(E)-2-hexenal-induced DNA damage and formation of cyclic 1,N2-(1,3- propano)-2'-deoxyguanosine adducts in mammalian cells

Article abstract of DOI:10.1021/tx9600107

(E)-2-Hexenal (hexenal), a natural flavor compound, acts as directly genotoxic agent and forms cyclic 1,N²-propano adducts with deoxyguanosine. Formation of this adduct in isolated DNA and in cells was studied with a modified ³²P-postlabeling procedure including HPLC separation, nuclease P1 enrichment, two-dimensional TLC of adducted nucleotide bisphosphates on PEI- cellulose, and quantification of adduct spots by liquid scintillation counting. Adduct formation with the more reactive crotonaldehyde was included for comparison. Synthesized adducted dG-3'-phosphates served as external standards for identification and quantification. In calf thymus DNA, hexenal (0.2 mM) shows a time dependent formation of adducts, yielding 1.55 pmol/μmol of DNA at 5 h incubation. With crotonaldehyde (0.2 mM) the adduct rate was about 10-fold higher. Hexenal also generated 1,N²-propano-dG adducts in the human lymphoblastoid Namalva cell line (0.2 mM, 1 h, 86 fmol/μmol of DNA) and in primary rat colon mucosa cells (0.4 mM, 30 min, 50 fmol/μmol of DNA). In primary colon mucosa cells from rats and humans, hexenal and crotonaldehyde (0.4 mM, 30 min) induced DNA damage, detected by single cell microgel electrophoresis (comet assay). In primary rat gastric mucosa cells, hexenal was only weakly active, inducing detectable DNA damage in 20% of cells at 0.8 mM concentration. In contrast, primary mucosa cells from rat esophagus were as sensitive as colon cells. After single oral application of hexenal to rats (up to 320 mg/kg body wt) DNA damage was not detectable in gastrointestinal mucosa. Analysis of hexenal in selected flavored foods revealed concentrations up to 14 ppm (0.14 mM) that are comparable to its natural occurrence in some fruits and vegetables (up to 30 ppm). Thus, the concentration range selected for the toxicological studies described here clearly is relevant: Hexenal, at concentrations found in food, exerts genotoxic effects in cells from rat and human gastrointestinal tract.

Full text of DOI:10.1021/tx9600107

Chem. Res. Toxicol. 1996, 9, 1207-1213
1207
(E)-2-Hexen a l-In d u ced DNA Da m a ge a n d F or m a tion of
Cyclic 1,N2-(1,3-P r op a n o)-2′-d eoxygu a n osin e Ad d u cts in
Ma m m a lia n Cells
Petra Go¨lzer,† Christine J anzowski,*,† Beatrice L. Pool-Zobel,‡ and
Gerhard Eisenbrand†
Department of Chemistry, Division of Food Chemistry and Environmental Toxicology, University of
Kaiserslautern, Erwin-Schro¨dinger-Strasse, Building 52, D-67663 Kaiserslautern, Germany, and
Institute of Nutritional Physiology, Federal Research Center for Nutrition, Karlsruhe, Germany
Received J anuary 19, 1996X
(E)-2-Hexenal (hexenal), a natural flavor compound, acts as directly genotoxic agent and
forms cyclic 1,N2-propano adducts with deoxyguanosine. Formation of this adduct in isolated
DNA and in cells was studied with a modified 32P-postlabeling procedure including HPLC
separation, nuclease P1 enrichment, two-dimensional TLC of adducted nucleotide bisphosphates
on PEI-cellulose, and quantification of adduct spots by liquid scintillation counting. Adduct
formation with the more reactive crotonaldehyde was included for comparison. Synthesized
adducted dG-3′-phosphates served as external standards for identification and quantification.
In calf thymus DNA, hexenal (0.2 mM) shows a time dependent formation of adducts, yielding
1.55 pmol/µmol of DNA at 5 h incubation. With crotonaldehyde (0.2 mM) the adduct rate was
about 10-fold higher. Hexenal also generated 1,N2-propano-dG adducts in the human
lymphoblastoid Namalva cell line (0.2 mM, 1 h, 86 fmol/µmol of DNA) and in primary rat
colon mucosa cells (0.4 mM, 30 min, 50 fmol/µmol of DNA). In primary colon mucosa cells
from rats and humans, hexenal and crotonaldehyde (0.4 mM, 30 min) induced DNA damage,
detected by single cell microgel electrophoresis (comet assay). In primary rat gastric mucosa
cells, hexenal was only weakly active, inducing detectable DNA damage in 20% of cells at 0.8
mM concentration. In contrast, primary mucosa cells from rat esophagus were as sensitive as
colon cells. After single oral application of hexenal to rats (up to 320 mg/kg body wt) DNA
damage was not detectable in gastrointestinal mucosa. Analysis of hexenal in selected flavored
foods revealed concentrations up to 14 ppm (0.14 mM) that are comparable to its natural
occurrence in some fruits and vegetables (up to 30 ppm). Thus, the concentration range selected
for the toxicological studies described here clearly is relevant: Hexenal, at concentrations found
in food, exerts genotoxic effects in cells from rat and human gastrointestinal tract.
use by the US Food and Drug Administration; the
Council of Europe included it in the list of artificial
flavoring substances (14, 15). However, the extent to
which hexenal is used as natural food additive in the
European Community is not known at present. The C4
homologue crotonaldehyde ((E)-2-butenal) occurs as com-
bustion product and has been detected in tobacco smoke
and other environmental media (16-18).
Both aldehydes are directly acting mutagens in bac-
teria and mammalian cells (19-22). They also induce
DNA single-strand breaks, chromosomal aberrations,
micronuclei, and sister chromatid exchanges in different
mammalian cells (23, 24). Crotonaldehyde induced liver
tumors in F344 rats after chronic administration in
drinking water at a dosage of 42 mg/L (25). Carcinoge-
nicity experiments with hexenal have not been described
yet. Organ toxicity was not observed in rats after feeding
hexenal in the diet (4000 ppm) for 13 weeks, which
corresponds to a total dose of approximately 250 mg/kg
body wt (26).
