Histochemistry and Cell Biology

https://doi.org/10.1007/s00418-019-01775-7

ORIGINAL PAPER

High-throughput quantification of the effect of DMSO

on the viability of lung and breast cancer cells using an easy-to-use
spectrophotometric trypan blue-based assay
Sarah Musa Hammoudeh1 · Arabella Musa Hammoudeh1 · Rifat Hamoudi1,2

Accepted: 7 February 2019

© The Author(s) 2019

Abstract
One of the main aspects investigated in potential therapeutic compounds is their effect on cells viability and proliferative abil-
ity. Although various methods have been developed to investigate these aspects, these methods present with shortcomings in
terms of either cost, availability, accuracy, precision, or throughput. This study describes a simple, economic, reproducible,
and high-throughput assay to quantify cell death and proliferation. In this assay, adherent cells are fixed, stained with trypan
blue, and measured for trypan blue internalization using a spectrophotometric absorbance plate reader. Corresponding cell
counts to the absorbance measurements are extrapolated from a standard curve. This assay was used to measure the effect
of dimethyl sulfoxide (DMSO) on the viability of breast and lung cancer cells. Decrease in cell count associated with the
increase in DMSO percentage and exposure time. The assay’s results closely correlated with the conventional trypan blue
exclusion assay (Pearson correlation coefficient (r) > 0.99; p < 0.0001), but with higher precision. The assay developed in
this study can be used for various applications such as optimization, cell treatment investigations, proliferation, and cyto-
toxicity studies.

Keywords  Cell count · Trypan blue · Spectrophotometric assay · DMSO · High-throughput screening

Introduction Originally, the use of haemocytometers and counting

chambers was the standardized cell-counting approach.
Cell counting is a common, important practice in cell culture However, due to their various limitations (i.e., low accuracy,
used on daily basis for different applications. The increased low precision, low throughput, high probability of varia-
incorporation of cell count in various techniques and assays tions, and susceptibility to various human error sources), the
imposed growing needs for higher levels of cell-count accu- use of these methods became limited to simple applications
racy, precision, and throughput than those offered by the (Ongena et al. 2010; Tucker et al. 1994). An improved, auto-
originally developed and standardized methods. Hence, vari- mated modification of these methods was developed aiming
ous alternative approaches were developed aiming at fulfill- at increasing the cell-count throughput. However, it failed
ing these requirements and overcoming original methods’ to overcome the other shortcomings due to the similarities
shortcomings. These approaches were based on various in sample preparation and mounting onto the special count-
concepts including automation, flow cytometry, impedance, ing slides/chambers (Camacho-Fernandez et al. 2018). Fur-
tetrazolium salts, methylene blue, microscopy, fluorescence, thermore, despite the various adjustment options, automated
and protein quantification. counters remain prone to fail in realizing varying cellular
shapes and sizes pooled in the same sample as well as out-
of-focus cells in different focal planes (Camacho-Fernandez
* Rifat Hamoudi et al. 2018).
[email protected] Other approaches aiming at increasing the accuracy and
1
Sharjah Institute for Medical Research, College of Medicine, precision of counting were based on passing and counting
University of Sharjah, 27272 Sharjah, UAE each cell in the sample discretely (e.g., flow cytometry and
2
Department of Clinical Sciences, College of Medicine, impedance-based Coulter cell counting) (Collins et al. 2010;
University of Sharjah, Sharjah, UAE Coulter 1953; DeBlois and; Bean 1970). However, the use

