer wells are not used because of possible evaporation.
242 van Meerloo, Kaspers, and Cloos

100

90

80

% of growth relative to control

70
Drug C
60
Drug E
50
Drug D
40

30

20

10

0
0.0012 0.0025 0.005 0.01 0.02 0.04
−10
Drug concentrations

Fig. 2. Dose–response curve of cell lines C, D, and E. (C) cell line with drug C with a
dose–response curve from control growth without drugs (set at 100%) to complete cell
kill at the highest drug concentrations, (D) cell line with drug D with a dose–response
curve showing no growth inhibition, (E) cell line with drug E with a dose–response curve
showing dose-dependent modest growth inhibition (IC50 is not reached).

to determine IC50 values of drugs D and E, different dose ranges

are required.
When measured in a plate reader, these measurements can
then be used to determine the IC50 for cell lines and LC50 for
patient material. The results can be plotted into a graph as shown
in Fig. 2. The wells containing untreated cells are set as 100% and
the dose–response curves of the three different drugs are depicted.
It is clearly shown in the picture that for drugs D and E, the IC50
is not reached.

4. Notes

1. The cell concentration that is plated may vary for different

cell lines and primary cells. For leukemic cell lines, we use
concentrations in the range of 3.0–4.0 × 103/well, while primary
AML cells are plated in 0.08–0.12 × 106 cells/well and
primary ALL cells in 0.16 × 106 cells/well. For cells that have
not been used before in this assay, several cell concentration
experiments have to be done before starting the drug experi-
ments as recommended below.
2. Cells are plated in duplicate or triplicate, and in each plate,
4–6 control wells are included that contain the same amount
of cells as the experimental wells and which will not be
exposed to drugs. Additional 4–6 wells are needed for
 Cell Sensitivity Assays: The MTT Assay 243

measuring the blanks, which contain the culture medium

only. Some drugs can have an OD value of their own that can
influence the measurement when diluted, and therefore a
dilution of the drugs has to be measured without the adding
cells. If there is a high difference in OD between the different
drug dilutions, extra drug dilution control wells have to be
added to the plate setup. The absorbance value for the blanks
should be between 0.001 and 0.1 OD units as measured on a
Microplate reader using filters for 540 and 720 nm. In addi-
tion, the absorbance range for untreated cells should typically
be between 0.75 and 1.25 OD units (see Notes 3 and 4).
3. To determine whether cells are suitable for use in the MTT
assay, several parameters need to be verified:
(a) Linearity of viable cell number with OD level. To deter-
mine this, several cell concentrations are plated and
related to OD values. For cell types where the mitochon-
drial activity of cells is not constant, the MTT assay can-
not be used in to measure the influence of the drugs.
(b) Cell concentrations. The optimal cell concentration to be
plated is dependent on the basic level of mitochondrial
activity and the rate of proliferation. In order to establish
this, several concentrations of cells should be plated in,
for instance, seven plates, and measured daily to deter-
mine the growth curve of the cell line to prevent over-
growth, which will influence the experiment. The starting
OD value of day 0 should not exceed 0.125. By con-
structing a growth curve, the logarithmic cell phase is
determined at which the cells duplicate. At a certain time
point the cell growth will plateau due to exhaustion of
the medium, contact inhibition, and exceeding of the
maximal OD value that can accurately be measured. The
most optimal concentration of plating is when cells have
almost no lag phase and the assay should not proceed
after the log phase.
(c) Cell culture. Cells should be cultured for a relevant time
period to be able to demonstrate the effect of the drug.
For cell lines this should only be during the log phase,
while for primary cells the MTT assay has to be completed
before all untreated cells are dead. Only for cell lines, a day
0 plate is used to precisely measure the activity of the
starting cell dilution at day 0 without the drugs’ effects.
Moreover, when using primary leukemic cells, in this
assay, the blast count at day 4 should be ³ 70% leukemic
cells; this will be a blast count of ³ 80–90% for ALL (7, 8)
and ³ 70–75% for AML at day 0. Samples can be purified
with immunomagnetic beads directed against contaminat-
ing nonmalignant cells (8).
244 van Meerloo, Kaspers, and Cloos