In tr od u ction
(E)-2-Hexenal (hexenal)1 is an important flavoring
compound which occurs naturally in many fruits and
vegetables (1-5). Maximum concentrations (up to 76
ppm) were found in bananas (3). Predominantly, hexenal
and other C6 aldehydes are formed by enzymatic lipid
peroxidation from C18 acids after disruption of cells (6-
8). The release of hexenal and other aldehydes by
lipoxygenase-hydroperoxide lyase from wounded plant
tissue is discussed as possible primary defense mecha-
nism against microbial infection (9, 10). In minor
amounts, hexenal is also generated by autoxidation
processes (11). Because of its “green” and fruity flavor,
hexenal is used as natural flavoring compound to gener-
ate “green notes” (12, 13). Hexenal was approved for food
* Author to whom correspondence should be addressed.
† University of Kaiserslautern.
‡ Federal Research Center for Nutrition, Karlsruhe.
X Abstract published in Advance ACS Abstracts, September 15, 1996.
1
Abbreviations: hexenal, (E)-2-hexenal; ATP, adenosine 5′-triph-
osphate; body wt, body weight; CrotdG, crotonaldehyde derived 1,N2-
(1,3-propano)dG; CrotdG3′p, crotonaldehyde derived 1,N2-(1,3-propano)-
dG-3′-phosphate; CrotdG5′p, crotonaldehyde derived 1,N2-(1,3-pro-
pano)dG-5′-phosphate; dG, 2′-deoxyguanosine; dG3′p, 2′-deoxygua-
nosine 3′-phosphate; dG5′p, 2′-deoxyguanosine 5′-phosphate; GC-MS,
gas chromatography/mass spectrometry; HexdG, hexenal derived 1,N2-
(1,3-propano)dG; HexdG3′p, hexenal derived 1,N2-(1,3-propano)dG-3′-
phosphate; HexdG5′p, hexenal derived 1,N2-(1,3-propano)dG-5′-phos-
phate; LMA, low melting agarose; MD, migration distance; MNNG,
N-methyl-N′-nitro-N-nitrosoguanidine; PEI-cellulose, poly(ethylen-
imine)-cellulose; SSB, single-strand breaks.
2-Alkenals react with cellular nucleophiles, predomi-
nantly in a Michael reaction where the â-carbon is the
prime site of nucleophilic attack. With glutathione, the
respective conjugates are formed predominantly under
catalysis by glutathione transferases, a process which
mainly leads to detoxification (24). With proteins, 1,4-
addition products and Schiff’s bases are formed, leading
to cytotoxicity. Both reaction types were observed in vitro
S0893-228x(96)00010-0 CCC: $12.00 © 1996 American Chemical Society

1208 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Go¨lzer et al.
Sch em e 1. F or m a tion of
1,N2-(1,3-P r op a n o)-2′-d eoxygu a n osin e 3′-P h osp h a te
Ad d u cts w ith Hexen a l (R1) (Hexd G3′p ) a n d
Cr oton a ld eh yd e (R2) (Cr otd G3′p )
Ma ter ia ls a n d Meth od s
Ch em ica ls. Hexenal (purity 99%) was obtained from Aldrich
(Steinheim, FRG), crotonaldehyde (purity > 99.5%) from Fluka
(Buchs, Switzerland), and (E)-2-octenal from Alfa (Karlsruhe,
FRG). Ca u tion : 2-Alkenals have been shown to be irritant and
genotoxic and therefore should be handled carefully. 2′-Deox-
yguanosine 3′-phosphate and 5′-phosphate (both sodium salts)
were purchased from Sigma Chemie (Deisenhofen, FRG), [γ-32P]-
ATP triethylammonium salt (specific activity >5000 Ci/mmol)
was from Amersham Buchler GmbH (Braunschweig, FRG), TLC
plates PEI-cellulose CEL 300 from Macherey & Nagel (Du¨ren,
FRG) and scintillation cocktail Aqualuma were from Baker
(Gross-Gerau, FRG). Autoradiographic films were X-omat AR
(Kodak) and Hyperfilm MP (Amersham). Proteinase K, ribo-
nuclease A type II A, micrococcal nuclease, spleen phosphodi-
esterase, nuclease P1, and nuclease S1 were provided from
Sigma, and cloned T4-polynucleotide kinase was obtained from
Amersham. Calf thymus DNA, highly polymerized, was from
Sigma, RPMI essential medium with L-glutamine was from
Gibco (Eggenstein, FRG), and low and normal melting agarose
was from Biozym Diagnostik GmbH (Hess. Oldendorf, FRG).
All other chemicals were of analytical grade.
Sch em e 2. 32P -P ostla belin g Assa y for
1,N2-P r op a n od eoxygu a n osin e Ad d u cts of Hexen a l
a n d Cr oton a ld eh yd e (Hexd G, Cr otd G)
In str u m en ta tion . Single cell electrophoresis was performed
with a Biometra electrophoresis unit. GC/FID analysis of
hexenal was carried out on a Hewlett Packard HP 5890 Series
II with autosampler HP 18596 B.
Syn th esis of Cr otd G- a n d Hexd G-P h osp h a tes. CrotdG-
and HexdG-3′- and -5′-phosphates (CrotG3′p, HexdG3′p,
CrotdG5′p, HexdG5′p) were prepared by reacting deoxygua-
nosine 3′-phosphate and 5′-phosphate (20 mM) with a 5-fold
molar excess of alkenal (100 mM) at 80 °C for 24 h in phosphate
buffer (0.01 M, pH 7.4) (36). Hexenal was predissolved in
acetonitrile (15% of incubation mixture). Samples were ex-
tracted twice with hexane to eliminate residual aldehyde,
evaporated to dryness by vacuum drying, redissolved in water,
and submitted to HPLC system 1 (Constametric LDC pump, λ:
254 nm; column: Hibar RP 18, 250 × 10 mm, 7 µm; eluent:
0.02 M ammonium formate, pH 5.4; flow rate: 6 mL/min;
sample volume: 1 mL). For separation of HexdGp, acetonitrile
(10%) was added to the eluent. Adduct purity was controlled
with HPLC system 2 (Hewlett Packard 1050, diode array
detection; column: Lichrospher RP-18 select B, 250 × 4 mm, 5
µm; eluent A: 0.02 M ammonium formate, pH 5.4; eluent B:
0.02 M ammonium formate, pH 5.4/acetonitrile (80/20), gradi-
ent: 50% B to 80% B in 30 min; flow rate: 0.7 mL/min).
