Compound cytotoxicity profiling using quantitative high-throughput screening.

Abstract

The propensity of compounds to produce adverse health effects in humans is generally evaluated using animal-based test methods. Such methods can be relatively expensive, low-throughput, and associated with pain suffered by the treated animals. In addition, differences in species biology may confound extrapolation to human health effects.
The National Toxicology Program and the National Institutes of Health Chemical Genomics Center are collaborating to identify a battery of cell-based screens to prioritize compounds for further toxicologic evaluation.
A collection of 1,408 compounds previously tested in one or more traditional toxicologic assays were profiled for cytotoxicity using quantitative high-throughput screening (qHTS) in 13 human and rodent cell types derived from six common targets of xenobiotic toxicity (liver, blood, kidney, nerve, lung, skin). Selected cytotoxicants were further tested to define response kinetics.
qHTS of these compounds produced robust and reproducible results, which allowed cross-compound, cross-cell type, and cross-species comparisons. Some compounds were cytotoxic to all cell types at similar concentrations, whereas others exhibited species- or cell type-specific cytotoxicity. Closely related cell types and analogous cell types in human and rodent frequently showed different patterns of cytotoxicity. Some compounds inducing similar levels of cytotoxicity showed distinct time dependence in kinetic studies, consistent with known mechanisms of toxicity.
The generation of high-quality cytotoxicity data on this large library of known compounds using qHTS demonstrates the potential of this methodology to profile a much broader array of assays and compounds, which, in aggregate, may be valuable for prioritizing compounds for further toxicologic evaluation, identifying compounds with particular mechanisms of action, and potentially predicting in vivo biological response.

Full text

Page 1

284
VOLUME 116 | NUMBER 3 | March 2008 • Environmental Health Perspectives
Research
Animal toxicity data is used to predict human
toxicity, based on the assumption that adverse
effects in laboratory animals indicate the
potential for adverse effects in humans.
Various animal models have been developed
to evaluate a broad range of toxicologic
responses in order to classify compounds by
their potential for causing adverse health
effects in humans. These animal models
include acute, subchronic, and/or chronic
tests for end points such as oral, dermal, and
ocular toxicity; immunotoxicity; genotoxicity;
reproductive and developmental toxicity; and
carcinogenicity (Chhabra et al. 2003). Animal
tests, while clearly useful, can be relatively
expensive and low throughput. Furthermore,
intrinsic differences in species sensitivity can
confound the extrapolation of certain test
results to human health effects. Also, there is
increasing societal concern about the use of
animals in testing, especially in test methods
that might induce pain and suffering in the
treated animals (e.g., ocular toxicity). Thus,
there is increased interest among the interna-
tional scientific community in the develop-
ment, translation, validation, and use of
nonanimal alternative test methods for mak-
ing regulatory decisions [European Center for
the Validation of Alternative Methods
(ECVAM) 2007; Interagency Coordinating
Committee on the Validation of Alternative
Methods (ICCVAM) 2007; National
Research Council 2007; Tweats et al. 2007].
Given these scientific and societal issues, and
the increasing number of new compounds
requiring toxicity testing, the National
Toxicology Program (NTP) recently began a
major initiative to develop a high-throughput
screening (HTS) program to prioritize com-
pounds for further in-depth toxicologic evalu-
ation, identify mechanisms of action for
further investigation, and develop predictive
models for in vivo biological response (NTP
2004a; Tice et al. 2007).
In support of this initiative, the NTP and
the National Institutes of Health (NIH)
Chemical Genomics Center (NCGC) formed
a partnership in 2005 to a) develop a library
of compounds suitable for HTS that had
been characterized to some degree by tradi-
tional toxicologic testing methods, b) identify
and/or develop cell-based or biochemical
HTS assays potentially informative for in vivo
toxicologic effects, and c) profile the com-
pound library in these HTS assays. The ulti-
mate goal of this collaboration is to establish
in vitro “signatures” of in vivo rodent and
human toxicity by comparing the data gener-
ated in HTS assays with the rich historical
database generated by the NTP using tradi-
tional in vivo and in vitro toxicologic assays.
The U.S. Environmental Protection Agency
(EPA) has also recognized the potential of
high-throughput screening in toxicology test-
ing and has initiated the ToxCast program
for prioritizing the toxicity testing of environ-
mental chemicals (Dix et al. 2007).
HTS was developed by the pharmaceuti-
cal industry to evaluate the biological activity
of thousands of chemicals to identify poten-
tial drug candidates. Because HTS for this
purpose is generally performed at a single
concentration only (typically 10 µM), the
approach is characterized by a high preva-
lence of false positives and negatives. To
address these limitations and make HTS use-
ful for toxicology and chemical genomics, the
NCGC developed the quantitative high-
throughput screening (qHTS) paradigm
(Inglese et al. 2006). With this approach, all
compounds are screened for a concentration-
dependent response, which allows for a more
accurate assessment of biological activity.
Here, we report on the use of qHTS to
profile the cytotoxicity (the term “cytotoxicity”
is used to describe the cumulative effect of a
compound over a given period of time on cell
number, whether due to apoptosis, necrosis, or
a reduction in the rate of cell proliferation) of
1,408 compounds in 13 cell types using a
homogeneous, luminescent cell viability assay
Address correspondence to C.P. Austin, Chemical
Genomics Center, National Institutes of Health,
9800 Medical Center Dr., MSC 3370, Bethesda,
MD 20892-3370 USA. Telephone: (301) 217-5733.
Fax: (301) 217-5736. E-mail: austinc@mail.nih.gov
We thank A. Yasgar and P. Shinn for the com-
pound management.
This research was supported by the Intramural
Research Programs of the National Toxicology
Program of the NIEHS, and the National Human
Genome Research Institute of the NIH, and the
NIH Roadmap for Medical Research Molecular
Libraries Program
The authors declare they have no competing
financial interests.
Received 31 July 2007; accepted 21 November 2007.
Compound Cytotoxicity Profiling Using Quantitative High-Throughput
Screening
Menghang Xia,1Ruili Huang,1Kristine L. Witt,2Noel Southall,1Jennifer Fostel,3Ming-Hsuang Cho,1Ajit Jadhav,1
Cynthia S. Smith,2James Inglese,1Christopher J. Portier,2Raymond R. Tice,2and Christopher P. Austin1
1NIH Chemical Genomics Center, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland, USA;
2National Toxicology Program, and 3National Center for Toxicogenomics, National Institute of Environmental Health Sciences,
National Institutes of Health, Department of Health and Human Services, Research Triangle Park, North Carolina, USA
BACKGROUND: The propensity of compounds to produce adverse health effects in humans is gener-
ally evaluated using animal-based test methods. Such methods can be relatively expensive, low-
throughput, and associated with pain suffered by the treated animals. In addition, differences in
species biology may confound extrapolation to human health effects.
OBJECTIVE: The National Toxicology Program and the National Institutes of Health Chemical
Genomics Center are collaborating to identify a battery of cell-based screens to prioritize com-
pounds for further toxicologic evaluation.
METHODS: A collection of 1,408 compounds previously tested in one or more traditional toxico-
logic assays were profiled for cytotoxicity using quantitative high-throughput screening (qHTS) in
13 human and rodent cell types derived from six common targets of xenobiotic toxicity (liver,
blood, kidney, nerve, lung, skin). Selected cytotoxicants were further tested to define response
kinetics.
RESULTS: qHTS of these compounds produced robust and reproducible results, which allowed
cross-compound, cross-cell type, and cross-species comparisons. Some compounds were cytotoxic to
all cell types at similar concentrations, whereas others exhibited species- or cell type–specific cyto-
toxicity. Closely related cell types and analogous cell types in human and rodent frequently showed
different patterns of cytotoxicity. Some compounds inducing similar levels of cytotoxicity showed
distinct time dependence in kinetic studies, consistent with known mechanisms of toxicity.
CONCLUSIONS: The generation of high-quality cytotoxicity data on this large library of known com-
pounds using qHTS demonstrates the potential of this methodology to profile a much broader
array of assays and compounds, which, in aggregate, may be valuable for prioritizing compounds
for further toxicologic evaluation, identifying compounds with particular mechanisms of action,
and potentially predicting in vivo biological response.
KEY WORDS: 1,536-well, cell viability, NTP 1,408 compound library, PubChem, qHTS, RT-CES.
Environ Health Perspect 116:284–291 (2008). doi:10.1289/ehp.10727 available via
http://dx.doi.org/ [Online 22 November 2007]