13
Vol.:(0123456789)
 Histochemistry and Cell Biology

of these approaches requires special devices and reagents Table 1  Cell concentrations Con- Standard curve serial
that are relatively costly and often unavailable for use. Fur- seeded to constitute the standard cen- concentrations (cells/
curve tra- ml)
thermore, some of these methods focus on measuring cell
tion
number and size, but lack the ability to distinguish viable
num-
cells and cell aggregates. ber
Less demanding approaches with high accuracy and pre-
cision include the use of colorimetric assays (e.g., tetrazo- 1 350,000
lium salt-based approaches) (Mosmann 1983; Roehm et al. 2 262,500
1991). However, these methods require the use of reagents 3 196,875
and kits that are costly. Furthermore, as those approaches are 4 147,656
based on the measurement of the colored formazan products, 5 110,742
they are prone to artifacts affecting the NAD(P)H-dependent 6 83,056
cellular oxidoreductase enzymes’ activity and formazan pro- 7 62,292
duction (Maioli et al. 2009). These assays could as well be 8 46,719
cell-type dependent with varying efficiencies. Colorimetric Following the first concentra-
methylene blue-based counting assays were developed as tion, a ratio of 3:4 is used to
simple cell-counting assays with relatively high through- construct the subsequent serial
dilutions
put and relatively low cost (Felice et al. 2009; Oliver et al.
1989). However, the requirement of the elution step in these
Table 2  Chosen arbitrary values seeded to assess the performance of
processes results in variations in readings and requires inten-
the assay
sive optimization efforts (Felice et al. 2009).
In accordance with these shortcomings, we aimed to Concentration number Chosen arbitrary values (seeded
count (cells/ml) ± DMSO treat-
develop an easy method that would only require daily used ment)
reagents in cell culture while still providing a close effi-
ciency to that of the other developed approaches. Applica- 1 200,000 – DMSO
tions of such an approach can vary to include proliferation 2 200,000 + DMSO
studies, cytotoxicity studies, seeding count optimizations, 3 150,000 – DMSO
and treatment optimizations. The capacity of this method as 4 150,000 + DMSO
a cell counting and toxicity assay will be assessed through 5 100,000 − DMSO
the application of DMSO to lung and breast cancer cells. 6 100,000 + DMSO
DMSO is a common reagent used in various cell-culture 7 50,000 − DMSO
applications including the preparation of treatments and 8 50,000 + DMSO
cryopreservation. However, increased exposure to DMSO Each concentration is seeded with (+) or without (−) DMSO treat-
can lead to increased cell damage resulting in pseudo-effects ment to investigate the potential use of the assay for cytotoxicity stud-
in treatments as well as cell loss in cryopreservation. ies

Materials and methods 5% dimethyl sulfoxide (DMSO; Sigma, USA), similarly in

triplicates. The 96-well plate was left for 10–15 min in the
Cell culture and standard curve preparation biological cabinet before transferring to the incubator to
for the assay avoid aggregation of cells in the middle of the well. A549
cells were incubated for 6 h, whereas MDA-MB-231 cells
A549 and MDA-MB-231 cell lines (ATCC, USA) were cul- were incubated for 20 h. Cells were monitored throughout
tured in RPMI-1640 media supplemented with 10% FBS for adherence and normal morphology adaptation and the
and 1% penicillin–streptomycin (Sigma, USA) and main- experiment was stopped (fixation/trypsinization) once these
tained at 37 °C and 5% CO2. Cell suspension concentrations conditions were fulfilled.
were prepared in 8 serial dilutions at a ratio of 3:4 starting
with 350,000 cells/ml (Table 1). Cells were then seeded in Traditional trypan blue cell counting
2 sets of triplicates (one set for traditional counting and the
other for the trypan blue colorimetric assay) in a 96 well Cells were harvested with 0.05% Trypsin–EDTA (Sigma,
plate, 100 µl per well. Additional arbitrary concentrations USA) and enzymatic activity was neutralized with FBS-
(Table 2) were chosen to assess the efficiency of the assay containing RPMI-1640 Media. Cell densities were measured
in estimating cell count and were seeded with or without using a haemocytometer after the addition of trypan blue.