Collected adduct fractions were identified by characteristic
differential UV absorption (λmax: 259 nm) compared to dG3′p
(λmax: 253 nm) and checked by 1H-NMR spectroscopy (21, 30,
37). Yields (% of theory) were as follows: CrotdG3′p, 8%;
CrotdG5′p, 30%; HexdG3′p, 1%; HexdG5′p, 4%. Identity of
nucleotides was confirmed after apyrase induced dephos-
phorylation: Adduct nucleotides (1 nmol) were incubated in 100
µL of acetate buffer (40 mM, pH 4.5) with 0.2 U of apyrase
(grade III, Sigma) at 37 °C for 2 h. Cochromatography of
reference nucleosides was performed on HPLC system 2.
Isola tion a n d Digestion of DNA. Cells (Namalva cells,
primary rat mucosa cells from colon and stomach) were lysed
in EDTA buffer (25 mM EDTA, 75 mM NaCl, pH 8.0) with 1%
SDS and 200 µg/mL proteinase K (38). After incubating
overnight at 55 °C, saturated NaCl solution was added to give
a final concentration of 1.5 M. Proteins were extracted by
adding 1 volume of chloroform, mixing (60 min), and centrifuga-
tion at 10 000 rpm (10 min). DNA was precipitated in the
aqueous supernatant by adding 1 volume of 2-propanol and
gently mixing (5 min, room temperature). The pellet was
washed twice with 70% ethanol and vacuum dried. DNA was
dissolved in 250 µL of citrate buffer (150 mM NaCl, 15 mM citric
acid, pH 7.4), and residual RNA was digested at 37 °C for 30
min with 40 µg of RNase A type II a (Sigma) (39). Saturated
NaCl solution was added to a final concentration of 0.6 M.
Proteins were extracted by adding 1 volume of phenol/isoamyl
alcohol (24/1) and mixing (15 min). DNA was precipitated from
the aqueous supernatant with 1 volume of 2-propanol (room
temperature), transferred into 70% ethanol, washed twice, and
on incubation of hemoglobin with hexenal (27). With 2′-
deoxyguanosine (dG) and dG-5′-phosphate (dG5′p) hex-
enal forms cyclic trans/ cis 1,N2-(1,3-propano)dG (HexdG)
adducts (28, 29) by ring closure at the 1 and N2 positions
of guanine from opposite directions as also described for
other alkenals (28, 30, 31). In general, pairs of diaster-
eomers were found for the respective trans/ cis adducts,
trans isomers being preferentially formed. The respective
crotonaldehyde 1,N2-(1,3-propano)dG adducts have been
detected in calf thymus DNA (30), in chinese hamster
ovary cells (32), and in vivo in mice skin after topical
treatment (33). There is some evidence that this adduct
is also formed endogenously in the liver of untreated rats
(31). Cyclic 7,8-dG adducts are also formed from hexenal
and crotonaldehyde (28).
We here describe the formation of the HexdG adduct
(see Scheme 1) in calf thymus DNA and in cells at low
hexenal concentrations. A modified 32P-postlabeling
method (34) was established for sensitive adduct deter-
mination (see Scheme 2). The more reactive homologue
crotonaldehyde was included for comparison. To further
evaluate the genotoxic potential of hexenal, induction of
DNA fragmentation was measured with single cell mi-
crogel electrophoresis (35). Primary gastrointestinal
mucosa cells (rat, human) were used in addition to a
sensitive human lymphoblastoid cell line. Selected
samples of processed foods and flavoring concentrates
were analyzed for hexenal in order to obtain more
information about the relevant concentration range to be
found in various foods.

Hexenal-Induced DNA Damage: 1,N2-Propano-dG Adducts
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1209
evaporated to dryness. DNA (10-100 µg) was dissolved in
buffer (2.5 mM Tris-HCl, 1.2 mM CaCl2, pH 8.8) (40) and
digested with micrococcal nuclease (5 U) and spleen phosphodi-
esterase (0.5 U) in a total volume of 60 µL (9 h at 37 °C).
Ad d u ct En r ich m en t. Digested DNA was chromatographed
on HPLC system 2, and adduct fractions were collected at
retention time intervals, known from separate experiments with
UV detectable amounts of 5′-nucleotide adducts. Hydrolyzed
DNA was quantified using 2′-deoxyadenosine 3′-phosphate as
external standard. Adduct fractions were evaporated to dryness
and dissolved in 20 µL of water. An aliquot (5 µL) was
submitted to digestion by nuclease P1 (10 U in 6 µL of sodium
acetate buffer: 100 mM, 55 µM ZnCl2, pH 5.0, 37 °C). After 1
h reaction was stopped with 1 µL of Tris base (0.5 M).
positive control N-methyl-N′-nitro-N-nitrosoguanidine (MNNG)
was dissolved in 0.7% NaCl. Colon and esophagus cells were
incubated at 37 °C for 30 min, gastric mucosa cells for 45 min.
Aliquots were checked for viability by trypan blue exclusion.
For postlabeling, cells were washed twice with RPMI and sub-
mitted to isolation of DNA. For microgel electrophoresis, cells
were centrifuged and resuspended to 2 × 105 cells in 10 µL.
In Vivo Exp er im en ts. Male Sprague-Dawley rats (6-8
weeks old) were gavaged with hexenal (40-320 mg/kg body wt),
dissolved in corn oil. MNNG (5 mg/kg body wt in 0.9% NaCl
solution) was used as positive control, corn oil as solvent control.
Rats were sacrificed 1 and 16 h after treatment for preparation
of stomach and colon, respectively. Mucosa cells were isolated
as described above and submitted to 32P-postlabeling and to
microgel electrophoresis.
