The genotoxicity of 3-nitrobenzanthrone and the nitropyrene lactones in human lymphoblasts

Article abstract of DOI:10.1016/S1383-5718(00)00135-2

Polycyclic aromatic hydrocarbons (PAH) and nitrated polycyclic aromatic compounds (nitro-PAC) have been found to be mutagenic in bacterial and human cells as well as carcinogenic in rodents. In this investigation, the genotoxic effects of 3-nitrobenzanthrone (3NB) and a mixture of nitropyrene lactones (NPLs) were determined using forward mutation assays performed in two human B-lymphoblastoid cell lines, MCL-5 and h1A1v2, which are responsive to the nitro-PAC class of compounds. Mutagenicity of the compounds was determined at the heterozygous tk locus and the hemizygous hprt locus, thus, identifying both large-scale loss of heterozygosity (LOH) events as well as intragenic mutagenic events. Genotoxicity was also determined using the CREST modified micronucleus assay, which detects chromosomal loss and breakage events. Results indicate 3NB is an effective human cell mutagen, significantly inducing mutations at the tk and hprt loci in both cell lines, and inducing micronuclei in the h1A1v2 cell line. The NPL isomers are also mutagenic, inducing mutations at the two loci as well as micronuclei in both cell lines. Because of their mutagenic potencies and their presence in ambient air, further assessments should be made of human exposures to these nitro-PAC and the potential health risks involved. (C) 2000 Elsevier Science B.V.

Full text of DOI:10.1016/S1383-5718(00)00135-2

Mutation Research 472 (2000) 93–103
The genotoxicity of 3-nitrobenzanthrone and the nitropyrene
lactones in human lymphoblasts
Patricia T. Phousongphouanga, Andrew J. Grosovskya,b, David A. Eastmonda,b
,
Maricela Covarrubiasa, Janet Areya,c,∗
a
Environmental Toxicology Graduate Program, Air Pollution Research Center,
University of California, Riverside, CA 92521, USA
Department of Cell Biology and Neuroscience, Air Pollution Research Center,
b
University of California, Riverside, CA 92521, USA
Department of Environmental Sciences, Air Pollution Research Center,
c
University of California, Riverside, CA 92521, USA
Received 22 March 2000; received in revised form 8 August 2000; accepted 21 September 2000
Abstract
Polycyclic aromatic hydrocarbons (PAH) and nitrated polycyclic aromatic compounds (nitro-PAC) have been found to
be mutagenic in bacterial and human cells as well as carcinogenic in rodents. In this investigation, the genotoxic effects of
3-nitrobenzanthrone (3NB) and a mixture of nitropyrene lactones (NPLs) were determined using forward mutation assays
performed in two human B-lymphoblastoid cell lines, MCL-5 and h1A1v2, which are responsive to the nitro-PAC class of
compounds. Mutagenicity of the compounds was determined at the heterozygous tk locus and the hemizygous hprt locus,
thus, identifying both large-scale loss of heterozygosity (LOH) events as well as intragenic mutagenic events. Genotoxicity
was also determined using the CREST modified micronucleus assay, which detects chromosomal loss and breakage events.
Results indicate 3NB is an effective human cell mutagen, significantly inducing mutations at the tk and hprt loci in both cell
lines, and inducing micronuclei in the h1A1v2 cell line. The NPL isomers are also mutagenic, inducing mutations at the two
loci as well as micronuclei in both cell lines. Because of their mutagenic potencies and their presence in ambient air, further
assessments should be made of human exposures to these nitro-PAC and the potential health risks involved. © 2000 Elsevier
Science B.V. All rights reserved.
Keywords: 3-Nitrobenzanthrone; Nitropyrene lactones; Nitro-PAC; Human cells; Mutagenicity; Micronucleus
1. Introduction
lactones and nitro-PAH ketones) are carcinogenic and
mutagenic [1–14]. The role of genotoxic nitro-PAH
in the development of lung cancer in humans exposed
to diesel exhaust has been extensively studied but
remains controversial [10,15]. Because the ambient
concentrations of nitro-PAC are controlled by com-
bustion emissions [8,14,15] and atmospheric forma-
tion reactions [7,16–19], it is important to determine
the contribution of both atmospherically formed and
Many nitrated polycyclic aromatic hydrocarbons
(nitro-PAH) and nitrated polycyclic aromatic com-
pounds (nitro-PAC) (including, for example, nitro-PAH
∗ Corresponding author. Tel.: +1-909-787-3502;
fax: +1-909-787-5004.
E-mail address: janet.arey@ucr.edu (J. Arey).
1383-5718/00/$ – see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S1383-5718(00)00135-2

94
P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
directly-emitted nitro-PAC to the mutagenicity of am-
bient air. Improved understanding of the nitro-PAC
contribution to ambient mutagenic activity is essen-
tial in evaluation of the risk associated with human
exposure to ambient atmospheric mutagens.
Recent mutagenicity studies in the human B-lym-
phoblastoid cell lines, h1A1v2 and MCL-5, which
focused on 2-nitronaphthalene (2NN), a two-ring
nitro-PAH, and 2-nitrodibenzopyranone (2NDBP), a
three-ring nitro-PAH lactone, identified some unex-
pected results [6,12,13]. These compounds were mu-
tagenic only at the autosomal, heterozygous tk locus
but not at the X-linked, hemizygous hprt locus. These
data suggest that large-scale loss of heterozygosity
(LOH) mechanisms, available only for heterozygous
sequences, such as tk, were the principal or exclusive
mutagenic pathway for 2NN and 2NDBP. Further-
more, several lines of evidence suggested an important
role for oxidative rather than reductive metabolism in
the mutagenicity of these compounds [6,13]. 2NN and
2NDBP have also recently been shown by our group
[6,12,13] to induce micronuclei in the same cell lines.