13
Histochemistry and Cell Biology

Densities were then adjusted for the final count per well by and very low confluency) were plated in a 96-well plate.
multiplying by the suspension volume in each well. Accordingly, the rough borderlines of the standard curve
range were determined, ~ 35,000 cells/well at full confluency
Trypan blue staining and ~ 5000 cells/well at low confluency, for both cell lines.
This range was then used to identify the dynamic range of
The cells were fixed with 4% Paraformaldehyde (PFA) the assay by testing its accuracy at the two previously iden-
(Sigma, USA) (prepared in PBS) for 20 min at room tem- tified borderlines. Subsequently, a serial dilution ratio (3:4
perature followed by 2 washed with PBS. The cells were dilution starting from 35,000 cells/well) was chosen to create
then stained with 0.1, 0.25, or 0.4% Trypan Blue (Sigma, a standard curve spanning the chosen range (Table 1). The
USA) diluted in PBS for 10, 30, or 60 min at room tempera- cells were plated according to the dilutions and monitored
ture. The cells were then washed twice with PBS to com- regularly for signs of adherence and adaptation of natural
pletely remove any remaining trypan blue residues and avoid morphology in culture. Once the cells fulfilled these condi-
an artifact signal. The absorbance was then measured using tions (after 6 h for A549 cells and 20 h for MDA cells), the
BioTek’s plate reader ELx808 at absorbance of 450 nm, cells are directly fixed for staining to avoid increase in cell
490 nm, and 630 nm. count.
Beyond fixing the cells, the application of PFA suffi-
Microscopy ciently permeabilizes the cell membrane to allow for the
entrapment of trypan blue intracellularly and the full cov-
Some of the images were acquired with Olympus IX53 erage of the cell body (Fig. 1). This method of staining
inverted microscope equipped with Olympus DP75 cam- results in dye density signals proportional to the volume
era (resolution of 5760 × 3600 pixels and pixel size of
5.86 × 5.86  µm) with Cell Sens Entry software (version
1.17). Images were acquired with 10X (numerical aperture
0.50) objective lenses at 1600 × 1200 pixels and 72.0 1/in
resolution in both X and Y axis. The rest of the images were
acquired using Olympus IX73 inverted microscope with
Olympus DP22 camera (resolution of 1920 × 1440 pixels
and pixel size of 3.69 × 3.69 µm) and DP camera acquisition
software. Images were acquired with 4X (numerical aperture
0.13) or at 1920 × 1440 pixels resolution.

Statistical analysis

T test and one-way ANOVA tests were conducted to statis-

tically analyze the cell-count data and identify differences
between the results of the different counting methods. The
significance was taken to be p < 0.05. Excel was used to con-
duct the goodness-of-fit test (R2) as a part of the standard
curve linear regression analysis. Otherwise, all statistical
analyses (including Pearson’s correlation) were carried out
using GraphPad Prism.

Results

Optimization of the trypan blue assay parameters

The use of this assay requires the optimization of multiple

parameters: (1) standard curve range and ratio of serial dilu-
tions; (2) standard curve plating duration; and (3) duration
Fig. 1  Dye absorption by a A549 and b MDA-MB-231 cells after fix-
of fixing and staining.
ation and staining with trypan blue. Scale bars representing 100 µm.
To optimize the standard curve range, multiple concentra- Images acquired with 10X magnification through Olympus IX53
tions of cells approaching the two extremes (full confluency inverted microscope

13
 Histochemistry and Cell Biology

and thickness of the cell; rounded cells are less spread but Triton X-100) to lightly permeabilize cell membrane and
thicker resulting in a darker shade of blue and the opposite increase dye internalization.
for cells that adapted to the normal morphology. This bal- To achieve the optimum staining procedure, several
ance between color density and surface area coverage com- parameters were optimized (Fig. 2). Initially, the staining
pensates for differences in cell morphologies and allows for resulting from ascending concentrations of trypan blue
a better cell-count estimation. (0.1, 0.25, and 0.4%) was investigated showing a positive
Staining duration was optimized over 0.5, 1, 2, 5 h, and concordance between the concentration of trypan blue and
overnight. As shown in Fig. 2, although the trypan blue sig- dye absorption (Fig. 2a). With the use of the highest con-
nal could be visualized even at shorter staining duration, centration, 0.4% of trypan blue, the staining duration was
the strength of the signal directly correlated with the stain- optimized over 10, 30, and 60 min (Fig. 2b). Positive cor-
ing duration. Therefore, to achieve the best visualization of relation was similarly observed between the staining dura-
the cells, overnight staining was applied to the subsequent tion and dye absorption. The additional permeabilization
experiments. Inefficiencies in the trypan blue internaliza- of the cells with 0.5% Triton-X100 prior to the staining
tion into the cells might require optimization of the fixa- was found to further increase trypan blue uptake (Fig. 2c).
tion method and staining duration. Otherwise, one might Therefore, permeabilization with Triton X-100 can be used
consider the application of a highly diluted detergent (e.g., to reduce the staining duration; 30 min dye uptake by Tri-
ton X-100 permeabilized cells closely approaches that of