Micr ogel Electr op h or esis. The assay was performed basi-
cally according to Singh et al. (44). Cell suspensions (10 µL)
were mixed with 65 µL of 0.7% low melting agarose (LMA) and
distributed on slides covered with 0.5% normal melting agarose
(45). After addition of a third layer of LMA, cells were lysed at
pH 10 and subjected to electrophoresis at 25 V/300 mA under
alkaline conditions. Then DNA was stained with ethidium
bromide, and migration distance (MD) of DNA (comet) was
determined by image analysis. Cells were classified according
to their image length as undamaged (<40 µm), moderately
damaged (40-80 µm), and severely damaged (>80 µm) (35). A
total of 100 cells/slide (2-3 slides/experimental point) was
evaluated. The proportion of damaged to undamaged cells was
expressed in % of total.
32P -P ostla belin g. The sample fraction was incubated (3 h,
37 °C) in 3 µL of labeling buffer (60 mM MgCl2, 60 mM
dithiothreitol, 6 mM spermidine, 60 mM Bicine/NaOH, pH 9.5)
containing T4 polynucleotide kinase (5 U) and [γ-32P]ATP
(specific activity 5000 Ci/mmol; 5 µCi). Subsequently, 4 µL of
nuclease P1 mix (200 mM sodium acetate, pH 4.5; 40 mM MgCl2,
5 U nuclease P1) was added. After incubation (1 h, 37 °C) the
mixture was reacted with 3 µL of apyrase (5 mU, 0.8 M Tris,
pH 8.5; 37 °C, 15 min); an aliquot (10 µL) was submitted to TLC
separation on prewashed 10 × 20 cm poly(ethylenimine) (PEI)-
cellulose plates (first direction: 0.2 M sodium phosphate buffer,
pH 6.5). The upper part of the plate was cut off and the
remaining (10 cm) was dried and developed in the second
direction (90°) using 2-propanol/water/25% NH3 (80/58/22).
After drying it was autoradiographed with intensifying screen
for up to 24 h. Adduct spots were cut out and quantified by
liquid scintillation counting. Each 32P-postlabeling assay in-
cluded one untreated control (untreated DNA, obtained from
the same source as the samples and subjected to the whole
procedure). Radioactivity of the respective TLC spot area was
determined and used to establish the detection limit (2-fold
background). Appropriate amounts of synthetic adduct stan-
dard, added to untreated controls (n ) 3) prior to nuclease P1
enrichment, were analyzed to correct for adduct losses.
Deter m in a tion of Hexen a l in F ood a n d F la vor in g Con -
cen tr a tes. Samples (food: 20 g; flavor concentrates: 20 mg)
were homogenized/diluted with varying amounts of water, and
hexenal was extracted into hexane. After centrifugation, the
organic layer was directly submitted to GC/FID (capillary
column: DB Wax: 50 m × 0.32 mm × 1 µm; Macherey & Nagel)
using (E)-2-octenal as internal standard. Recovery ranged about
90%, and the detection limit varied from 0.2 to 0.5 ppm. Positive
results were confirmed by GC-MS.
In cu ba tion of Ca lf Th ym u s DNA. Calf thymus DNA (100
µg), dissolved in 200 µL Tris-HCl (0.125 M, pH 7.4), was
incubated with hexenal (0.02-2 mM), added in 50 µL DMSO
(20 h at 37 °C). The solution was adjusted to 1.5 M NaCl, and
DNA was precipitated by addition of 1 volume of 2-propanol and
washed twice with 70% ethanol before evaporation to dryness.
In cu ba tion of Na m a lva Cells. Human lymphoblastoid
Namalva cells, originating from a Burkitt lymphoma (41), were
cultivated at 37 °C in 5% CO2 atmosphere using RPMI 1640
essential medium with glutamine, supplemented with 5% new-
born calf serum and antibiotics. Cell suspensions (10 mL, 2 ×
106 cells/mL) were incubated with hexenal, predissolved in 100
µL of DMSO, in serum free medium under shaking (37 °C for
1-20 h). Viability was assessed by trypan blue exclusion. After
incubation, cells were washed twice with RPMI medium, centri-
fuged (10 min, 200g, 20 °C), and submitted to isolation of DNA.
Isola tion a n d In cu ba tion of P r im a r y Mu cosa Cells fr om
Colon , Stom a ch , a n d Esop h a gu s. Mucosa cells were isolated
according to Burlinson (42) with slight modifications (35). Colon
and stomach were excised, perfused with Ca2+, Mg2+ free PBS
for 5 min, filled with Ca2+ and Mg2+ free HBSS containing 50
U/mL protease K, and incubated at 37 °C for 30 (colon) or 45
min (stomach). Then the organs were opened and cells were
isolated by gentle agitation. The suspensions were centrifuged
at 800 rpm (colon) or 500 rpm (stomach); and the pellet was
resuspended in 10 mL of RPMI 1640 and adjusted to 2 × 106
cells/mL. Esophagus was directly filled with proteinase K
solution and incubated 30 min at 37 °C. Then isolated cells were
centrifuged at 800 rpm and resuspended in 1 mL of RPMI.
Human colon mucosa biopsy samples were incubated for 30 min
at 37 °C with HBSS containing 50 U/mL protease K. Cell
suspensions were centrifuged and adjusted to 1 × 105 cells/mL
in RPMI 1640 (43). Viability of cells, as determined by trypan
blue exclusion, exceeded 80%.
Resu lts
To investigate the formation of hexenal adducts in
DNA, a modified 32P-postlabeling method was developed.
Generation of the respective crotonaldehyde adduct was
included for comparison. Reference nucleotides obtained
from reaction of aldehyde with dG-3′- and -5′p (CrotdG3′p;
CrotdG5′p; HexdG3′p; HexdG5′p) showed the expected
pairs of signals indicative for mixtures of diastereomers
for each adduct. dG3′p adducts were used as external
standard to control the 32P-postlabeling procedure (see
Scheme 2): After enzymatic digestion of DNA into 3′-
nucleotides, the sample was submitted to adduct enrich-
ment. HPLC separation of adducted 3′-nucleotides was
achieved on RP-18, using an ammonium formate/aceto-
nitrile gradient and the respective 5′p adducts as UV
markers. The collected adduct fraction containing both
isomers was incubated with nuclease P1 to further reduce
the amount of normal nucleotides (34). This two-step
adduct enrichment allowed subsequent 32P-postlabeling
under ATP excess without addition of nonlabeled ATP.