The human B-lymphoblastoid cell line MCL-5
has facilitated the investigation of the mutagenicity
of PAH [6,12,20] and nitro-PAC [1,2,6,12,13] since
the cell line has been transfected with two plasmids
encoding multiple cytochrome P450 and epoxide hy-
drolase genes, and also has inducible endogenous
cytochrome P450 1A1 (CYP1A1) [21]. Therefore, ex-
ogenous metabolic activation of PAH is not required.
A related cell line, derived from AHH-1, h1A1v2,
which is transfected with a single cytochrome P450
enzyme, CYP1A1 [22], has also been employed
in studies of the genotoxicity of PAH [3,6,23] and
nitro-PAC [2,3,6,13]. The differing transfected enzy-
matic activities in the two cell lines can potentially
result in differing mutagenic metabolites, and there-
fore, differing mutagenic profiles. The h1A1v2 and
MCL-5 cell lines also differ in their p53 status [24].
MCL-5 is wild type for p53, whereas h1A1v2 carries
a heterozygous p53 mutation [24], and exhibits a de-
layed apoptotic response and an increased recovery
of chromosomal scale LOH mutants [25,26].
Fig. 1. Structure and numbered NO2 positions of the nitropyrene
lactones (I). Structure of 3-nitrobenzanthrone (II).
sen for this study are potent bacterial mutagens that
were identified through the use of bioassay-directed
chemical analysis of ambient particles [5,7], diesel
exhaust [5] and atmospheric chamber reactions [11].
The nitropyrene lactones (NPLs; Fig. 1(I)), the 4-ring
analogues of 2NDBP, were identified in environmen-
tal chamber simulations of atmospheric reactions of
pyrene as mutagenic reaction products [11] and have
also been reported in ambient air, although the specific
isomers involved have not yet been isolated [7,11].
Also chosen for study was 3-nitrobenzanthrone (3NB;
Fig. 1(II)), identified in diesel exhaust and ambient
air particles and representing a new chemical class of
mutagenic nitro-PAC, a nitro-PAH ketone [5].
Researchers employing the h1A1v2 cell line have
reported that urban airborne particles contain hu-
man cell mutagens, including unsubstituted PAH and,
in the more polar liquid chromatographic fractions,
nitro-PAH and PAH ketones [27,28]. However, a large
portion of the mutagenicity attributed to semipo-
lar and polar compounds remains unassigned [28].
Because of their strong bacterial mutagenicity and
their presence in ambient air, the NPLs and 3NB are
candidate contributors to the unassigned human cell
mutagenicity of ambient particles. Here, the geno-
toxicity of 3NB and the NPLs was evaluated in two
human cell lines using mutagenicity assays at tk and
hprt, as well as the CREST modified micronucleus
assay as an indicator of clastogenic or aneugenic
properties [29,30]. The results suggest that these com-
pounds induce a broader range of genotoxic effects
than previously observed for 2NN and 2NDBP and
their contribution to the human cell mutagenicity of
ambient particles should be further assessed.
The studies presented here were designed to eval-
uate whether the findings for 2NN and 2NDBP of
human cell mutagenicity at the heterozygous tk locus
but not at the hemizygous hprt locus could be gener-
alized to include other nitro-PAC. The nitro-PAC cho-

P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
95
2. Materials and methods
chromatographic peaks had area percents (in order of
their elution on a DB-1701 capillary column) of 7.0,
46.3, 27.5, 0.7, 17.4 and 1.1. The NPLs were identi-
fied by their characteristic mass spectra, which were
similar to those reported for the nitrophenanthrene
lactone isomers, including 2NDBP [34]. All the NPL
isomers had a strong molecular ion peak at m/z =
265, a base peak at m/z = 163 which corresponds to
the loss of C2NO4 (−102), and additional fragment
ions from the loss of CNO2 (−58) and CNO3 (−74).
3NB was synthesized by reacting benzanthrone
(Bz) with N2O4 in CH2Cl2 at room temperature [33].
This reaction formed 3NB as the major mononitro-Bz
isomer, along with one minor mononitro-Bz isomer
and several dinitro-Bz isomers. The 3NB was sep-
arated from the mononitro- and dinitro-Bz isomers
by flash chromatography, and the fractions collected
were analyzed by GC–MS. The fraction containing
the desired 3NB was then further purified by recrys-
tallization and quantified by GC–FID (>99% purity).
3NB was conclusively identified by HH-COSY and
1H-NMR (CDCl3): δ = 7.69 (dt, J = 7.89, 1H), 7.84
(dt, J = 8.63, 2.01, 1H), 8.02 (dt, J = 7.73, 1.32,
1H), 8.38 (d, J = 8.15, 1H), 8.43 (d, J = 8.09, 1H),
8.52–8.57 (m, 2H), 8.87 (dd, J = 7.35, 1.47, 1H),
8.98 (dd, J = 8.82, 1.47, 1H), which was in general
agreement with a previously reported spectra [35].
2.1. Chemicals
Benzo[a]pyrene (BaP) [50-32-8] (98% stated
purity), dimethyl sulfoxide [67-68-5] (99.5% stated
purity), and melphalan [148-82-3] (95% stated pu-
rity) were purchased from Sigma (St. Louis, MO).
Benzanthrone [82-05-3] (technical grade) was pur-
chased from Aldrich (Milwaukee, WI), and pyrene
lactone (>99% purity) was synthesized as described
elsewhere [31].
2.2. Synthesis and purification of
3-nitrobenzanthrone and the nitropyrene lactones
Two convenient methods of synthesizing nitro-PAC
are the nitration of a parent compound either with
dinitrogen pentoxide (N2O5) in carbon tetrachloride
solution at room temperature [32] or with dinitrogen
tetroxide (N2O4) in methylene chloride (CH2Cl2)
at room temperature [33]. The N2O5 reaction often
produces isomer profiles different from those formed
in the N2O4 reaction. In aprotic solution, the N2O5
undergoes dissociation to NO3 and NO2, allowing
nitration by way of a radical reaction mechanism,
whereas the N2O4 reaction in CH2Cl2 is believed to
proceed via electrophilic nitration.