Fig. 2  Optimization of trypan blue staining parameters including a Triton X-100, and d absorbance acquisition wavelength. Data points
trypan blue concentration (0.1, 0.25, and 0.4% trypan blue), b stain- presented as mean± SEM
ing duration (10, 30, and 60 min), c pre-permeabilization with 0.5%

13
Histochemistry and Cell Biology

non-permeabilized cells stained for 1 h. Taken together, Standard curve accuracy verification
the optimum staining conditions were found to be staining
with 0.4% trypan blue at room temperature for 30 min if The gradual reduction in cell density along the standard
preceded by 20 min of Triton X-100 permeabilization or curve serial dilutions and the corresponding decrease in
1 h without permeabilization. trypan blue staining were visually verified using phase
Furthermore, absorbance measurement at the three dif- contrast microscopy (Fig. 3). The trypan blue absorbance
ferent wavelengths (450, 490, and 630 nm) was found to signal was then quantified using BioTek’s plate reader
provide sufficient signal to estimate cell count, indicating the Elx808. As seen in the absorbance measurements of the
flexibility of the assay (Fig. 2d). However, as the absorbance standard curves for both cell lines (Fig. 4a, b), a significant
measurement at wavelength of 630 nm provided the strong- correlation [correlation results for the A549 cell line: Pear-
est signal, and the highest correlation and accuracy, it was son’s correlation: 0.9971; p < 0.0001; goodness of fit (R2):
used in this paper for all subsequent measurements. 0.9942; correlation results for the MDA-MB-231 cell line:
Pearson’s correlation: 0.9919, p < 0.0001; goodness of fit

Fig. 3  Standard curve serial densities of A549 cells before and after fixation and staining with trypan blue. Scale bars representing 200  µm.
Images were taken at a 4X magnification using Olympus IX73 inverted microscope

13
 Histochemistry and Cell Biology

counting to avoid cell damage due to prolonged cell incuba-

tion at room temperature.
These cell-count measurements were then compared to
those of the new trypan blue colorimetric assay. As can be
seen in the graphs (Fig. 5b, c), the measurement from both
methods was relatively close and fluctuates narrowly around
the initially seeded counts. However, in comparison with the
haemocytometer’s measurements, the counts from the new
assay were found to be more representative of the initially
seeded cell counts and more precise in terms of triplicate
measurements. This indicates the comparatively higher
accuracy and precision of the new trypan blue colorimetric
assay.

Efficiency of the assay in estimating arbitrary

samples with unknown counts

Arbitrary values were chosen to assess the power of the

assay and its ability to estimate unknown cell counts. As
seen in the graphs of the two cell lines (Fig. 6), the results
from the two assays fluctuate closely around the initially
seeded counts. However, at multiple data points, the trypan
blue assay proved to yield closer cell counts to those initially
seeded. Furthermore, the ranges of inter-triplicate variations
are significantly smaller in the trypan blue spectrophotomet-
ric assay’s measurements.