Control experiments showed a maximal adduct-labeling
efficiency of about 60%. After labeling, samples were
treated again with nuclease P1 to dephosphorylate traces
of unmodified nucleotides, while the adducted bisphos-
phates remain unchanged. This results in a better
adduct separation on PEI-cellulose; adduct isomers,
however, are not separated under the described condi-
tions. Adduct spots were identified by autoradiography,
cut out, and quantified by liquid scintillation counting
in comparison to untreated and adduct-spiked control
samples. The detection limit for hexdG3′p was ap-
proximately 1 adduct/5 × 107 nucleotides. For CrotdG3′p
Test compounds, dissolved in DMSO, were added to the cell
suspension to give a final DMSO concentration of 1%. The

1210 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Go¨lzer et al.
Ta ble 1. F or m a tion of 1,N2-P r op a n od eoxygu a n osin e
Ad d u cts in Ca lf Th ym u s DNA by Hexen a l (Hexd G) a n d
Cr oton a ld eh yd e (Cr otd G)a
Ta ble 3. F or m a tion of 1,N2-P r op a n od eoxygu a n osin e
Ad d u cts (Hexd G) in P r im a r y Ra t Colon Cells by Hexen a l
hexenal (mM)
viability (%)
HexdG (fmol/µmol of DNA)
aldehyde incubation
HexdG
CrotdG
0.2
0.4
0.8*
83 ( 5
87 ( 2
80 ( 10
n/d
50 ( 13
162 ( 33
(mM)
time (h) (pmol/µmol of DNA) (pmol/µmol of DNA)
0.02
0.2
20
2
0.36*
n/a
0.30 ( 0.16
1.55 ( 0.89
6.9 ( 2.6
8.9 ( 0.7
9.2 ( 3.0
16.8*
n/a
a
n/d ) not detectable (<20 fmol/µmol of DNA, corresponding
to 1 adduct/5 × 107 nucleotides); (*): autoradiogram shown in
Figure 1. Means and ranges were obtained from two independent
experiments. Incubation: 30 min, 37 °C, 2 × 106 cells/mL.
5
13.9*
n/a
n/a
72.6 ( 6.6
n/a
20
2
5
20
2
a
n/a ) not analyzed. Means and ranges were obtained from 2
experiments except for (*): n ) 1. After incubation (37 °C; 100
µg of DNA/250 µL of incubation mixture) and digestion of DNA
an aliquot =20 µg of DNA was submitted to 32P-postlabeling.
Ta ble 2. F or m a tion of 1,N2-P r op a n od eoxygu a n osin e
Ad d u cts (Hexd G) in Na m a lva Cells by Hexen a la
hexenal
(mM)
incubation
time (h)
viability
(%)
HexdG
(fmol/µmol of DNA)
F igu r e 1. Detection of HexdG in primary rat colon mucosa
cells. Autoradiograms of two-dimensional TLC separations. (A)
Untreated control. (B) Untreated control, fortified with adduct
standard (34 fmol of HexdG3′p) prior to 32P-postlabeling. (C)
Sample: 2 × 106 cells/mL incubated with hexenal (0.8 mM) for
30 min at 37 °C (see Table 3). The whole sample was submitted
to adduct determination by 32P-postlabeling. Or: origin. Add:
HexdG-3′,5′-bisphosphate. Autoradiography: 20 h at 22 °C,
Amersham Hyperfilm MP, with intensifying screens.
0
0.2
86 ( 4
88
1
3
5
20
1
3
5
86 ( 3
211 ( 122
135 ( 33
72 ( 26
171 ( 43
297 ( 50
n/a
85
81
44
0.4
80
65
30
a
n/a ) not analyzed. Means and ranges were obtained from
2-4 experiments. Incubation: 37 °C, 2 × 107 cells/10 mL of
incubation mixture.
it was approximately 1 adduct/106 nucleotides because
of higher background of untreated control in the respec-
tive TLC region.
Results of adduct formation experiments in calf thymus
DNA (Table 1) show that adduct levels increased with
incubation time at 0.2 mM hexenal, but not at 2 mM.
Crotonaldehyde induced about 1 order of magnitude more
DNA adducts than hexenal. At high aldehyde exposure
incomplete hydrolysis due to strongly modified DNA
might lead to underestimation of adduct level.
Incubations of RNA with hexenal had shown that
adducted nucleotides are not separated from the respec-
tive DNA adducts during enrichment, labeling, and TLC
separation. Thus, in experiments with cells, RNase
treatment is required before hydrolysis of DNA. Incuba-
tion of Namalva cells with 0.2 and 0.4 mM hexenal for 1
h resulted in mean adduct levels of 86 and 171 fmol/µmol
of DNA, respectively (Table 2). Under these conditions,
viability of cells was not significantly reduced, compared
to the control. Higher concentrations or prolonged
incubation at 0.4 mM hexenal, however, exerted cytotoxic
effects with relative viabilities below 70%. Incubation
with 50 µM hexenal (5 h) indicated minute adduct
formation, below the determination limit.
Incubation of colon mucosa cells for 30 min with 0.4
and 0.8 mM hexenal induced detectable adducts (see
Table 3). As an example, the autoradiogram resulting
from 0.8 mM incubation is shown in Figure 1C, in
comparison to untreated control (Figure 1A) and control
spiked with adducted standard (Figure 1B). In gastric
mucosa cells incubated for 1 h with hexenal (up to 1.2
mM) adducts were not detectable, however. Autoradio-
grams from these cells showed a background spot in the
adduct area, which compromised the detection limit. The
structure of this background adduct cannot be identified.
The spot might, however, well originate from endogenous
sources. 1,N2-Propano-dG adducts from other alkenals
(acrolein, crotonaldehyde) have been detected in DNA
F igu r e 2. Concentration dependent induction of DNA damage
with hexenal and crotonaldehyde in primary rat and human
cells. (A) Primary rat colon mucosa cells: hexenal (0) and
crotonaldehyde (b); human colon biopsy samples: hexenal (×).