The synthesis of the NPLs was accomplished by
reaction of the parent compound, pyrene lactone (PL),
with both N2O4 and N2O5. Each reaction formed
multiple mononitro-PL and dinitro-PL isomers. The
NPLs were separated from the dinitro-PL isomers
by flash chromatography and purified by recrystal-
lization. Because, the specific NPL isomers found in
ambient air extracts [7] and as pyrene atmospheric
reaction products [11] are not yet known, rather than
separate the mononitro-derivatives and test each in-
dividually, it was decided to assay the NPLs as a
mixture. By mixing aliquots from the two nitrations
after removal of the dinitro-PL isomers, a mixture was
produced that contained the mononitro-PL isomers
expected to be formed by electrophilic nitration and
thus representative of NPLs emitted in combustion
sources, plus additional isomers formed by a radical
reaction mechanism, as often occurs in atmospheric
reactions [16]. The NPLs purity as determined by
gas chromatography was >99% and the six NPL
2.3. Cell cultures and genotoxicity assays
2.3.1. Derivation of the cell lines
AHH-1 is a clonal isolate derived from RPMI 1788
cells (a human B-lymphoblastoid cell line obtained
from a healthy male donor), selected for sensitivity
to BaP. The h1A1v2 cell line was derived by fur-
ther transfecting AHH-1 with a plasmid encoding
additional inducible CYP1A1 activity [22]. The L3
variant of the AHH-1 cell line was selected in benzo-
[ghi]perylene for elevated levels of endogenous
CYP1A1 activity and has a lower spontaneous mutant
fraction at the tk locus [36]. MCL-5 was derived from
L3 by transfection with plasmids encoding epoxide
hydrolase as well as CYP1A2, CYP2A6, CYP2E1,
and CYP3A4 genes [21].
2.3.2. Maintenance and growth of cell cultures
The lymphoblastoid cell lines h1A1v2 and MCL-5
were maintained and grown as recently described

96
P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
[25] in suspension cultures in Gentest RPMI 1640
media (without histidine and with 2 mM histidi-
nol required for plasmid selection) supplemented
with 9% horse serum, l-glutamine, 100 ␮g/ml peni-
cillin and 100 ␮g/ml streptomycin sulfate. Cells
were grown to a density of 106 cells/ml and subcul-
tured to 2.5 × 105 cells/ml every other day. At this
time, MCL-5 cells were further supplemented with
100 ␮g/ml hygromycin B for plasmid selection.
5 × 105 cells/ml daily during the phenotypic expres-
sion period required for selection of TK−/− and
HPRT− mutants.
2.3.5. Selection of TK−/− and HPRT− mutants
The cells were allowed a phenotypic expression
period of 5 days prior to selection of TK−/− mutants
and 7 days prior to selection of HPRT− mutants as
determined previously by Sasaki et al. [6,12]. Se-
lection was performed by preparing 96-well plates
at a cell density of 105 cells/ml (20,000 cells/well,
4 plates/culture) in media containing 4 ␮g/ml trifluo-
rothymidine and 9% heat-inactivated horse serum for
TK−/− mutants, and a cell density of 5 × 104 cell/ml
(10,000 cells/well, 4 plates/culture) in media contain-
ing 0.6 ␮g/ml 6-thioguanine and 9% horse serum
not heat-inactivated for HPRT− mutants. Cloning
efficiency plates were prepared in parallel at a cell
density of 10 cells/ml (2 cells/well, 2 plates/culture)
in non-selective media. Selection plates and cloning
efficiency plates were scored 14 and 12 days after
seeding for TK−/− and HPRT− mutants, respectively.
2.3.3. Chemical exposure
Prior to dosing the cells with the test chemicals,
the cells were maintained for 3 days with HAT
(containing 2 × 10−4 M hypoxanthine, 8 × 10−7 M
aminopterin and 3.5 × 10−5 M thymidine) to elim-
inate any pre-existing mutants. On day 3, the cells
were centrifuged out of HAT media, and resuspended
in fresh media containing 2 × 10−4 M hypoxanthine
and 3.5 × 10−5 M thymidine.
Dosing solutions of the compounds were made up
in dimethyl sulfoxide (DMSO) at 1000× the final de-
sired concentration of the compound in cell culture.
Approximately 3 × 107 cells were centrifuged out of
maintenance media and resuspended in 40 ␮l of Cell-
gro RPMI 1640 media containing histidine without
horse serum. Following a protocol previously evalu-
ated using BaP, the chemical exposure was performed
in the absence of serum to maximize the bioavail-
ability of the test compound [6,12]. The cell culture
exposure was then limited to 10 h at 37◦C to minimize
effects due to growth in serum-free conditions [6,12].
To enhance the solubility of test chemical in media,
an aliquot of 120 ␮l DMSO was added to each cell
culture prior to the addition of the 40 ml aliquot of test
chemical, resulting in 0.4% DMSO in all treatments
[6,12]. BaP (15 ng/ml, determined to be the minimum
mutagenic concentration in h1A1v2 [3]) was used as
the positive control, and DMSO (0.4%) was used as
the negative control. At the completion of the expo-
sure period, the cells were centrifuged out of chemical
and resuspended in fresh maintenance media.
2.3.6. Preparation of cell cultures for the CREST
modified micronucleus assay
The same cell lines and chemical exposure param-
eters used in the mutagenicity assay were employed
for the CREST modified micronucleus assay. The
CREST modified micronucleus assay employs the
CREST stain, an antikinetochore antibody obtained
from patients with the autoimmune disease sclero-
derma CREST (calcinosis, Raynaud’s phenomenon,
esophageal dysmotility, sclerodactyly and telangiec-
tasia) syndrome [37–39]. Melphalan (4 ␮M) was used
as the positive control, and DMSO (0.4%) was used
as the negative control. After the 10 h chemical ex-
posure period, ∼106 cells were resuspended in fresh
media with horse serum and 3 ␮g/ml cytochalasin
B was added to each culture. After further incuba-
tion of the cells for 36 h at 37◦C, allowing for cell
recovery and the maximum number of cells to go
through one cell division, the cells were harvested
by cytocentrifugation. The slides were fixed in 100%
methanol for 10 min and allowed to air dry. The slides
were then stored in a nitrogen atmosphere at −20◦C
until later use. The CREST staining and scoring
procedures were performed as previously described
[29,30].