Cytotoxicity assay

The arbitrarily chosen values above were treated with 5%

DMSO to assess the ability of the trypan blue spectropho-
Fig. 4  Absorbance standard curve and goodness-of-fit test (R2) as a tometric assay in measuring treatment-induced cytotoxicity.
measure of correlation for a A549 and b MDA-MB-231 cells. Data
points presented as mean± SEM The assay measurements of the DMSO treated cells (Fig. 7)
show a clear reduction in cell count, conforming to the
results of the standard haemocytometer counting.
(R2): 0.9839] is achieved with low variations amongst the Furthermore, the resolution power of the assay was
triplicate measures at each point. assessed by measuring the cytotoxic effect of 3 consecu-
tive concentrations of DMSO, 1, 2, 3, 4, 5, 7.5, and 10%
Comparison between cell‑count measurements (Fig. 8a). Starting at an initial count of ~ 40,000 cells per
using the conventional haemocytometer well, a gradual decrease is seen in cell count in correspond-
and the new trypan blue assay ence with the increase in the applied DMSO concentra-
tion. Hence, this demonstrates the ability of the assay to
Standard curve cell-count measurements were similarly done detect and differentiate between relatively close cell counts.
using the traditional haemocytometer counting method. Sev- Moreover, the cell-count measurements obtained with the
eral precautions were taken to increase the accuracy and new trypan blue assay positively correlated with those from
precision of these counts including: (1) ensuring proper traditional haemocytometer counting (Pearson’s correlation
detachment of cells and homogenous suspension; (2) con- coefficient: 0.9771; p < 0.0001; R2: 0.9546), verifying the
ducting triplicate measurements at each point; (3) counting validity of the new assay’s measurements.
cells in the four outer edges squares of the haemocytometer
grid; and (4) including cells only on the top and right edges
of each of these squares (Fig. 5a). Furthermore, images were
taken of the haemocytometer grid instead of instantaneous

13
Histochemistry and Cell Biology

Fig. 5  a Representative image of hemocytometer grid taken for tradi- obtained using haemocytometer and the new trypan blue assay. Data
tional hemocytometer counting and comparison between cell count, b points presented as mean± SEM (horizontal error bars representing
A549 (Pearson correlation: 0.9931 and p value for is > 0.0001), and SEM of haemocytometer count while the vertical ones representing
c MDA-MB-231 cells (Pearson correlation: 0.9866 and p < 0.0001) that of the new trypan blue assay)

Measurement of the cytotoxic effect of DMSO Discussion

exposure one range of treatment durations
Although various cell-counting approaches were developed
The assay was used to assess the gravity of damage inflicted to fulfil application-specific needs, these approaches are
on the cells in correlation with the increase in the incubation either low throughput (e.g., haemocytometer) or relatively
duration with 10% DMSO—commonly used concentration cost-demanding (e.g., flow cytometry). In this paper, we
for cryopreservation. As shown in Fig. 8b, increased incu- present a simple, cost-efficient, time-saving, high through-
bation with 10% DMSO is correlated with increased dam- put, and reasonably precise and accurate assay. The concept
age to the cells. This damage is more drastic in the initial behind this assay is to measure cell density by measuring the
10 min after which the rate of cell loss decreases. However, internalization of trypan blue by fixed, adherent cells. This
the viability of the cells must be taken into consideration as measurement is then used to infer the cell number in the well
the cells might be damaged regardless of their adherence. by plotting to a standard curve.
These results were found to highly correlate with those of As shown in the results’ section, the proposed trypan blue
the traditional haemocytometer cell count (Pearson’s corre- assay provided close estimation of the arbitrarily selected
lation coefficient: 0.9947; p < 0.0001; R2: 0.9894), verifying cell counts. Moreover, it efficiently captured the cytotoxic
the capacity of the assay to reflect and quantify cell damage. effect of DMSO treatment on cells which positively cor-
related with the increase in the applied DMSO percentage.
The negative effect of DMSO on cell viability was found as
well to correlate with the exposure duration. 10% DMSO

13
 Histochemistry and Cell Biology

Fig. 6  Cell-count measurements of arbitrary seeded counts of a A549

and b MDA-MB-231 cells using traditional haemocytometer count- Fig. 7  Cell-count measurements of the chosen arbitrary cell counts
ing (counted) and the new trypan blue assay (calculated from stand- after treatment with 5% DMSO calculations for a A549 cell line and
ard curve) in comparison with originally seeded counts. Data points b MDA-MB-231 cell line using traditional haemocytometer counting
presented as mean± SEM; **represents p ≤ 0.01, and ***represents (counted) and the new trypan blue assay (calculated from standard
p ≤ 0.001 curve). Data points presented as mean± SEM; *represents p ≤ 0.05,
** represents p ≤ 0.01, *** represents p ≤ 0.001, and **** represents
p ≤ 0.0001