Page 2

that measures the intracellular levels of adeno-
sine triphosphate (ATP) as an indicator of the
number of metabolically active cells. Selection
of the 1,408 compounds was based in part on
the availability of toxicologic data from stan-
dard tests for carcinogenicity, genotoxicity,
immunotoxicity, and/or reproductive and
developmental toxicity; all compounds were
tested at 14 concentrations from 0.59 nM to
92 µM. The cell types used in this evaluation
include corresponding human and rodent cells
derived from six tissues (liver, blood, kidney,
nerve, lung, skin) that are common targets of
xenobiotic toxicity. Using this approach, we
developed species- and cell type–specific
cytotoxicity profiles for each compound.
Furthermore, we demonstrate that compounds
with similar end point toxicity may exhibit dif-
ferent cytotoxicity kinetics, suggestive of differ-
ent mechanisms of action. In vitro profiling of
compounds promises to provide information
on molecular mechanisms of toxicity, and may
allow the creation of algorithms for predictive
in vivo toxicology.
Materials and Methods
The NTP 1,408 compound library. A collec-
tion of 1,408 substances (except where noted,
the term “substance” is used interchangeably
with “compound” here) was constructed for
characterization in qHTS assays (Smith et al.
2007; Tice et al. 2007); 1,408 is the number
of substances that can fit in a single 1,536-
well plate exclusive of controls. To allow eval-
uation of assay reproducibility, 55 of the
compounds were represented twice in the col-
lection, giving a total of 1,353 unique com-
pounds. Of these, 1,206 had been tested by
the NTP in one or more in vitro and/or
in vivo assays, including those for Salmonella
typhimurium mutagenicity (68%), chronic
toxicity/carcinogenicity (23%), reproductive
toxicity (3%), developmental toxicity (3%),
and immunotoxicity (1%). Also included
were 147 reference compounds identified by
the ICCVAM for the development and/or
validation of alternative in vitro test methods
for dermal corrosivity, acute toxicity, and
endocrine activity. Molecular weights of all
compounds ranged from approximately 32
(methanol) to 1,300 (actinomycin D), with
95% of the compounds having a molecular
weight that was < 400. Functionally, the
NTP library of 1,408 compounds includes
solvents, fire retardants, preservatives, flavor-
ing agents, plasticizers, therapeutic agents,
inorganic and organic pollutants, drinking-
water disinfection by-products, pesticides,
and natural products. Compounds excluded
from this NTP collection were those consid-
ered excessively volatile and those not soluble
in dimethylsulfoxide (DMSO), the solvent
used for compound transfer. A complete list
of the NTP 1,408 compounds and full
chemical descriptions are publicly available
(PubChem 2007a).
All compounds were received from suppli-
ers via the NTP chemistry support contract in
1-mL aliquots at 10 mM dissolved in DMSO
and stored at –80°C in Matrix TrakMates 2D
bar-coded storage tubes (Thermo Fisher
Scientific, Hudson, NH). Purity and identity
information for the compounds was obtained
from the suppliers and, in the case of com-
pounds used in NTP studies, from the charac-
terizations performed in support of those
studies. With the exception of natural com-
pounds and other known mixtures, most com-
pounds were > 90% pure.
Sets of compounds prepared as 10-mM
stock solutions and stored in 96-well plates
were compressed into 384-well plates. From
these plates, fifteen 384-well plates containing
the 1,408 compounds at 2.236-fold dilutions
were prepared using an Evolution P3system
(PerkinElmer, Inc., Wellesley, MA). The sets
of 384-well plates composing the dilution
series were then compressed into multiple
1,536-well plates by interleaved quadrant
transfer. During screening, working copies of
the 1,536-well compound plates were stored at
room temperature for up to 6 months; back-up
copies were heat sealed and stored at –80°C.
Cell types and culture conditions. Human
embryonic kidney cells (HEK293), human
hepatocellular carcinoma cells (HepG2),
human neuroblastoma cells (SH-SY5Y and
SK-N-SH), human leukemia T cells (Jurkat,
clone E6-1), normal human foreskin fibrob-
lasts (BJ), normal human lung fibroblasts
(MRC-5), normal human vascular endothelial
cells (HUVEC), rat hepatoma cells (H4-II-E),
mouse neuroblastoma cells (N2a), and mouse
fibroblast cells (NIH 3T3) were purchased
from the American Type Culture Collection
(ATCC, Manassas, VA). Human renal mesan-
gial cells obtained from adult kidney tissue
were kindly provided by J. Kopp (National
Institute of Diabetes and Digestive and Kidney
Diseases, NIH, Bethesda, MD). Rat renal
proximal tubule cells were freshly isolated from
rat kidney by In Vitro ADMET Laboratories,
LLC (Rockville, MD). All cells chosen for test-
ing represent target tissues of interest in toxi-
cology from both human and rodent sources;
most were transformed lines, but some were
nontransformed or primary cells. All are well-
characterized cells, produce uniform popula-
tions, are relatively easy to culture, and are
amenable to a 1,536-well format. The origin
and status of each cell type are presented in
Table 1. We determined the doubling times
for each cell line using a hemocytometer at des-
ignated intervals (McAteer and Davis 2002).
We cultured cells in ATCC complete
Eagle’s minimal essential medium (N2a,
H4-II-E, SK-N-SH, MRC-5, BJ, HepG2, and
HEK293), ATCC complete Dulbecco’s mini-
mal essential medium (DMEM) (NIH 3T3),
RPMI 1640 (Jurkat and human renal mesan-
gial cells), or ATCC complete DMEM/F-12
(SH-SY5Y and rat renal proximal tubule cells).
These media were supplemented with 10%
fetal bovine serum (Invitrogen, Carlsbad, CA),
50 U/mL penicillin, and 50 µg/mL strepto-
mycin (Invitrogen). Human HUV-EC-C cells
were cultured in ATCC Kaighn’s F12K
medium supplemented with 0.1 mg/mL
heparin (Sigma-Aldrich, St. Louis, MO),
0.04 mg/mL endothelial cell growth supple-
ment (Sigma-Aldrich). Each cell type was
maintained at 37°C under a humidified atmos-
phere and 5% CO2.
Cell viability assay. We measured cell
viability using a luciferase-coupled ATP quanti-
tation assay (CellTiter-Glo; Promega, Madison,
WI). In this assay, luminescent signal is propor-
tional to amount of ATP and thus to the num-
ber of metabolically competent cells; cell injury
and death result in a dramatic decrease in intra-
cellular ATP levels (Crouch et al. 1993). Cells
were dispensed at 1,000–2,000 cells/5 µL/well
in tissue-culture treated 1,536-well white/solid
bottom assay plates (Greiner Bio-One North
America, Monroe, NC) using a Flying Reagent
Dispenser (Aurora Discovery, Carlsbad, CA).
All but Jurkat cells (which are grown in
suspension) were incubated at 37°C for 5–6 hr
to allow for cell attachment, followed by addi-
tion of compounds via pin tool (Kalypsys,
Compound cytotoxicity profiling using qHTS
Environmental Health Perspectives • VOLUME 116 | NUMBER 3 | March 2008
285
Table 1. Cell types tested.
Species
Human
Human
Human
Human
Human
Human
Human
Human
Human
Rat
Rat
Mouse
Mouse
Cell type
HEK293
HepG2
SH-SY5Y
SK-N-SH
Jurkat
BJ
HUV-EC-C
MRC-5
Mesangial
Proximal tubules
H-4-II-E
N2a
NIH 3T3
OriginStatusa
T
T
T
T
T
NT
NT
NT
NT
P
T
T
NT
Doubling time (hr)
28
37
61
56
22
147
148
132
22
ND
22
25
28
Embryonic kidney cells
Hepatocellular carcinoma
Neuroblastoma
Neuroblastoma
T-cell leukemia
Foreskin fibroblasts
Vascular endothelial cells
Lung fibroblasts
Renal glomeruli
Cells from kidney
Hepatoma
Neuroblastoma
Fibroblasts, embryonic
Abbreviations: ND, not determined; NT, nontransformed; P, primary; T, transformed.