4. Drug incubations.
(a) Control medium. In order to account for possible influences
of the dissolvent of the drugs on the background OD, the
control medium should contain the concentrations of dis-
solvent of drugs.
(b) Stability of the drugs. To determine if cytostatic drugs
have to be added fresh or if the drugs can be stored in
plates at −20°C, drugs should be added into the 96-well
plates and stored at −20°C in advance of the experiments.
Possible differences in the dose–response curves will
imply whether the plates with drugs can be stored or that
the drugs need to be added fresh at the day of plating.
(c) Drug combinations of drug A + drug B can be measured
in several ways of which two are the most common (1)
One fixed concentration of drug A can be used (usually
in a concentration between the IC20 and IC50) in combi-
nation with a dose range of drug B. (2) Mix both drugs
at IC50 as highest dose and then make a dose–response
curve by diluting this mixture further to obtain a
combined dose–response curve. Synergism or antago-
nism can then be investigated, for instance, by the use
of “calcusyn” software (Biosoft, Cambridge, UK) which
can deal with both methods.
(d) Reproducibility is important for an accurate measure-
ment and so the use of at least a duplicate row setup of
each drug, and if possible a triplicate row setup is pre-
ferred within each plate. For cell lines we always perform
the assay in triplicate while for primary material this is
commonly not possible.
(e) Plate set-up. It is our experience that some drugs can
affect surrounding cells possibly by aerosols; therefore,
the wells with drugs have to be separated by PBS to make
sure that the drugs will not affect the adjacent wells. The
plate set-up also needs to include 4–6 control wells with
the used culture medium, which are used to measure the
background of culture medium.

Acknowledgments

This work was supported by KiKa “Stichting Kinderen

Kankervrij” – Dutch Children Cancer-free foundation and
VONK “VUmc Onderzoek Naar Kinderkanker” foundation.
 Cell Sensitivity Assays: The MTT Assay 245

References
1. Kaspers, G.J., Pieters, R., Van Zantwijk, CH., Gibson, B.E., Creutzig, U., Janka-Schaub,
Van Wering, E.R., Van Der Does-Van Den G.E., den Boer, M.L., Pieters, R., Kaspers,
Berg, A., Veerman, A.J. (1998) Prednisolone G.J. (2006) In vitro sensitivity and cross-resis-
resistance in childhood acute lymphoblastic tance to deoxynucleoside analogs in childhood
leukemia: vitro-vivo correlations and cross- acute leukemia. Haematologica. 91, 17–23.
resistance to other drugs. Blood. 92, 259–266. 6. Oerlemans, R., Franke, N.E., Assaraf, Y.G.,
2. Salomons, G.S., Smets, L.A., Verwijs-Janssen, Cloos, J., van Zantwijk, I., Berkers, C.R.,
M., Hart, A.A., Haarman, E.G., Kaspers, G.J., Scheffer, G.L., Debipersad, K., Vojtekova, K.,
Wering, E.V., Der Does-Van Den Berg, A.V., Lemos, C., van der Heijden, J.W., Ylstra, B.,
Kamps, W.A. (1999) Bcl-2 family members in Peters, G.J., Kaspers, G.J., Dijkmans, B.A.,
childhood acute lymphoblastic leukemia: rela- Scheper, R.J., Jansen, G. (2008) Molecular
tionships with features at presentation, in vitro basis of bortezomib resistance: proteasome
and in vivo drug response and long-term clini- subunit beta5 (PSMB5) gene mutation and
cal outcome. Leukemia. 13, 1574–1580. overexpression of PSMB5 protein. Blood. 112,
3. Kaspers G.J., Veerman A.J., Pieters R, Van 2489–2499.
Zantwijk CH., Smets L.A., Van Wering E.R., 7. Klumper E., Pieters R., Veerman A.J.,
Van Der Does-Van Den Berg A. (1997) In Huismans D.R., Loonen A.H., Hählen K.,
vitro cellular drug resistance and prognosis in Kaspers G.J., van Wering E.R., Hartmann R.,
newly diagnosed childhood acute lymphoblas- Henze G. (1995) In vitro cellular drug resis-
tic leukemia. Blood. 90, 2723–2729. tance in children with relapsed/refractory
4. Kaspers, G.J., Wijnands, J.J., Hartmann, R., acute lymphoblastic leukemia. Blood. 86,
Huismans, L., Loonen, A.H., Stackelberg, A., 3861–3868.
Henze, G., Pieters, R., Hählen, K., Van 8. Kaspers G.J., Veerman A.J., Pieters R,
Wering, E.R., Veerman, A.J. (2005) Immuno- Broekema G.J., Huismans D.R., Kazemier
phenotypic cell lineage and in vitro cellular K.M., Loonen A.H., Rottier M.A., van
drug resistance in childhood relapsed acute Zantwijk CH., Hählen K. (1994) Mononuclear
lymphoblastic leukaemia. Eur J Cancer. 41, cells contaminating acute lymphoblastic leu-
1300–1303. kaemic samples tested for cellular drug resis-
5. Hubeek, I., Peters, G.J., Broekhuizen, R., tance using the methyl-thiazol-tetrazolium
Zwaan, C.M., Kaaijk, P., van Wering, E.S., assay. Br J Cancer. 70, 1047–1052.

View publication stats