is a commonly used concentration in preparation of cells

for cryopreservation. However, prolonged exposure to this the apoptosis-induced factor (AIF) from the mitochondria,
percentage is shown above to induce cell damage propor- activation of poly-(ADP-ribose)-polymerase (PARP), and
tional to the exposure duration. These results reflect the stimulation of caspase-3-independent apoptosis pathways
previous findings on the various mechanisms stimulated (Galvao 2014).
by DMSO to induce cell damage in correspondence to its In comparison with the haemocytometer standardized
concentration. For instance, DMSO application at concen- counting, the trypan blue colorimetric assay counting results
trations > 10% results in the perforation of the plasma mem- are relatively higher in accuracy, precision, and resolution.
brane and eventually apoptosis (De Ménorval MA 2010; Furthermore, the assay only requires manual counting of a
Notman 2006). On the other hand, lower concentrations of single value (the highest value in the standard curve) which
DMSO were found to result in the nuclear translocation of would be used as the reference count for all subsequent

13
Histochemistry and Cell Biology

experiments, proliferation, and cytotoxicity assessment).

To further increase the cost efficiency of the assay, the
diluted trypan blue dye can be recycled from the previous
experiments. Another advantage of this approach is that
in case of weak trypan blue signal, the cells can be simply
re-incubated with the trypan blue dye for a longer time
period. Furthermore, the experiment plate stained with
the trypan blue can be saved for back-referencing by stor-
ing it at 4 °C in PBS with the option of re-staining in case
dye is eluted.
Due to the cell-volume dependence of the assay, the
measurements done through this assay are relatively repro-
ducible. This results in various advantages; for example,
the optimized standard curve could be reused for the calcu-
lations for subsequent assays on the condition that the fixa-
tion and staining processes are identical. This is especially
the case with stable cell lines with uniform morphology,
adherence, and growth rates. Furthermore, cell popula-
tions with irregular shapes would be accounted for in this
assay. Shape irregularities will result in changes in cell
thickness which will be directly accounted for by changes
the trypan blue signal (positive correlation between thick-
ness and signal). Moreover, it might be possible to use
the assay for counting cell populations heterogeneous in
volume and shape as these differences will be considered
in the standard curve and reflect on the count of the target
samples; this require on proper homogenous distribution
of these cells throughout the suspension. Third, overlap-
ping or clustering of seeded cells would result in a correla-
tive increase in dye intensity. Thus, the overlapped cells
would be as well accounted for in the final count of cells
per well.
Successfully conducting this assay requires several
Fig. 8  Cell-count measurements of A549 cells treated with a 0, 1, precautions and requirements: (1) ensuring homogenous
2, 3, 4, 5, 7.5, and 10% of DMSO for 24  h and b 10% DMSO for
increasing time periods (0–60  min) measured with the traditional
distribution of cells on the well surface by allowing the
hemocytometer counting (counted) and the new trypan blue assay cells to settle and adhere to the plate before transferring it
(calculated from standard curve). Data points presented as mean± to the incubator; (2) pipetting gently to avoid cell loss due
SEM to mechanical pressure; and (3) reducing the incubation
time of the cells in suspension before plating to enhance
calculations. Hence, the assay has a significantly increased cell adhesion and ensure proper morphology adaptation.
throughput and time efficiency. Standard counting with Although this approach has the shortcoming of being
haemocytometer, as well as many of the other approaches, a fixed-cell assay, its value lies in its potential use as a
requires the trypsinization and detachment of cells. This proliferation/cytotoxicity assay similarly to, but at a
process increases the risk of cell loss and affects cell count. lower cost than currently existing colorimetric assays
In this approach, we eliminated this risk by directly stain- (e.g., tetrazolium salt-based approaches). The use of this
ing and measuring the absorbance on the same cell-culture approach can be ideal for optimization procedures requir-
surface of the 96-well plate. ing multiple repetitions and assessing multiple parameters
Moreover, the use of this assay requires reagents used with wide ranges. In such cases, the use of the relatively
daily in cell-culture practice (i.e., trypan blue and para- expensive approaches would result in a high economic
formaldehyde) and a simple spectrophotometric plate burden. While economic options, such as the haemocy-
reader, usually at disposal to most. Therefore, the assay is tometer, would be impractical due to its low throughput
developed to be user-friendly, economic, and suitable for and potential time consumption. Accordingly, this assay
frequent, simple counting procedures (i.e., optimization can provide measurements relatively better than those of