Page 3

San Diego, CA). After compound addition,
plates were incubated for 40 hr at 37°C; incu-
bation duration was based on results of assay
optimization experiments (data not shown) and
is the longest duration that can be used without
evaporation-induced edge effects occurring in
1,536-well plates. At the end of the incubation
period, 5 µL of CellTiter-Glo reagent was
added, plates were incubated at room tempera-
ture for 30 min, and the luminescence intensity
of each well was determined using a ViewLux
plate reader (PerkinElmer, Shelton, CT).
To avoid false positive or false negative
results due to luciferase inhibition or signal
interference from optically active compounds,
respectively, we tested all compounds in the
presence of a physiologically relevant concen-
tration of ATP in each well (data not shown).
Of the 25 compounds that inhibited luciferase
to any degree, the extent of luciferase inhibi-
tion was not sufficient to interfere with meas-
urement of cytotoxicity.
qHTS. Compound formatting and qHTS
were performed as described previously
(Inglese et al. 2006). Two positive control
compounds were dispensed on each plate: dox-
orubicin in a concentration series from 0.7 nM
to 100 µM in DMSO in column 1, and
tamoxifen in concentration series from 0.23 to
100 µM in DMSO in column 2. DMSO only
was dispensed into column 3, and tamoxifen at
100 µM in DMSO in column 4.
Transfer of 23 nL of experimental com-
pounds was performed via pin tool (Cleveland
and Koutz 2005), resulting in final concentra-
tions of 0.59 nM–46 µM of compound, and
0.45% DMSO. To achieve a highest final
compound concentration of 92 µM (DMSO
concentration, 0.9%), 23 nL was transferred
twice from the highest concentration mother
plate into each well of the assay plate; control
plates using DMSO only at this higher con-
centration were also included. DMSO toler-
ance experiments with each cell type showed
no effect on viability at concentrations up to
1% (data not shown).
Inclusion of 55 duplicate compounds in
the compound plate allowed for measurement
of intraexperimental reproducibility in every
assay at every concentration. To test inter-
experimental reproducibility, qHTS of
HepG2 cells was performed across the entire
1,408 collection 3 times, with one screen
being done on each of 3 different weeks using
identical assay conditions.
Real-time cell electronic sensing (RT-CES)
cytotoxicity assay. HepG2 cells in the presence
of compounds at concentrations from 0.14 to
100 µM were monitored by measurements of
electrical impedance (ACEA Biosciences Inc.,
San Diego, CA) every 10 min for the first hour,
then every hour for 70–90 hr. Continuous
recording of impedance in cells was reflected by
cell index value (Xing et al. 2005).
Data analysis and curve fitting. Analysis
of compound concentration–response data
was performed as previously described
(Inglese et al. 2006). Raw plate reads for each
titration point were first normalized relative
to the tamoxifen control (100 µM, 100%)
and DMSO-only wells (basal, 0%), then cor-
rected by applying a pattern correction algo-
rithm using compound-free control plates
(i.e., DMSO-only plates) at the beginning
and end of the compound plate stack.
Concentration–response titration points for
each compound were fitted to the Hill equa-
tion (Hill 1910), yielding concentrations of
half-maximal inhibition (IC50) and maximal
response (efficacy) values. Compounds were
designated as classes 1–4 according to the type
of concentration–response curve observed
(Inglese et al. 2006): Class 4 compounds show
no concentration response or have no signifi-
cant activity point, that is, ≥ 3 SD of the activ-
ity at the lowest concentration tested; class 3
compounds display significant activity only at
the highest concentration tested; class 2 com-
pounds have incomplete curves (i.e., no
low-concentration asymptote) and class 1
compounds have complete response curves
(i.e., two asymptotes). Class 1 or 2 com-
pounds were further divided into subclasses
based on efficacy and quality of fit (R2).
Compounds with R2> 0.9 and high (> 80%)
efficacy were designated as subclass a, and
compounds with R2> 0.9 and low efficacies
(30–80%) as subclass b. Compounds with
class 1a, 1b or 2a curves were generally
selected for follow-up analyses as they repre-
sented high-confidence data. Compound
activity correlations between different assays or
experiments were assessed by calculating the
Pearson correlation coefficient. Hierarchical
clustering of compound activity patterns across
different assays was performed within Spotfire
DecisionSite 8.2 (Spotfire Inc., Cambridge,
MA) using the correlation of log IC50values as
the similarity metric. All the normalized cyto-
toxicity data obtained for the 1,408 compounds
tested in the 13 cell types have been deposited
into PubChem (http://www.ncbi.nlm.nih.gov/
sites/entrez?db=pcassay, search term “NCGC
[sourcename] AND Viability [AssayName]”)
(PubChem 2007b).
Results
A total of 1,408 known compounds previ-
ously studied by the NTP were successfully
arrayed in 1,536-well plates in uniform sol-
vent and concentrations. A luminescent cell
viability assay was selected to investigate the
feasibility of using 14-concentration qHTS
for in vitro toxicity characterization of these
Xia et al.
286
VOLUME 116 | NUMBER 3 | March 2008 • Environmental Health Perspectives
Figure 1. Intraexperiment reproducibility. The figure shows example replicate dose–response curves for
colchicine, cycloheximide, progesterone, tetraethylene glycol diacrylate, and sodium dichromate dehydrate
in rat primary kidney proximal tubule cells.
20
0
–20
–40
–60
–80
–100
–120
Percent activity
Colchicine
Colchicine
Cycloheximide
Cycloheximide
Progesterone
Progesterone
Tetraethylene glycol diacrylate
Tetraethylene glycol diacrylate
Sodium dichromate dihydrate
Sodium dichromate dihydrate
–10–9 –8–7 –6–5 –4 –3
Compound log (M)
Table 2. Sensitivity (mean ± SD) of the 13 cell types to the tamoxifen positive control.
Cell type
Jurkat
HepG2
Rat kidney proximal tubule
HUV-EC-C
SK-N-SH
H-4-II-E
SH-SY5Y
MRC-5
N2a
NIH 3T3
HEK293
BJ
Mesangial
R2
Hill coefficient
2.0 ± 0.2
4.4 ± 0.5
4.7 ± 0.3
4.2 ± 0.5
3.4 ± 0.7
2.7 ± 0.8
3.1 ± 0.7
2.9 ± 1.2
4.0 ± 0.8
2.6 ± 0.8
3.8 ± 1.0
3.1 ± 1.0
2.4 ± 0.8
IC50(µM)
8 ± 1
43 ± 8
44 ± 5
46 ± 5
57 ± 13
63 ± 15
65 ± 15
68 ± 17
72 ± 13
79 ± 4
79 ± 10
79 ± 10
79 ± 10
CV Z’ factor
0.86 ± 0.06
0.87 ± 0.03
0.88 ± 0.02
0.82 ± 0.07
0.84 ± 0.06
0.91 ± 0.02
0.71 ± 0.06
0.83 ± 0.07
0.89 ± 0.05
0.44 ± 0.10
0.84 ± 0.08
0.80 ± 0.08
0.91 ± 0.03
S/B
0.98 ± 0.00
0.96 ± 0.02
0.96 ± 0.01
0.95 ± 0.02
0.92 ± 0.02
0.93 ± 0.02
0.91 ± 0.03
0.90 ± 0.05
0.95 ± 0.05
0.91 ± 0.02
0.91 ± 0.04
0.89 ± 0.04
0.85 ± 0.03
10.0 ± 8.7
8.8 ± 5.7
7.6 ± 5.1
8.9 ± 4.3
10.3 ± 5.7
8.5 ± 7.3
12.4 ± 7.1
9.3 ± 4.7
8.6 ± 6.7
9.7 ± 6.5
9.8 ± 5.1
10.4 ± 4.6
6.9 ± 4.6
83 ± 10
25 ± 1
22 ± 1
26 ± 1
27 ± 1
25 ± 1
12 ± 3
26 ± 1
25 ± 1
4 ± 1
25 ± 1
20 ± 4
25 ± 1
S/B, signal-to-background ratio.