(B) Primary rat stomach mucosa cells: hexenal (0) and cro-
tonaldehyde (b). Means and SD were obtained from 600 cells
of 2 independent experiments, each experiment performed in
triplicate (100 cells/slide; 3 slides/experiment). Viability of cells
>80%; undamaged cells: MD <40 µm.
from untreated rodents and humans. Induction of DNA
damage, assayed by single cell electrophoresis, was
investigated in parallel. Results are expressed as ratio
of undamaged cells in % of total (Figure 2). Rat colon
mucosa cells (Figure 2A) were sensitive toward hexenal
and crotonaldehyde. Incubation with both aldehydes (0.4
mM for 30 min) clearly induced genotoxic effects (DNA

Hexenal-Induced DNA Damage: 1,N2-Propano-dG Adducts
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1211
Since hexenal is less reactive than the shorter chain
homologues, a sensitive detection method was required.
The 32P-postlabeling procedure was therefore modified
accordingly. A two-step adduct enrichment by reversed-
phase HPLC and nuclease P1 treatment was found
appropriate; adducted bisphosphates were resistant to
dephosphorylation (34). TLC separation, a critical step
for small adducts, was facilitated by a second nuclease
P1 treatment after the 32P-postlabeling step. A similar
approach has recently been published for cyclic adducts
of acrolein and crotonaldehyde (31).
Adduct formation by hexenal was observed in Namalva
cells, a human lymphoblastoid cell line poor in glu-
tathione, glutathione S-transferase, and detoxifying en-
zymes (24, 50). Hexenal has been shown earlier to induce
DNA single-strand breaks (SSB) in Namalva cells (24).
As hexenal is mainly ingested with food, primary gastric,
colon, and esophagus mucosa cells were examined for
DNA adduct formation and DNA damage (comet assay).
The comet assay has been found more appropriate for
mucosa cells than alkaline elution, since a smaller
number of cells is required and secretion of mucous
material causes fewer interferences. Adduct levels in
primary rat colon cells, extrapolated to an incubation
time of 1 h, were about 50% those found for Namalva
cells. This might be due to a higher capacity of colon
cells for detoxification by quenching reactions with
noncritical cellular nucleophiles or by deactivating
biotransformation, as described earlier for primary hepa-
tocytes (24). It is not clear why primary gastric mucosa
cells were only weakly sensitive toward hexenal but
highly sensitive to crotonaldehyde. Substrate specific
differences in uptake or detoxification might play a role.
F igu r e 3. DNA damage with hexenal and crotonaldehyde (both
0.8 mM) in primary rat mucosa cells of different tissues. Ratios
of undamaged cells (MD <40 µm), moderately damaged cells
(MD 40-80 µm), and severely damaged cells (MD >80 µm) are
given in percent of total. Bars: (1) untreated control, esophagus,
n ) 1*; (2) stomach: hexenal, n ) 6; (3) colon: hexenal, n ) 6;
(4) esophagus: hexenal, n ) 1*; (5) stomach: crotonaldehyde,
n ) 6; (6) colon: crotonaldehyde, n ) 6; n ) number of pooled
slides from repetitive experiments. (*) Preliminary pilot experi-
ment. Viability of cells >80%.
damage in more than 40% of cells). In primary human
colon mucosa cells hexenal was as potent as in rat colon
cells. In rat stomach mucosa cells (Figure 2B) hexenal
was only weakly effective (DNA damage in 20% of cells
at 0.8 mM) whereas crotonaldehyde was equally active
in both cell types. Effects of hexenal and crotonaldehyde
at 0.8 mM in rat tissues are presented in detail in Figure
3. The percentage of undamaged cells (MD <40 µm) after
crotonaldehyde treatment (bars 5 and 6) is significantly
different from that obtained with hexenal (bars 2 and 3)
in the corresponding rat tissues (Student’s T test, p
<0.01). A pilot experiment with primary mucosa cells
from rat esophagus, included for comparison, suggests
comparable sensitivity toward hexenal.
Depending on the respective aldehyde and also on cell
type, the in vitro results show distinctly differential
sensitivity with regard to DNA interactions. Similar
differences have also been described for mutagenicity of
acrolein in mammalian cells (32). Significant induction
of DNA damage was only observed at alkenal concentra-
tions rather close to the cytotoxic concentration limit, as
evidenced by a reduction in the percentage of surviving
cells. This is in line with in vitro mutagenicity experi-
ments in chinese hamster lung V79 cells with homologous
alkenals (22). Hexenal and crotonaldehyde (5-250 µM)
have shown concentration dependent induction of micro-
nuclei, sister chromatid exchanges, and numerical aber-
rations in Namalva cells and also in human blood
lymphocytes (23). Hexenal has been found to be more
potent than crotonaldehyde in induction of numerical
chromosomal aberrations and inhibition of tubulin po-
lymerization (23, 51).
After oral application of hexenal to rats, DNA damage
was not detectable in mucosa cells of stomach (160 mg/
kg body wt) and colon (320 mg/kg body wt) whereas the
positive control MNNG (5 mg/kg body wt) was clearly
effective in both tissues.
GC analysis of 26 selected foods and flavoring concen-
trates showed hexenal in substantial amounts in banana
(42 ppm), endive (27 ppm), kiwi-apple fruit preparation
(7 ppm), alcoholic beverage with apple taste (14 and 1.9
ppm), and apple flavor concentrate (11 400 ppm). Thus,
the hexenal concentrations in apple-kiwi fruit prepara-
tion and in the alcoholic beverages with apple taste were
comparable to the natural occurrence in some fruits or
vegetables such as banana or endive. In other food
samples such as baby foods, milk products, and alcoholic
and nonalcoholic drinks hexenal was not detectable. One
apple flavor concentrate contained more than 1% hex-
enal, while a sample of banana flavor concentrate tested
in parallel was negative.
The negative results of the in vivo experiments with
hexenal in rats need further investigation. It is not clear
whether the time interval between application and
isolation of cells (gastric mucosa: 1 h; colon: 16 h) was
appropriate. These time points had been shown to be
suitable for MNNG in gastric and colon, respectively (35),
but hexenal may require different exposure lengths.
Moreover, a single dose experiment might not reflect
effects of chronic exposure. Organs such as the liver, the
target organ of crotonaldehyde carcinogenicity, might be
more relevant. In volunteers, exposure to hexenal (mouth
washing) indicated cytogenetic effects in human oral
mucosa cells.2
Discu ssion
To evaluate the potency of hexenal to interact with
DNA, we focused on cyclic 1,N2-propano-dG adducts as
it was known that this adduct type, formed in vivo from
acrolein and crotonaldehyde (33, 40), induces mutations.