13
 Histochemistry and Cell Biology

the haemocytometer and at a faster rate and economic cost. DeBlois R W, Bean P C (1970) Counting and sizing of submicron par-
Moreover, this assay could potentially be modified and ticles by the resistive pulse technique. Rev Sci Instrum 41(7):909–
916. https​://doi.org/10.1063/1.16847​24
expanded to other cell viability exclusion dyes with simi- Felice DL, Sun J, Liu RH (2009) A modified methylene blue assay
lar properties to those of trypan blue according to their for accurate cell counting. J Funct Foods 1:109–118. https​://doi.
availability in the lab, further increasing the feasibility org/10.1016/j.jff.2008.09.014
of the assay. Galvao JDB, Tilley M, Normando E, Duchen MR, Cordeiro MF (2014)
Unexpected low-dose toxicity of the universal solvent DMSO.
FASEB J 28:1317–1330
Acknowledgements R.H. is funded by Al Jalila Foundation Maioli E, Torricelli C, Fortino V, Carlucci F, Tommassini V, Pacini
(AJF201741), University of Sharjah and Boehringer Ingelheim A (2009) Critical appraisal of the MTT assay in the presence of
(120102). We thank Dangoor Education for supporting this work. rottlerin and uncouplers. Biol Proced Online 11:227–240
Mosmann T (1983) Rapid colorimetric assay for cellular growth and
OpenAccess  This article is distributed under the terms of the Crea- survival: application to proliferation and cytotoxicity assays. J
tive Commons Attribution 4.0 International License (http://creat​iveco​ Immunol Methods 65:55–63
mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribu- Notman RNM, O’Malley B, Anwar J (2006) Molecular basis for
tion, and reproduction in any medium, provided you give appropriate dimethylsulfoxide (DMSO) action on lipid membranes. J Am
credit to the original author(s) and the source, provide a link to the Chem Soc 128:13982–13983
Creative Commons license, and indicate if changes were made. Oliver MH, Harrison NK, Bishop JE, Cole PJ, Laurent GJ (1989) A
rapid and convenient assay for counting cells cultured in micro-
well plates: application for assessment of growth factors. J Cell
Sci 92:513–518
References Ongena K, Das C, Smith JL, Gil S, Johnston G (2010) Determining
cell number during cell culture using the Scepter cell counter. J
Camacho-Fernandez C, Hervas D, Rivas-Sendra A, Marin MP, Segui- Vis Exp 26:6
Simarro JM (2018) Comparison of six different methods to cal- Roehm NW, Rodgers GH, Hatfield SM, Glasebrook AL (1991) An
culate cell densities. Plant Methods 14:018–0297 improved colorimetric assay for cell proliferation and viability uti-
Collins CE, Young NA, Flaherty DK, Airey DC, Kaas JH (2010) A lizing the tetrazolium salt XTT. J Immunol Methods 142:257–265
rapid and reliable method of counting neurons and other cells in Tucker KG, Chalder S, al-Rubeai M, Thomas CR (1994) Measurement
brain tissue: a comparison of flow cytometry and manual counting of hybridoma cell number, viability, and morphology using fully
methods. Front Neuroanat 4:55 automated image analysis. Enzyme Microb Technol 16:29–35
Coulter WH (1953) Means for counting particles suspended in a fluid.
US Patent, 1953 Publisher’s Note Springer Nature remains neutral with regard to
De Ménorval MAML, Fernández ML, Reigada R (2010) Effects of jurisdictional claims in published maps and institutional affiliations.
dimethyl sulfoxide in cholesterol-containing lipid membranes: a
comparative study of experiments in silico and with cells. PLoS
One 7:e41733

13