Page 4

compounds and was easily adapted to a
1,536-well format. Robust and reproducible
data were obtained on the NTP compound
library in this cytotoxicity assay. These data
demonstrate the utility of the qHTS approach
to toxicity investigations and will facilitate
interpretation of data from subsequent cell-
based qHTS assays conducted at NCGC on
the NTP compound collection.
Establishment of a positive control for cyto-
toxicity. qHTS is optimally informative when
results are compared to a positive control.
Establishment of a universal positive control
for these studies was challenging because toxic-
ity was to be tested across a wide variety of cell
types. Initial testing of the 1,408 compounds
identified two, tamoxifen and doxorubicin,
that were toxic in all cell types and were there-
fore included as positive controls on all assay
plates. Tamoxifen was used as the control to
normalize response unless otherwise noted.
Tamoxifen concentration–response curves
were very consistent among plates within a cell
type although potency was cell-type specific,
with Jurkat cells being the most sensitive (IC50
= 8.3 µM), and NIH3T3, HEK293, BJ, and
mesangial cells least sensitive (IC50= 79 µM).
The rank order of potency among the 13 lines
is presented in Table 2. Z´-factors (Zhang et al.
1999) were between 0.44 and 0.91, and coeffi-
cients of variation (CV) were between 6.9 and
12.4, based on 234 plates used to test
1,408 compounds in 13 cell types (Table 2).
Both sets of Z-factors and CV values indicate
that the assays under the qHTS platform are
very robust and suitable for accurately identify-
ing cytotoxic compounds.
Assay reproducibility. To evaluate intra-
experimental reproducibility, we compared
results for the 55 compounds present in dupli-
cates in the collection. The correlation of IC50
for these replicates in all 13 cell lines was 0.71
(p < 0.001). In rat renal proximal tubule cells,
the IC50values for the 11 duplicate com-
pounds that displayed a concentration–
response relationship showed an excellent cor-
relation (R2= 0.97); all compounds cytotoxic
in one replicate were cytotoxic in the other,
and the same was true for noncytotoxic com-
pounds. Examples of concentration–response
curves for duplicate cytotoxic compounds in
rat renal proximal tubule cells are shown in
Figure 1.
To evaluate interexperimental reproducibil-
ity, we performed qHTS on the 1,408 com-
pounds in HepG2 cells once per week for
3 weeks. The concentration–response relation-
ship of each compound was classified (see
details in “Materials and Methods”). The
outcome for each compound in each weekly
iteration was designated as cytotoxic (classes 1a,
1b, and 2a compounds), not cytotoxic (class 4)
or inconclusive (other curve classes). Of
the 1,353 unique compounds, 90% of the
compounds had the same outcome in all three
runs; 9.3% of the compounds had the same
outcome in two runs only; and 0.7% of the
compounds had different outcomes in all three
runs. Of the 116 compounds that were cyto-
toxic in at least one run, 102 (87.9%) were
cytotoxic in all three runs. The regression cor-
relations of the IC50values for the three inde-
pendent runs were calculated (IC50value for
nontoxic compounds was set to 92 µM, the
maximum concentration used in all experi-
ments) and good linear correlations were
observed, with an average correlation coeffi-
cient (R2) of 0.74 (p < 0.001; n > 1000).
Cytotoxicity of the 1,408 NTP compounds.
Concentration–response relationship classifica-
tion, potency, efficacy, and Hill coefficient
were determined for all compounds and used
to evaluate cytotoxicity of each compound in
the 13 cell types tested. Of the 1,353 com-
pounds tested, 428 showed cytotoxicity in at
least one cell type. Results in rat primary kid-
ney proximal tubule cells are presented to illus-
trate the type of data obtained for each cell
type (Table 3). Seventy-nine compounds (6%
of the total) produced class 1a, 1b, or 2a
curves; these represent the highest confidence
data for cytotoxicity. Another 96 (7%) pro-
duced class 2b or 3 curves, which represent
lower confidence data for cytotoxic activity.
The remaining 1,233 compounds (87%) did
not induce a concentration-related increase in
cytotoxicity and were classified as class 4. By
comparing the percentage of compounds that
produced class 1a, 1b, or 2a curves (percent-
ages of all class 1–3 compounds are shown in
parentheses) in each of the cell types
(Figure 2A), a rank order of sensitivity was
derived: Jurkat: 10.6% (17.2%) ≈ SH-SY5Y:
10.5% (20.1%) > N2a: 9.3% (17.0%) ≈
NIH 3T3: 9.1% (17.6%) > H-4-II-E: 8.5%
(18.4%) > SK-N-SH: 6.4% (10.6%)
> HEK293: 5.7% (11.9%) ≈ rat primary prox-
imal tubule: 5.5% (12.4%) > HUV-EC-C:
4.5% (9.4%) ≈ mesangial: 4.3% (8.7%) ≈
HepG2: 4.3% (7.8%) ≈ BJ: 4.0% (9.1%) ≈
MRC-5: 3.8% (7.2%).
The IC50values of the cytotoxic com-
pounds ranged from double-digit nM to
80 µM (Figure 2B). For example, in rat pri-
mary kidney proximal tubule cells, three com-
pounds had IC50values <100 nM, 5 had
values between 100 nM and 1 µM, 10 had
values between 1 and 10 µM, and 60 had val-
ues between 10 and 80 µM (Table 3).
Compound profiles across cell types. We
used the differential cytotoxicity of com-
pounds across cell types and species to con-
struct profiles of cell lines according to the
compounds that were toxic to them and pro-
files of compounds according to the cell lines
in which they were toxic.
To create compound profiles, compounds
that demonstrated activity in at least one
cell type were clustered by their IC50values
(Figure 3) across the 13 cell types. Some
Compound cytotoxicity profiling using qHTS
Environmental Health Perspectives • VOLUME 116 | NUMBER 3 | March 2008
287
Table 3. Curve classification and potency distribution of the NTP 1,408 compounds in rat primary kidney
proximal tubule cells.
Curve classificationa
IC50range1a1b
< 100 nM (%) 1 (0.07)2 (0.1)
1 µM–100 nM (%) 5 (0.4)0
10 µM–1 µM (%)10 (0.7)0
80 µM–10 µM (%)23 (1.6) 7 (0.5)
> 92 µM (%)
2a
0
0
0
2b
0
0
3
0
0
0
4
4 (0.3)
71 (5.0) 30 (2.1)21 (1.5)
1,233 (87.6)
aSee “Materials and Methods” for curve class definitions.
Figure 2. Pharmacological profile of compound activity. (A) Percentage of activity in each class identified
from all compounds in 13 cell lines. (B) Potency distribution of all compounds in 13 cell lines.
100
20
15
10
5
0
HEK293
HepG2
SH-SY5Y
SK-N-SH
Jurkat
BJ
HUV-EC-C
MRC-5
Mesangial
Proximal tubules
H-4-II-E
N2a
NIH 3T3
HEK293
HepG2
SH-SY5Y
SK-N-SH
Jurkat
BJ
HUV-EC-C
MRC-5
Mesangial
Proximal tubules
H-4-II-E
N2a
NIH 3T3
Percent activity
AB
25
20
15
10
5
0
Percent activity
Class 4
Classes 2b, 3
Classes 1a, 1b, 2a
10 μM–80 μM
1 μM–10 μM
100 nM–1μM
< 100 nM