2.3.4. Clonogenic survival assay
At the completion of the dosing period, survival
plates were prepared at 10 cells/ml (2 cells/well,
2 plates/culture) and scored for clonogenic survival
after 14 days. The remaining cells were grown
to a density of 106 cells/ml and subcultured to

P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
97
2.4. Statistical methods
its apoptotic deficient status [24].
The results of the mutagenicity assays indicate
a general dose-dependent increase in mutation fre-
quency for 3NB (Fig. 2) and the NPLs (Fig. 3), in
each cell line at both the tk and hprt loci. A sig-
nificant increase in the TK−/− mutant fraction was
induced by 3NB in h1A1v2 (Fig. 2A, P = 0.0085)
and MCL-5 (Fig. 2B, P = 0.0001). 3NB also in-
duced a significant increase in the HPRT− mutant
fraction in h1A1v2 (Fig. 2A, P = 0.0001) and
MCL-5 (Fig. 2B, P = 0.0001). The NPLs induced a
significant increase in the TK−/− mutant fraction in
h1A1v2 (Fig. 3A, P = 0.0383) and MCL-5 (Fig. 3B,
P = 0.0058). The NPLs also induced a significant in-
crease in the HPRT− mutant fraction in h1A1v2 (Fig.
3A, P = 0.0001) and MCL-5 (Fig. 3B, P = 0.0027).
BaP induced a significant increase in the TK−/−
mutant fraction in MCL-5 (P = 0.05) but not in
h1A1v2 (P = 0.147). BaP also induced a significant
increase in the HPRT− mutant fraction in MCL-5
(P = 0.0022) but not in h1A1v2 (P = 0.353). Our
results are in general agreement with previously de-
termined mutation frequencies for BaP using the same
protocol [12,13].
The program StatView SE + Graphics was used to
calculate the statistical values for the tk and hprt muta-
genicity assays. A test for linear trend was performed
using a 95% confidence interval.
The dose-related increases in the induction of
micronuclei were tested for significance using the
Cochran–Armitage test for trend in binomial propor-
tions. Following the observation of a significant trend,
a one-tailed Fisher exact test or Chi-square test was
used as a post hoc test to determine the significance
of each test concentration.
3. Results
3.1. Mutagenicity of 3-nitrobenzanthrone and the
nitropyrene lactones at the tk and hprt loci
Clonogenic cytotoxicity was determined in parallel
with mutagenicity and is presented in Table 1. In both
cell lines, 3NB is more cytotoxic than the mixture of
NPLs tested, and BaP, the positive control, was not
toxic at the dose tested. At the highest dose tested,
3NB exposure resulted in 46 and 7% cell survival
in h1A1v2 and MCL-5 cells, respectively, compared
to 80 and 37% cell survival for the corresponding
NPLs exposure. The h1A1v2 cell line sustained a
lower amount of cytotoxicity, which may be due to
A comparison of the mutagenicity of 3NB and the
NPLs within each cell line (Table 2) reveals a three-
to four-fold higher mutation frequency at tk than hprt
in the h1A1v2 cell line (Figs. 2A and 3A), although
comparable levels of mutagenicity were observed at tk
and hprt in the MCL-5 cell line (Figs. 2B and 3B). The
Table 1
Fraction of cell survivala,b after treatment with 3-nitrobenzanthrone and the nitropyrene lactones
Compound
Concentration (␮M)
Fraction of cell survival ꢀ S.E.M.
h1A1v2
MCL-5
DMSOc
–
1.00
1.00
3-Nitrobenzanthrone
0.364
1.82
3.64
0.92 ꢀ 0.12
0.60 ꢀ 0.09
0.46 ꢀ 0.07
0.61 ꢀ 0.12
0.21 ꢀ 0.04
0.07 ꢀ 0.05
Nitropyrene lactones
Benzo[a]pyrened
0.377
1.89
3.77
1.20 ꢀ 0.18
0.92 ꢀ 0.15
0.80 ꢀ 0.10
1.08 ꢀ 0.09
1.06 ꢀ 0.17
0.66 ꢀ 0.11
0.37 ꢀ 0.06
1.23 ꢀ 0.12
0.060
a Fraction of cell survival was calculated with data from three independent experiments.
b Fraction of cell survival is given ꢀ S.E.M.
c Negative control (0.4% DMSO; clonogenic survival defined as 1.00).
d Positive control.

98
P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
Fig. 2. TK−/− and HPRT− mutation frequency per 106 viable cells (MF × 10−6) induced in (A) h1A1v2 and (B) MCL-5 cells treated
with 3-nitrobenzanthrone. The mutation frequency was calculated with data from three independent experiments.
increased mutagenicity at tk in h1A1v2 cells may be
due to the p53-deficient background in these cells [24]
and a resultant increased recovery of recombinational
LOH events [25,26].
Comparing the results between the two cell lines,
the recovery of induced TK−/− mutants in h1A1v2
was consistently higher than the recovery of induced
TK−/− mutants in MCL-5 cells exposed to equal con-
centrations of mutagen (Table 2). In contrast, for both
3NB and the NPLs, the recovery of induced HPRT−
mutants was greater in MCL-5 than in h1A1v2
(Table 2), perhaps reflecting the production of addi-
tional mutagenic metabolites in MCL-5 associated
with its broader metabolic capacity.
3.2. Micronuclei induced by 3-nitrobenzanthrone
and the nitropyrene lactones
Table 3 lists the nuclear division index (NDI), which
is indicative of cell growth or cell division, for h1A1v2
and MCL-5 cells after treatment with 3NB and the
NPLs. Treatment of both cell lines with 3NB resulted
in a linear decrease in NDI, indicating that 3NB in-
hibits cell growth in both cell lines. Treatment with
Fig. 3. TK−/− and HPRT− mutation frequency per 106 viable cells (MF × 10−6) induced in (A) h1A1v2 and (B) MCL-5 cells treated
with the nitropyrene lactones. The mutation frequency was calculated with data from three independent experiments.