With the model adduct, primarily 1,N2-propano-dG G f
T transversions were observed in Escherichia coli and
simian kidney cells (COS) (46-48). In certain sequences
1,N2-propano-dG also induces frame shift mutations (49).
2
Dittberner et al., unpublished results.

1212 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Go¨lzer et al.
(14) Opdyke, D. L. J . (1975) Monographs on fragrance raw materials.
Human exogenous exposure to hexenal mainly results
from food. In addition to its natural occurrence in a
variety of fruits and vegetables, hexenal is widely used
as a flavoring agent. Potential adduct formation by
hexenal in vivo needs further elucidation, also with
respect to the endogenous adduct burden resulting from
potential in vivo formation of hexenal during lipid
peroxidation. The respective crotonaldehyde adducts
have been detected in liver DNA of untreated rodents
and humans (52). Hexenal has been identified as lipid
peroxidation product in the liver of bromobenzene-
poisoned mice (53). Our in vitro results show that
hexenal forms DNA adducts and induces DNA damage
in human lymphoblastoid cells and in colon mucosa cells
from rats and humans. Appropriate in vivo studies are
required to evaluate the risk of chronic hexenal exposure
in food.
Hexen-2-al. Food Cosmet. Toxicol. 13, 453-454.
(15) Council of Europe (1981) Flavouring substances and natural
sources of flavourings, 3rd ed., Strasbourg, Maisonneuve, France.
(16) Wynder, E. L., Goodman, D. A., and Hoffmann, D. (1965)
Ciliatoxic compounds in cigarette smoke. II. Carboxylic acids and
aldehydes. Cancer 18, 505-509.
(17) Sakuma, H., Murakata, S., and Sugawara, S. (1981) Volatile
products of cellulose pyrolysis. Agric. Biol. Chem. 45, 443-451.
(18) Simmon, V. F., Kauhanen, K., and Tardiff, R. G. (1977) Mutagenic
activity of chemicals identified in drinking water. In Progress in
genetic toxicology (Scott, D., Bridges, B. A., and Sobels, F. H., Eds.)
pp 249-258, Elsevier, Amsterdam.
(19) Lutz, D., Eder, E., Neudecker, T., and Henschler, D. (1982)
Structure mutagenicity relationship in R,â-unsaturated carbonylic
compounds and their corresponding allylic alcohols. Mutat. Res.
93, 305-315.
(20) Marnett, L. J ., Hurd, H. K., Hollstein, M. C., Levin, D. E.,
Esterbauer, H., and Ames, B. N. (1985) Natural occurring
carbonyl compounds are mutagens in Salmonella tester strain TA
104. Mutat. Res. 148, 25-34.
(21) Eder, E., Deininger, C., Neudecker, T., and Deininger, D. (1992)
Mutagenicity of â-alkylsubstituted acrolein congeners in the
Salmonella typhimurium strain TA 100 and genotoxicity testing
in the SOS-chromotest. Environ. Mol. Mutagen. 19, 338-345.
(22) Canonero, R., Martelli, A., Marinari, U. M., and Brambilla, G.
(1990) Mutation induction in Chinese hamster lung V79 cells by
five alk-2-enals produced by lipid peroxidation. Mutat. Res. 244,
153-156.
Ack n ow led gm en t. This work was supported in part
by the Deutsche Forschungsgemeinschaft (DFG), Grant
Ei 172/3-2. We thank M. Diehl, A. Kissels, H. Schneider,
and D. Scha¨fer for competent assistance, M. Lorez for
performing the animal experiments, Prof. M. Metzler and
E. Pfeiffer for GC-MS analyses, M. Knoll and R. Lam-
bertz for microgel analysis, and I. Hemm for help with
the manuscript preparation.
(23) Dittberner, U., Eisenbrand, G., and Zankl, H. (1995) Genotoxic
effects of the R,â-unsaturated aldehydes 2-trans-butenal, 2-trans-
hexenal and 2-trans,6-cis-nonadienal. Mutat. Res. 335, 259-265.
(24) Eisenbrand, G., Schuhmacher, J ., and Go¨lzer, P. (1995) The
influence of glutathione and detoxifying enzymes on DNA damage
induced by 2-alkenals in primary rat hepatocytes and human
lymphoblastoid cells. Chem. Res. Toxicol. 8, 40-46.
(25) Chung, F.-L., Tanaka, T., and Hecht, S. S. (1986) Induction of
liver tumors in F344 rats by crotonaldehyde. Cancer Res. 46,
1285-1289.
(26) Gaunt, I. F., Colley, J ., Wright, M., Creasy, M., Grasso, P., and
Gangolli, S. D. (1971) Acute and short-term toxicity studies of
trans-2-hexenal. Food Cosmet. Toxicol. 9, 775-786.
(27) Kautiainen, A. (1992) Determination of hemoglobin adducts from
aldehydes formed during lipid peroxidation in vitro. Chem.-Biol.
Interact. 83, 55-63.
Refer en ces
(1) Go¨tz-Schmidt, E.-M., and Schreier, P. (1986) Neutral volatiles
from blended endive (Cichorium endivia, L.). J . Agric. Food Chem.
34, 212-215.
(2) De Vincenzi, M., Badellino, E., Di Folco, S., Dracos, A., Magliola,
M., Stacchini, A., Stacchini, P., and Silano, V. (1989) A basis for
estimation of consumption: literature values for selected food
volatiles, Part III. Food Addit. Contam. 6, 235-267.
(3) Feron. V. J ., Til, H. P., de Vrijer, F., Woutersen, R. A., Cassee, F.
R., and van Bladeren, P. J . (1991) Aldehydes: occurrence,
carcinogenic potential, mechanism of action and risk assessment.
Mutat. Res. 259, 363-385.
(4) Luning, P. A., de Rijk, T., Wichers, H. J ., and Roozen, J . P. (1994)
Gas chromatography, mass spectrometry, and sniffing port
analyses of volatile compounds of fresh bell peppers (Capsicum
annuum) at different ripening stages. J . Agric. Food Chem. 42,
977-983.
(5) van Ruth, S. M., Roozen, J . P., and Cozijnsen, J . L. (1995) Volatile
compounds of rehydrated French beans, bell peppers and leeks.