Page 5

compounds, such as digitonin and the three
rosaniline derivatives (hexamethyl-p-rosaniline
chloride, pararosaniline hydrochloride, and
malachite green oxalate) produced cytotoxicity
in all cell types at similar concentrations
(Figure 4). Other compounds, such as tetra-
methylthiuram disulfide and methyl mercuric
(II) chloride, were found to be selectively toxic
to a particular cell type. For example, tetra-
methylthiuram disulfide is 250-fold more
potent in SH-SY5Y than the next most
responsive cell type, and methyl mercuric (II)
chloride is 50-fold more potent in NIH 3T3
than the next most responsive cell type.
Comparing species, some compounds such
as zinc pyrithione showed similar toxicity in
analogous human and rodent cell types, while
others such as diphenylurea showed toxicity in
cells from one species but not the other. In
addition to compounds with differential toxic-
ity between particular cell types in the two
species, we were able to identify compounds
that were generally cytotoxic either toward
rodent or human cell types (determined by a
t-test on IC50values of the nine human versus
the four rodent cell types). Thirty-nine com-
pounds showed significant species selectivity
(p < 0.05), 34 with greater potency in the
rodent cells, and 5 with greater potency in
human cells. For example, dimethylamino-
azobenzene, a rodent liver carcinogen (Biswas
and Khuda-Bukhsh 2005), was more toxic to
rodent cells while digoxin, a Na+,K+-ATPase
inhibitor (Sweadner 1989), was more toxic to
human cells. Similar results for digoxin have
been noted previously (Hengstler 1999), and
other rodent-selective compounds have also
been identified (Klaunig 2003).
Cell-type profiles defined by compound
responses. Similar to what was observed for
cross-species comparisons, a range of related-
ness was seen in the comparison of cell types.
In some cases of closely related cell lines, such
as MRC-5 and BJ human fibroblasts, the pat-
tern of compound activity was quite similar
(compare columns MRC-5 and BJ in heat
map in Figure 3). However, in other closely
related cell lines, such as the SK-N-SH and
SH-SY5Y neuroblastoma cells (the former is
the parental line of the latter), activity patterns
were quite different (compare columns
SK-N-SH and SH-SY5Y in heat map in
Figure 3). Homologous cell types from differ-
ent species produced patterns that were
less related than those from nonhomologous
cell types from the same species (compare
columns SH-SY5Y and N2a, and HepG2 and
H-4-IIE in heat map in Figure 3). For exam-
ple, the human SH-SY5Y and rat N2a neuro-
blastoma cell lines produced quite different
response patterns, as did the human HepG2
and rat H-4-II-E hepatoma lines. In general,
rodent cells were more sensitive than human
cells: on average, 6% of the compounds were
Xia et al.
288
VOLUME 116 | NUMBER 3 | March 2008 • Environmental Health Perspectives
Figure 3. Compound activity patterns clustered by cell and species type. The compounds with an IC50of
< 92 µM in at least one cell type are selected and arranged in the order of a hierarchical clustering based
on their IC50 values as shown in the dendrogram on the left side of the heat map. Neighboring compounds
share similar activity patterns. In the figure, each row represents a compound, and each column is a cell
type. Compound activity in each cell line is colored according to potency (IC50) range. Potent compounds
are deeper shades of red and nontoxic compounds are white. Assays are also clustered by similarity in
their compound IC50patterns as shown in the dendrogram on the top of the heat map.
18 34102 446
> 50 μM
10–50 μM
1–10 μM
< 1 μM
Figure 4. Compound activity across different species. Compound activity in each cell line is colored
according to potency (IC50) range. The figure shows examples of compounds that are more cytotoxic to
human cells (top two rows) or to rodent cells (middle rows), and compounds with similar levels of cytotoxi-
city in human and rodent cells (bottom rows).
> 50 μM
10–50 μM
1–10 μM
< 1 μM

Page 6

designated as active (class 1a, 1b, and 2a)
against human cells, compared to 8% active
against rodent cells (Figure 2A). This differ-
ence is statistically significant (p < 0.001).
Structure–activity relationships. As the
1,353 compounds were chosen on the basis of
pre-existing toxicity data rather than struc-
tural relatedness, this modest set of com-
pounds was too diverse to rigorously define
any structure–activity relationships. However,
we did observe that some structurally related
compounds induced similar levels of cytotoxic-
ity across cell types. For example, the three
thiopurines—azathioprine, 6-mercaptopurine
monohydrate, and 6-thioguanine—and the
two organic mercury compounds—phenyl
mercuric acetate and methyl mercuric (II)
chloride—showed similar toxicity profiles
across the 13 cell types and were nearest neigh-
bors in the hierarchical clustering of activities.
Data for a larger set of compounds with struc-