P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
99
Table 2
Table 3
Induced mutation frequencya at tk and hprt at the highest concen-
Nuclear division indexes (NDI)a,b for h1A1v2 and MCL-5 cells
after treatment with 3-nitrobenzanthrone and the nitropyrene lac-
tones
tration tested
Compound
Concentra- h1A1v2
MCL-5
tion (␮M)
Test
compound
Concentra- NDI ꢀ S.E.M.
tion (␮M)
tk
hprt tk hprt
h1A1v2
MCL-5
3-Nitrobenzanthrone
Nitropyrene lactones
2-Nitrodibenzopyranoneb
2-Nitronaphthaleneb
3.64
3.77
4.15
156.1 35.6 62.6 86.7
37.6 14.3 27.1 46.4
3-Nitrobenzanthrone
0
1.97 ꢀ 0.13 1.61 ꢀ 0.04
1.88 ꢀ 0.05 1.54 ꢀ 0.15
1.78 ꢀ 0.02 1.47 ꢀ 0.07
1.66 ꢀ 0.06 1.36 ꢀ 0.11
1.47 ꢀ 0.06 1.31 ꢀ 0.07
1.31 ꢀ 0.07 1.18 ꢀ 0.05
1.96 ꢀ 0.03 1.69 ꢀ 0.05
2.11 ꢀ 0.08 1.58 ꢀ 0.06
1.97 ꢀ 0.04 1.61 ꢀ 0.04
1.92 ꢀ 0.09 1.63 ꢀ 0.02
16.7
96.7
0
0
7.2
83.6
0
0
0.364
1.82
3.64
5.45
7.27
462
a Induced mutation frequency is defined as the number of
mutants per 106 viable cells minus the background mutants per
106 viable cells.
b Data from Grosovsky et al. [6].
Nitropyrene lactones
Melphalanc
0
0.377
1.89
3.77
the NPLs, on the other hand, did not affect the NDI in
either cell line, indicating that the NPLs did not cause
growth inhibition in the cells at the doses tested. Treat-
ment with melphalan, the positive control, resulted
in substantial decreases in NDI at the dose tested in
h1A1v2 and MCL-5 cells.
The data in Figs. 4 and 5 represent the percent of
total micronuclei (total MN) induced by the expo-
sure of h1A1v2 and MCL-5 cells to 3NB (Fig. 4)
and the NPLs (Fig. 5). Of the total MN, MN− indi-
cates the number due to chromosomal breakage and
MN+ indicates the number due to chromosomal loss.
A significant dose-related increase was observed in
total MN in h1A1v2 (Fig. 4A) after treatment with
4.00
1.13 ꢀ 0.02 1.12 ꢀ 0.03
a NDI = {[N1+(2×N2)+(3×N3)+(4×N4)]/(N1+N2+
N3 + N4)} where N1–N4 represent the number of cells with 1–4
nuclei, respectively [29], and was calculated with data from three
independent experiments.
b NDI values are given ꢀ S.E.M.
c Positive control. NDI was calculated with data from six in-
dependent experiments.
3NB, but only a weak significant increase was ob-
served in MCL-5 (Fig. 4B). At the highest dose tested
in h1A1v2, 3NB induced a significant amount of both
chromosomal loss (MN+) and chromosomal breakage
Fig. 4. Percent micronuclei (% MN) induced in (A) h1A1v2 and (B) MCL-5 cells treated with 3-nitrobenzanthrone. The percent micronuclei
was calculated with data from three independent experiments. a: P < 0.05; b: P < 0.01; c: P < 0.001.

100
P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
Fig. 5. Percent Micronuclei (% MN) induced in (A) h1A1v2 and (B) MCL-5 cells treated with the nitropyrene Lactones. The percent
micronuclei was calculated with data from three independent experiments. a: P < 0.05; b: P < 0.01.
(MN−), although the significant increase in chromo-
somal loss (MN+) may be due to an abnormally low
background frequency compared to historical values.
In MCL-5, 3NB only induced a significant increase
in chromosomal breakage (MN−). Fig. 5 illustrates
micronuclei induced by the NPLs in h1A1v2 and
MCL-5 cells. In both cell lines there was a significant
dose-related increase in total MN, with comparable
induction of total MN at the highest dose tested. Of
the total MN induced, a significant amount were due
to chromosomal loss (MN+) in h1A1v2 (Fig. 5A) and
chromosomal breakage (MN−) in MCL-5 (Fig. 5B).
Melphalan induced a significant amount of total MN
in h1A1v2 (8.90 ꢀ 0.68%, P < 0.0001). This in-
duction was primarily due to chromosomal breakage
(MN−) (6.66 ꢀ 0.59%, P < 0.0001), although a sig-
nificant increase was due to chromosomal loss (MN+)
(2.24ꢀ0.63%, P < 0.0001). Melphalan also induced
a significant amount of total MN in MCL-5 cells
(7.67 ꢀ 0.90%, P < 0.0001). Again chromosomal
breakage (MN−) (6.33 ꢀ 0.95%,P < 0.0001) domi-
nated, but chromosomal loss (MN+) (1.34 ꢀ 0.41%,
P < 0.0001) was also significant.
TA98 and YG1024, comparable to 1,8-dinitropyrene,
one of the most mutagenic nitro-PAH reported to
date [14], and these investigators also reported that
3NB induced micronuclei in mouse peripheral blood
reticulocytes after intraperitoneal injections. Other
mononitro-Bz isomers were found to be less muta-
genic in TA98 [5]. Busby and co-workers evaluating
the mononitro- and dinitro-pyrenes, demonstrated
that potency in a bacterial assay is not necessarily a
good predictor of activity in human cell assays [1].