turally related series are required to develop
robust structural models of cytotoxicity or
differential toxicity between cell types.
Kinetic signatures of toxic compounds. To
further characterize selected compounds iden-
tified in the end point cytotoxicity assay, cells
exposed to selected compounds were dynami-
cally monitored over 70 hr using real-time
electrical impedance as a measure of viable
cell number (Xing et al. 2005). Using this
methodology, five well-characterized cyto-
toxic compounds were found to have differ-
ent response kinetics. Over the concentration
range tested, digitonin, potassium dichro-
mate, doxorubicin, and tamoxifen were cyto-
toxic in all 13 cell types, and cycloheximide
was cytotoxic in all but two cell types
(MRC-5 and SH-SY5Y). However, time-
course measurement of cytotoxicity in
HepG2 cells, where all five compounds were
cytotoxic, demonstrated distinct kinetic pro-
files that fit into three categories: acute (< 1 hr
to full toxicity), subacute (1–40 hr), and long
term (> 40 hr). Digitonin (Figure 5A), a mild
detergent (Schulz 1990), and tamoxifen
(Figure 5E), a Ca2+influx stimulator, were
fully cytotoxic to HepG2 cells after only
10 min of exposure. Potassium dichromate
(Figure 5B) and doxorubicin (Figure 5D),
both DNA-damaging agents (De Flora et al.
1990; Yang and Wang 1999), demonstrated a
slower onset of activity and induced complete
cytotoxicity only after 35 hr of exposure.
Finally, cycloheximide (Figure 5C), a protein
synthesis blocker (Liu et al. 1994), inhibited
cell proliferation but did not decrease cell
number below what was present at the time
of addition.
Discussion
More than 80,000 chemical compounds
are registered for use in the United States
(NTP 2004b). In addition, about 2,000 new
compounds are introduced into commercial
use each year that may pose hazards for
human health (NTP 2004b). Traditional toxi-
cologic methods cannot characterize and
define the toxicity of such a large number of
compounds in a cost-efficient and timely
manner. However, recently developed HTS
technologies may help to solve this problem
by identifying classes of compounds with simi-
lar activity profiles and by helping to select
and prioritize which compounds should
receive a comprehensive toxicologic evalua-
tion. As a first step in its HTS initiative, the
NTP selected 1,353 compounds for character-
ization in cell-based cytotoxicity screens. The
choice of compounds for this initial screening
set was based primarily on practical and opera-
tional considerations. For example, the com-
pounds had to be readily commercially
available, of adequate purity and quantity, and
soluble in DMSO up to the maximum stock
concentration of 10 mM. This, and the fact
that the entire universe of compounds of
potential toxicologic interest stretches into the
tens of thousands, means that many important
compounds were not included in this initial
set of 1,353. Now that proof of principle has
been established in this robust and repro-
ducible cell-based cytotoxicity assay, the com-
pound collection is being expanded to include
a greater number and diversity of compounds.
The human and rodent primary cells and
cell lines used in the study were selected to rep-
resent six tissues of interest in toxicology: liver,
blood, kidney, nerve, lung, and skin. Nine dif-
ferent cell types from human and four from
rodents were chosen to gather data on poten-
tially varied organ- and tissue-specific responses
to environmental agents. Because our goal with
this study was to investigate the feasibility of
performing high-throughput assays of toxico-
logic compounds, we conducted our initial
experiments with cell lines commonly used in
qHTS: HepG2, SH-SY5Y, HEK293, Jurkat,
H-4-II-E, and N2a (Hynes et al. 2006;
Knasmuller et al. 2004). These lines are trans-
formed or otherwise adapted to growing
in vitro, so presumably are less representative of
Compound cytotoxicity profiling using qHTS
Environmental Health Perspectives • VOLUME 116 | NUMBER 3 | March 2008
289
Figure 5. Kinetics of cytotoxicity responses for digitonin (A), potassium dichromate (B), cycloheximide (C),
doxorubicin (D), and tamoxifen (E) in HepG2 cells monitored by the RT-CES system. 10,000 cells/well were
plated in 16-well strips for the RT-CES cytotoxicity assay. ↓, time of compound addition. Different com-
pound concentrations are indicated by different colors. Data are normalized to the time of compound addi-
tion at 22–24 hr of cell culture.
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Cell index
Time (hr)
Cell index
Cell index Cell index
Cell index
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
010 20 304050 607080010 20 30 4050 6070 80
Time (hr)
Time (hr)
0 1020 30 40
Time (hr)
5060 7080 90100010 2030 405060 7080 90 100
0 1020 304050 60 70
Time (hr)
Digitonin Potassium dichromate
Doxorubicin Cycloheximide
Tamoxifen
AB
CD
E
100 μM
33.3 μM
11.1 μM
3.7 μM
1.2 μM
0.41 μM
0.14 μM
Basal

Page 7

in vivo responses than primary cells, but they
served to establish the qHTS methods used
here and provided a baseline against which
future studies with primary cells will be com-
pared. In addition, several types of nontrans-
formed and primary cells were used in this
study, including human umbilical vein
endothelial capillary cells (HUVECs) (Hoshi
and Mckeehan 1984), MRC-5 normal human
fetal lung fibroblasts (Jacobs et al. 1970), BJ
normal human foreskin fibroblasts (Steinert
et al. 2000), human kidney glomerular mesan-
gial cells, and primary rat renal proximal
tubule cells. The ability to use these cells in
these qHTS assays suggests that future efforts
in toxicologic profiling using primary cells will
be feasible.
Reproducibility of qHTS data has been
demonstrated previously (Inglese et al. 2006),
but given the particular importance of
reliability in toxicity profiling, we performed
additional studies of assay robustness and
reproducibility both within an assay and
between replicates of an assay over time. In
addition we evaluated the concentration–
response curves for the positive controls
included in each plate and found them to be
very consistent. IC50values of active com-
pounds among 55 duplicates included in the
compound library exhibited excellent correla-
tion (R2= 0.71; R2= 0.89 when compounds
with efficacies lower than 30% were removed
from the analysis), demonstrating high intra-
experiment reproducibility. To assess interassay
reproducibility, three independent runs of the
HepG2 qHTS were compared, producing a
good correlation and average R2of IC50values
of 0.74. Taken together, these data demonstrate
that the ATP-based cytotoxicity assay in the
qHTS platform is highly reproducible.
Of the 1,353 unique compounds tested,
only 428 produced a measurable response in
the cytotoxicity assays. Although this activity
rate is higher than the expected rate in high-
throughput screens for drug discovery, these
results represent the prevalence of cytotoxicity
at 40 hr of exposure in these cell types, meas-
ured with this readout, under these condi-
tions at these concentrations, in the absence
of metabolism. Many of the 1,353 com-
pounds tested have been associated only with
more chronic or subtle toxicity in vivo, so
only a small subset of the compounds might
reasonably be expected to be positive in this
measure of acute cell killing.
Within the subgroup of 428 cytotoxic
compounds, we were able to identify multiple
patterns of effects within and across com-
pound types, cell types, and species. Some
compounds (e.g., digitonin, phenyl mercuric
acetate) were uniformly toxic across all cell
types, whereas others showed selective toxicity
(e.g., 2-methyl-1-nitroanthraquinone in
HEK293 cells). To the degree that structurally
related compounds were present in this lim-
ited collection, nascent structure–activity rela-
tionships could sometimes be detected (e.g.,
the organic mercurials). These results suggest
that a full range of similarities and differences
in compound effects are potentially detectable
using cell-based qHTS.
A range of cytotoxicity response patterns
was also seen among cell types. Overall, the
human blood- (Jurkat) and neuron-derived
(SH-SY5Y) cells and rodent cells (N2a,
H-4-II-E and NIH 3T3) were most sensitive
to compound-induced cytotoxicity; kidney-
derived cells (HEK293, human mesangial,
and rat primary proximal tubule) were inter-
mediate in sensitivity; and human fibroblastic,
endothelial, and skin cells (HUV-EC-C, BJ,
and MRC-5) were least sensitive. However,
there was no one “sentinel” cell type sensitive
to all toxins, which could be used to triage
compounds for screening against additional
cell types and assays.
A striking finding from the current study
is the lack of similarity in the patterns of com-
pound activity in cells derived from the same
tissue but from different species (e.g., human
HepG2 and rat H-4-II-E hepatoma cells).
Even cells of similar tissue origin from the
same species sometimes showed considerable
differences in compound activity profiles. For
example, the human SK-N-SH and SH-SY5Y
lines showed quite different response patterns
across the 1,353 compounds, even though
SK-N-SH is the parental line of the SH-SY5Y.
Overall, SH-SY5Y cells were more sensitive to
compound-induced toxicity than SK-N-SH
cells; the former cells are more differentiated
and express a variety of neurotransmitters and
neuronal cell surface markers. Neuronal cells
are sensitive to environmental insults, perhaps
because of their high metabolic demands and
physiologic/morphologic specialization
(Jellinger 2006). In general, rodent cells were
more sensitive than the human cells in this
study, and in each case of homologous cells
from the two species, rodent cells were the
more sensitive cell type. Cell doubling times
did not correlate with sensitivity (as measured
by the number of positive compounds), but
future studies will investigate other possible
mechanisms for the differential toxicities
observed. In all, these studies demonstrate that
in vitro cytotoxicity is often cell-type specific
and that cytotoxicity in one cell type does not
necessarily predict cytotoxicity in another.
One important limitation of the assay used
here is that cytotoxicity was measured at a sin-
gle time point (40 hr) only. To address this
limitation, dynamic measurements of cellular
response to exposure over time were made with
selected compounds and the time-courses of
cytotoxicity correlated with known molecular
and cellular mechanisms of compound activity.
Although this system is low throughput, staged
screening of all compounds in the end point
cytotoxicity assay followed by selective testing
in the dynamic system may allow inference of
mechanism of cytotoxicity.
In this study we have shown that it is feasi-
ble to screen large numbers of compounds in a
titration-based HTS format and generate
robust and reproducible results that can be
analyzed to detect and compare cytotoxicity of
a large number of compounds rapidly in a vari-
ety of cell types. To facilitate the use of these
data by others, we have deposited all data from
this study into a public database (PubChem
2007b) in advance of publication, and we will
continue to do so as more compounds, condi-
tions, cell types, and assay readouts are tested.
Large qHTS data sets promise to provide a rich
source of information for the development of
in vitro toxicologic profiles that may prove
valuable for prioritizing compounds for more
intensive toxicologic investigation, and ulti-
mately, predicting in vivo toxicity.
REFERENCES
Biswas S J, Khuda-Bukhsh AR. 2005. Cytotoxic and genotoxic
effects of the azo-dye-p-dimethylaminoazobenzene in
mice: a time-course study. Mutat Res 587:1–8.
Chhabra RS, Bucher JR, Wolfe M, Portier C. 2003. Toxicity
characterization of environmental chemicals by the US
National Toxicology Program: an overview. Int J Hyg
Environ Health 206:437–45.
Cleveland PH, Koutz PJ. 2005. Nanoliter dispensing for uHTS
using pin tools. Assay Drug Dev Technol 3:213–25.
Crouch A, Kozlowski R, Slater K, Fletcher J. 1993. The use of ATP
bioluminescence as a measure of proliferation and cyto-
toxicity. J Immunol Methods 160:81–88.
De Flora S, Bagnasco M, Serra D, Zanacchi P. 1990. Geno-
toxicity of chromium compounds: a review. Mutat Res
238:99–172.
Dix DJ, Houck KA, Martin MT, Richard AM, Setzer RW, Kavlock
RJ. 2007. The ToxCast program for prioritizing toxicity test-
ing of environmental chemicals. Toxicol Sci 95:5–12.
ECVAM (European Center for the Validation of Alternative
Methods). 2007. Home Page. Available: http://ecvam.jrc.it/
index.htm [accessed 1 May 2007].
Hengstler JG, Van Der Burg B, Steinberg P, Oesch F. 1999.
Interspecies differences in cancer susceptibility. Drug
Metab Rev 31:917–970.
Hill AV. 1910. The possible effects of the aggregation of the
molecules of haemoglobin on its dissociation curves.
J Physiol (London) 40:4–7.
Hoshi H, McKeehan WL.1984. Brain- and liver cell-derived fac-
tors are required for growth of human endothelial cells in
serum-free culture. Proc Natl Acad Sci USA 81:6413–6417.
Hynes J, Hill R, Papkovsky DB. 2006. The use of a fluores-
cence-based oxygen uptake assay in the analysis of cyto-
toxicity. Toxicol in Vitro 20:785–792.
ICCVAM (Interagency Coordinating Committee on the Validation
of Alternative Methods). 2007. Home Page. Available: http://
iccvam.niehs.nih.gov [accessed 1 May 2007].
Inglese J, Auld, DS, Jadhav A, Johnson RL, Simeonov A, YasgarA,
et al. 2006. Quantitative high-throughput screening: a
titration-based approach that efficiently identifies biological
activities in large chemical libraries. Proc Natl Acad Sci USA
103:11473–11478.
Jacobs JP, Jones CM, Bailie JP. 1970. Characteristics of a
human diploid cell designated MRC-5. Nature 227:168–170.
Jellinger KA. 2006. Challenges in neuronal apoptosis. Curr
Alzheimer Res 3:377–391.
Klaunig JE, Babich MA, Baetcke KP, Cook JC, Corton JC,
David RM, et al. 2003. PPAR-alpha agonist-induced rodent
tumors: modes of action and human relevance. Crit Rev
Toxicol 33:655–780.
Knasmuller S, Mersch-Sundermann V, Kevekordes S, Darroudi F,
Huber WW, Hoelzl C, et al. 2004. Use of human-derived liver
Xia et al.
290
VOLUME 116 | NUMBER 3 | March 2008 • Environmental Health Perspectives