However, we tested the 3NB isomer in human cells,
because, in addition to its high bacterial mutagenic-
ity, it has been identified as a component of both
diesel exhaust particle extracts and ambient particle
extracts [5].
While two NPL isomers are formed in simulated
atmospheric reactions of pyrene [11], and two NPL
isomers have been reported in analyses of ambient
samples [7], the specific isomers have not yet been
identified. Two specific nitropyrene lactone isomers,
1-NPL and 3-NPL, have been found to be highly mu-
tagenic in the Ames Salmonella assay [4]. By testing
a mixture of six of the eight possible NPL isomers,
it is likely that isomers representative of both com-
bustion emission and atmospheric formation were
screened, allowing us to determine if NPLs are likely
to be important human cell mutagens. Because NPLs
are atmospheric reaction products of the abundant
PAH, pyrene [11], human populations downwind of
4. Discussion
Enya et al. [5] reported 3NB to be highly mu-
tagenic in the Ames Salmonella assay in strains

P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
101
sources of PAH such as vehicle emissions may be
exposed to increasing levels of NPLs.
more mutagenic at tk than hprt in the h1A1v2 cell line
(Table 2, Figs. 2A and 3A), suggesting an increment
of recoverable mutations attributable to large-scale
LOH events. Clearly, although the mutations induced
by 2NN and 2NDBP are restricted to large-scale LOH
events, this cannot be generalized to all nitro-PAC. A
comparison of the induced mutation frequency at tk
and hprt of comparable test concentrations (Table 2)
shows that 3NB is two- to four-fold more potent
than is the mixture of NPLs, and almost an order
of magnitude more potent than 2NDBP. The NPLs
are two times more mutagenic than is the individual
nitrophenanthrene lactone isomer, 2NDBP, at the tk
locus in h1A1v2, and four times more mutagenic in
MCL-5. 2NN (note higher test dose, Table 2) is the
least potent mutagen of the group.
The results of the CREST modified micronucleus
assay show that both compounds are inducers of
micronuclei. The results are more significant in the
h1A1v2 cell line than the MCL-5 cell line for both
compounds. Again this may be attributed to the apop-
totic deficiency in h1A1v2 cells, which may allow
survival of a greater number of damaged cells. No
conclusion can be made regarding which transfected
enzymes are required for metabolizing the nitro-PAC,
since both cell lines were responsive to both test
compounds. Both compounds can be classified as
clastogens as well as aneugens, since both compounds
induced both chromosomal loss and chromosomal
breakage. In the past, it has been found that alky-
lating agents are both clastogens and aneugens [29],
indicating that the tested nitro-PAC may be alkylating
agents.
Results presented here demonstrate that exposure of
3NB and the NPLs to two human B-lymphoblastoid
cell lines resulted in induction of mutations at the tk
and hprt loci. This pattern differs from previous studies
of smaller nitro-PAC such as 2NN, which induced mu-
tations only at the tk locus [12,13]. Thus, unlike 2NN
genotoxicity, which was restricted to the induction of
large-scale LOH events and micronuclei, 3NB and
the NPLs appear to be capable of inducing a broader
range of genetic alterations including smaller-scale
chromosomal changes and intragenic mutations. The
significant genotoxicity of 3NB and the NPLs in hu-
man cell assays verify that human risk from expo-
sure to these atmospheric pollutants should be further
assessed.
Our laboratory has previously shown that in both
MCL-5 cells and h1A1v2 cells 2NN and 2NDBP were
mutagenic at the tk but not the hprt locus, suggesting
that the human cell mutagenicity of these compounds
was exclusively attributable to LOH events not avail-
able at the hemizygous hprt locus [6,13]. In contrast,
3NB and the NPLs were strongly mutagenic at tk and
hprt in both MCL-5 and h1A1v2, suggesting a signif-
icantly different mutational spectrum (Table 2, Figs. 2
and 3).
Previous reports [25,26] have demonstrated that
higher mutation frequencies at tk but not hprt are char-
acteristic of h1A1v2, probably due to the abnormal
apoptotic response associated with a heterozygous
mutation at p53 [24]. The human B-lymphoblastoid
cell line MCL-5 is wild type for p53, whereas AHH-1
and its derivatives, h1A1v2 and h2E1v2, carry a het-
erozygous p53 mutation [24], and exhibit delayed
apoptotic responses and increased spontaneous and
induced mutation frequencies [6,13,25,26]. MCL-5
and h2E1v2 have similar HPRT− mutation frequen-
cies [25], but h2E1v2 has a four- to five-fold higher
TK−/− mutation frequency [25,26]. Molecular analy-
sis of the TK−/− mutants indicates the higher TK−/−
mutation frequency in h2E1v2 can be entirely at-
tributed to the recovery of large-scale LOH [25,26].
This pattern closely resembles the source of increased
TK−/− mutation frequency in p53 deficient WIL2-NS
human lymphoblasts and its derivative, WTK-1, as
compared to related p53 wild type TK6 cells [40–42].
Induced HPRT− mutation frequencies are similar in
the cell lines [41], but induced mutation frequencies
at the heterozygous tk locus are 20- to 50-fold higher
in the WIL2-NS cell lines than in the TK6 cells
[41,42]. These similar patterns in two independent
cell line pairs suggests that the p53 deficiency, and
the resultant abnormal apoptotic response, is respon-
sible for the increased recovery of large-scale LOH
mutants.
Summarized in Table 2 are the induced muta-
tion frequencies for 3NB, the NPLs, and the atmos-
pherically-formed [11] nitro-PAC, 2NDBP and 2NN,
tested previously [6]. In the MCL-5 cell line, both
3NB and the NPLs induced comparable mutation fre-
quencies at hprt and tk (Table 2; Figs. 2B and 3B). In
contrast, 3NB and the NPLs were three- to four-fold

102
P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
[11] J. Sasaki, J. Arey, W.P. Harger, Formation of mutagens from
Acknowledgements
the photooxidations of 2-4-ring PAH, Environ. Sci. Technol.
29 (1995) 1324–1335.
Myra Yuan is thanked for the synthesis of pyrene
lactone, which was required to further synthesize the
NPLs. Leslie Hasegawa and Dr. Maik Schuler are
thanked for their assistance with the CREST modi-
fied micronucleus assay. P.T. Phousongphouang was
partially funded by the University of California Toxic
Substances Research and Teaching Program. M. Co-
varrubias was funded by the California Alliance for
Minority Participation.
[12] J.C. Sasaki, J. Arey, D.A. Eastmond, K.K. Parks, A.J.
Grosovsky, Genotoxicity induced in human lymphoblasts by
atmospheric reaction products of naphthalene and phenan-
threne, Mutat. Res. 393 (1997) 23–35.
[13] J.C. Sasaki, J. Arey, D.A. Eastmond, K.K. Parks, P.T.
Phousongphouang, A.J. Grosovsky, Evidence for oxidative
metabolism in the genotoxicity of the atmospheric reaction
product 2-nitronaphthalene in human lymphoblastoid cell
lines, Mutat. Res. 445 (1999) 113–125.
[14] H. Tokiwa, Y. Ohnishi, Mutagenicity and carcinogenicity of
nitroarenes and their sources in the environment, CRC Crit.
Rev. Toxicol. 17 (1986) 23–60.
[15] Health Effects Institute, Diesel exhaust: a critical analysis of
emissions, exposure, and health effects; a special report of
the institute’s diesel working group. In: Diesel Exhaust: A
Critical Analysis of Emissions, Exposure, and Health Effects;
A Special Report of the Institute’s Diesel Working Group,
Cambridge, MA, 1995, 294 pp.
[16] J. Arey, Atmospheric reactions of PAHs including formation
of nitroarenes, in: A.H. Neilson (Ed.), The Handbook of
Environmental Chemistry: PAHs and Related Compounds,
Vol. 3, Part I, 1998, pp. 347–385.
References
[1] W.F. Busby Jr., B.W. Penman, C.L. Crespi, Human cell
mutagenicity of mono- and dinitropyrenes in metabolically
competent MCL-5 cells, Mutat. Res. 322 (1994) 233–242.
[2] W.F. Busby Jr., H. Smith, C.L. Crespi, B.W. Penman,
A.L. Lafleur, Mutagenicity of the atmospheric transformation
products 2-nitrofluoranthene and 2-nitrodibenzopyranone in
Salmonella and human cell forward mutation assays, Mutat.
Res. 389 (1997) 261–270.
[17] R. Atkinson, J. Arey, Atmospheric chemistry of gas-phase
polycyclic aromatic hydrocarbons: formation of atmospheric
mutagens, Environ. Health Perspect. 102 (1994) 117–126.
[18] P. Ciccioli, A. Cecinato, E. Brancaleoni, M. Frattoni, P.
Zacchei, P. De Castro Vasconcellos, The ubiquitous occu-
rrence of nitro-PAH of photochemical origin in airborne
particles, Ann. Chim. 85 (1995) 455–469.
[19] P. Ciccioli, A. Cecinato, E. Brancaleoni, M. Frattoni, P.
Zacchei, A.H. Miguel, P. De Castro Vasconcellos, Formation
and transport of 2-nitrofluoranthene and 2-nitropyrene of
photochemical origin in the troposphere, J. Geophys. Res.
101 (1996) 19567–19581.
[20] W.F. Busby Jr., H. Smith, C.L. Crespi, B.W. Penman,
Mutagenicity of benzo[a]pyrene and dibenzopyrenes in the
Salmonella typhimurium TM677 and the MCL-5 human cell
forward mutation assays, Mutat. Res. 342 (1995) 9–16.
[21] C.L. Crespi, F.J. Gonzalez, D.T. Steimel, T.R. Turner, H.V.
Gelboin, B.W. Penman, R. Langenbach, A metabolically
competent human cell line expressing five cDNAs
encoding procarcinogen-activating enzymes: application to
mutagenicity testing, Chem. Res. Toxicol. 4 (1991) 566–572.
[22] B.W. Penman, L. Chen, H.V. Gelboin, F.J. Gonzalez,
[3] J.L. Durant, W.F. Busby Jr., A.L. Lafleur, B.W. Penman, C.L.
Crespi, Human cell mutagenicity of oxygenated, nitrated and
unsubstituted polycyclic aromatic hydrocarbons associated
with urban aerosols, Mutat. Res. 371 (1996) 123–157.
[4] K. El-Bayoumy, S.S. Hecht, Mutagenicity of K-region deri-
vatives of 1-nitropyrene; remarkable activity of 1- and
3-nitro-5H-phenanthro[4,5-bcd]pyran-5-one, Mutat. Res. 170
(1986) 31–40.
[5] T. Enya, H. Suzuki, T. Watanabe, T. Hirayama, Y. Hisamatsu,
3-Nitrobenzanthrone, a powerful bacterial mutagen and sus-
pected human carcinogen found in diesel exhaust and airborne
particulates, Environ. Sci. Technol. 31 (1997) 2772–2776.
[6] A.J. Grosovsky, J.C. Sasaki, J. Arey, D.A. Eastmond, K.K.
Parks, R. Atkinson, Evaluation of the potential health effe-
cts of the atmospheric reaction products of polycyclic
aromatic hydrocarbons, Health Effect Institute Research
Report Number 84, March 1999, 27 pp.
[7] D. Helmig, J. Arey, W.P. Harger, R. Atkinson, J. López-
Cancio, Formation of mutagenic nitrodibenzopyranones and
their occurrence in ambient air, Environ. Sci. Technol. 26
(1992) 622–624.
[8] International Agency for Research on Cancer, Diesel and
gasoline engine exhausts and some nitroarenes, in: IARC
Monographs on the Evaluation of the Carcinogenic Risks to
Humans, Vol. 46, IARC, Lyon, France, 1989.
C.L. Crespi, Development of
a human lymphoblastoid
cell line constitutively expressing human CYP1A1 cDNA:
substrate specificity with model substrates and promutagens,
Carcinogenesis 15 (1994) 1931–1937.