Page 8

Compound cytotoxicity profiling using qHTS
Environmental Health Perspectives • VOLUME 116 | NUMBER 3 | March 2008
291
cell lines for the detection of environmental and dietary
genotoxicants; current state of knowledge. Toxicology
198:315–328.
Liu J, Li H, Tribolet ND, Jaufeerally R, Hamou MF, Meir Van EG.
1994. IL-6 stimulates growth and inhibits constitutive pro-
tein synthesis-independent apoptosis of murine B-cell
hybridoma 7TD1. Cell Immunol 155:428–435.
McAteer JA, Davis JM. 2002. Basic cell culture technique and
the maintenance of cell lines. In: Basic Cell Culture (Davis
JM, ed). New York:Oxford, 125–189.
National Research Council. 2007. Toxicity Testing in the Twenty-
first Century: A Vision and a Strategy. Washington, DC:
National Academies Press.
NTP. 2004a. A National Toxicology Program for the 21st Century:
A Roadmap to Achieve the NTP Vision. Research Triangle
Park, NC:National Toxicology Program, National Institute of
Environmental Health Sciences.
NTP. 2004b. Current Directions and Evolving Strategies.
Research Triangle Park, NC:National Toxicology Program,
National Institute of Environmental Health Sciences.
Available: http://ntp.niehs.nih.gov/ntp/Main_Pages/PUBS/
2004CurrentDirections.pdf [accessed 1 August 2007].
PubChem. 2007a. PubMed Home Page. Available: http://www.
ncbi.nlm.nih.gov/sites/ entrez?db=pcsubstance&term=
NTPHTS_NCGC [accessed 1 November 2007].
PubChem 2007b. PubMed Home Page. Available: http://www.
ncbi.nlm.nih.gov/sites/ entrez?db=pcassay, search term
“NCGC [sourcename] AND Viability [AssayName]”
[accessed 1 November 2007].
Schulz I. 1990. Permeabilizing cells: some methods and applica-
tions for the study of intracellular processes. Methods
Enzymol 192:280–300.
Smith CS, Bucher J, Dearry A, Portier C, Tice R, Witt K, et al.
2007. Chemical selection for NTP’s high throughput screen-
ing initiative [Abstract]. Toxicologist 46:247.
Steinert S, Shay JW, Wright WE. 2000. Transient expression of
human telomerase extends the life span of normal human
fibroblasts. Biochem Biophys Res Comm 273:1095–1098.
Sweadner KJ. 1989. Isozymes of Na+/K+-ATPase. Biochem
Biophys Acta 989:185–220.
Tice RR, Fostel J, Smith CS, Witt K, Freedman JH, Portier CJ,
et al. 2007. The National Toxicology Program high through-
put screening initiative: current status and future directions
[Abstract]. Toxicologist 46:246.
Tweats DJ, Scott AD, Westmoreland C, Carmichael PL. 2007.
Determination of genetic toxicity and potential carcino-
genicity in vitro - challenges post the Seventh Amendment
to the European Cosmetics Directive. Mutagenesis 22:5–13.
Xing JZ, Zheu L, Jackson J A, Gabos S, Sun XJ, Wang XB, et al.
2005. Dynamic monitoring of cytotoxicity on microelectronic
sensors. Chem Res Toxicol 18:154–161.
Yang XL, Wang AH. 1999. Structural studies of atom-specific anti-
cancer drugs acing on DNA. Pharmacol Ther 83:181–215.
Zhang JH, Chung TD, Oldenburg KR. 1999. A simple statistical
parameter for use in evaluation and validation of high
throughput screening assays. J Biomol Screen 4(2):67–73.