[9] H.S. Rosenkranz, R. Mermelstein, Mutagenicity and geno-
toxicity of nitroarenes. All nitro-containing chemicals were
not created equal, Mutat. Res. 114 (1983) 217–267.
[10] H.S. Rosenkranz, Mutagenic nitroarenes, diesel emissions,
particulate-induced mutations and cancer: an essay on
cancer-causation by a moving target, Mutat. Res. 367 (1996)
65–72.
[23] J.L. Durant, A.L. Lafleur, W.F. Busby Jr., L.L. Donhoffner,
B.W. Penman, C.L. Crespi, Mutagenicity of C24H14 PAH in
human cells expressing CYP1A1, Mutat. Res. 446 (1999)
1–14.
[24] S.M. Morris, M.G. Manjanatha, S.D. Shelton, O.E. Domon,
L.J. McGarrity, D.A. Casciano, A mutation in the p53 tumor

P.T. Phousongphouang et al. / Mutation Research 472 (2000) 93–103
103
suppressor gene of AHH-1 tk+/− human lymphoblastoid cells,
Mutat. Res. 356 (1996) 129–134.
[25] K.L. Dobo, C.R. Giver, D.A. Eastmond, H.S. Rumbos, A.J.
Grosovsky, Extensive loss of heterozygosity accounts for
differential mutation rate on chromosome 17q in human
lymphoblasts, Mutagenesis 10 (1995) 53–58.
[26] K.L. Dobo, D.A. Eastmond, A.J. Grosovsky, The influence
of cellular apoptotic capacity on N-nitrosodimethylamine-
induced loss of heterozygosity mutations in human cells,
Carcinogenesis 18 (1997) 1701–1707.
[27] J.L. Durant, A.L. Lafleur, E.F. Plummer, K. Taghizadeh, W.F.
Busby Jr., W.G. Thilly, Human lymphoblast mutagens in
urban airborne particles, Environ. Sci. Technol. 32 (1998)
1894–1906.
[28] M.P. Hannigan, G.R. Cass, B.W. Penman, C.L. Crespi, A.L.
Lafleur, W.F. Busby Jr., W.G. Thilly, B.R.T. Simoneit,
Bioassay-directed chemical analysis of Los Angeles airborne
particulate matter using a human cell mutagenicity assay,
Environ. Sci. Technol. 32 (1998) 3502–3514.
[29] D.A. Eastmond, J.D. Tucker, Identification of aneuploidy-
inducing agents using cytokinesis-blocked human lym-
phocytes and an antikinetochore antibody, Environ. Mol.
Mutagen. 13 (1989) 34–43.
[30] G. Krishna, R. Fiedler, J.C. Theiss, Simultaneous analysis of
chromosome damage and aneuploidy in cytokinesis-blocked
V79 Chinese hamster lung cells using an antikinetochore
antibody, Mutat. Res. 282 (1992) 79–88.
[31] R.G. Gillis, Q.N. Porter, 5-Methoxyphenanthrene-4-carbo-
xylic acid, Aust. J. Chem. 42 (1989) 1007–1010.
[32] B. Zielinska, J. Arey, R. Atkinson, T. Ramdahl, A.M.
Winer, J.N. Pitts Jr., Reaction of dinitrogen pentoxide
with fluoranthene, J. Am. Chem. Soc. 108 (1986) 4126–
4132.
[34] D. Helmig, J. Arey, Analytical chemistry of four nitrodi-
benzopyranone isomers for ambient air analysis, Int. J.
Environ. Anal. Chem. 49 (1992) 207–219.
[35] H. Suzuki, T. Enya, Y. Hisamatsu, Synthesis and charac-
terization of some nitrobenzanthrones: suspected new
mutagens in atmospheric environment, Synthesis (1997)
1273–1276.
[36] R.L. Davies, C.L. Crespi, K. Rudo, T.R. Turner, R.
Langenbach, Development of a human cell line by selection
and drug-metabolizing gene transfection with increased
capacity to activate promutagens, Carcinogenesis 10 (1989)
885–891.
[37] B.R. Brinkley, A. Tousson, M.M. Valdivia, The kinetochore of
mammalian chromosomes: structure and function in normal
mitosis and aneuploidy, in: V.L. Dellarco, P.E. Voytek, A.
Hollaender (Eds.), Aneuploidy: Etiology and Mechanisms,
Plenum Press, New York, 1985, pp. 243–267.
[38] S. Frackowiak, B. Labidi, D. Hernandez-Verdun, M. Bouteille,
Preservation of chromosome integrity during micronucleation
induced by colchicine in PtK1 cells, Chromosoma (Berlin)
94 (1986) 468–474.
[39] B.K. Vig, S.E. Swearngin, Sequence of centromere separation;
kinetochore formation in induced laggards and micronuclei,
Mutagenesis 1 (1986) 461–465.
[40] S.A. Amundson, F. Xia, K. Wolfson, H.L. Liber, Different
cytotoxic and mutagenic responses induced by X-rays in two
human lymphoblastoid cell lines derived from a single donor,
Mutat. Res. 286 (1993) 233–241.
[41] F. Xia, S.A. Amundson, J.A. Nickoloff, H.L. Liber, Different
capacities for recombination in closely related human
lymphoblastoid cell lines with different mutational responses
to X-irradiation, Mol. Cell. Biol. 14 (1994) 5850–5857.
[42] F. Xia, X. Wang, Y.H. Wang, N.M. Tsang, D.W. Yandell,
K.T. Kelsey, H.L. Liber, Altered p53 status correlates with
differences in sensitivity to radiation-induced mutation and
apoptosis in two closely related human lymphoblast lines,
Cancer Res. 55 (1995) 12–15.
[33] F. Radner, Nitration of polycyclic aromatic hydrocarbons
with dinitrogen tetroxide. A simple and selective synthesis
of mononitro derivatives, Acta Chem. Scand. B 37 (1983)
65–67.