Part 1. Flavor release in the mouth and in three mouth model
systems. Food Chem. 53, 15-22.
(6) Hatanaka, A., Kajiwara, T., and Sekiya, J . (1986) Fatty acid
hydroperoxide lyase in plant tissues: volatile aldehyde formation
from linoleic and linolenic acid. In Biogeneration of aromas
(Parliment, T. H., and Croteau, R., Eds) pp 167-175, American
Chemical Society, Washington, DC.
(7) Schreier, P., and Lorenz, G. (1981) Formation of “green-grassy”-
notes in disrupted plant tissues: characterization of the tomato
enzyme systems. In Flavour 81, Weurman Symp., 3rd (Schreier,
P., Ed.) pp 495-507, de Gruyter, Berlin, FRG.
(8) Wu, C.-M., and Liou, S.-E. (1986) Effect of tissue disruption on
volatile constituents of bell peppers. J . Agric. Food Chem. 34,
770-772.
(9) Andersen, R. A., McCracken, C. T., J r., Hamilton-Kemp, T. R.,
and Hildebrand, D. F. (1995) Nitrosation of tobacco alkaloids
during storage reduced by antimicrobial (E)-2-hexenal vapor. ACS
Symposium Series 584 (Synthesis and Chemistry of Agrochemi-
cals IV), 463-472.
(28) Eder, E., and Hoffman, C. (1993) Identification and characteriza-
tion of deoxyguanosine adducts of mutagenic â-alkyl-substituted
acrolein congeners. Chem. Res. Toxicol. 6, 486-494.
(29) Schuhmacher, J . (1990) Untersuchungen zur Gentoxizita¨t und
zum Metabolismus aromawirksamer R,â-ungesa¨ttigter Aldehyde,
Diss. D 386, Kaiserslautern, FRG.
(30) Chung, F.-L., Young, R., and Hecht, S. S. (1984) Formation of
2-propanodeoxyguanosine adducts in DNA upon reaction
cyclic 1,N
with acrolein or crotonaldehyde. Cancer Res. 44, 990-995.
(31) Nath, R. G., Chen, H.-J . C., Nishikawa, A., Young-Sciame, R.,
and Chung, F.-L. (1994) A 32P-postlabeling method for simulta-
neous detection and quantification of exocyclic etheno and pro-
pano adducts in DNA. Carcinogenesis 15, 979-984.
(32) Foiles, P. G., Akerkar, S. A., Miglietta, L. M., and Chung, F.-L.
(1990) Formation of cyclic deoxyguanosine adducts in Chinese
hamster ovary cells by acrolein and crotonaldehyde. Carcinogen-
esis 11, 2059-2061.
(33) Chung, F.-L., Young, R., and Hecht, S. S. (1989) Detection of cyclic
1,N2-propanodeoxyguanosine adducts in DNA of rats treated with
N-nitrosopyrrolidine and mice treated with crotonaldehyde. Car-
cinogenesis 10, 1291-1297.
(34) Go¨lzer, P., J anzowski, C., and Eisenbrand, G. (1993) Detection
of crotonaldehyde and E-2-hexenal 1,N2-deoxyguanosine adducts
in calf thymus DNA by 32P-postlabelling. J . Cancer Res. Clin.
Oncol. 119 (Suppl. 2), 56.
(35) Pool-Zobel, B. L., Guigas, C., Klein, R., Neudecker, Ch., Renner,
H. W., and Schmezer P. (1993) Assessment of genotoxic effects of
lindane. Food Chem. Toxicol. 31, 271-283.
(36) Eder. E., and Hoffmanm, C. (1992) Identification and character-
ization of deoxyguanosine-crotonaldehyde adducts. Formation of
7,8 cyclic adducts and 1,N2,7,8 bis-cyclic adducts. Chem. Res.
Toxicol. 5, 802-808.
(10) Hamilton-Kemp, T. R., McCracken, C. T., Andersen, R. A., and
Hildebrand, D. F. (1995) Antimicrobial properties of natural
volatile compounds. ACS Symposium Series 584 (Synthesis and
Chemistry of Agrochemicals IV), 449-462.
(11) Esterbauer, H., Eckl, P., and Ortner, A. (1990) Possible mutagens
derived from lipids and lipid precursors. Mutat. Res. 238, 223-
233.
(37) Chung, F.-L., and Hecht, S. S. (1983) Formation of cyclic 1,N2-
adducts by reaction of deoxyguanosine with R-acetoxy-N-nitros-
opyrrolidine, 4-(carbethoxynitrosamino)butanal, and crotonalde-
hyde. Cancer Res. 43, 1230-1235.
(12) Boelens, M. H., and van Gemert, L. J . (1987) Organoleptic
properties of aliphatic aldehydes. Perfum. Flavor. 12, 31-43.
(13) Ziegler, E. (1982) Die natu¨rlichen und ku¨nstlichen Aromen, Dr.
Alfred Hu¨thig Verlag GmbH, Heidelberg.

Hexenal-Induced DNA Damage: 1,N2-Propano-dG Adducts
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1213
(38) Mu¨llenbach, R., Lagoda, P. J . L., and Welter, C. (1989) An efficient
salt-chloroform extraction of DNA from blood and tissue. Trends
Genet. 5, 391.
(39) Davis, L. G., Dibner, M. D., and Battey J . F. (1986) Basic methods
in molecular biology, Elsevier, New York.
(40) Wilson, V. L., Foiles, P. G., Chung, F.-L., Povey, A. C., Frank, A.
A., and Harris, C. C. (1991) Detection of acrolein and crotonal-
dehyde DNA adducts in cultured human cells and canine periph-
eral blood lymphocytes by 32P-postlabeling and nucleotide chro-
matography. Carcinogenesis 12, 1483-1490.
(47) Moriya, M., Marinelli, E., Shibutani, S., and J oseph, J . (1989)
Site-specific mutagenesis using the model exocyclic DNA adduct,
1,N2-propanodeoxyguanosine. Proc. Am. Assoc. Cancer Res. 30,
A 555.
(48) Moriya, M., Zhang, W., J ohnson, F., and Grollman, A. P. (1994)
Mutagenic potency of exocyclic DNA adducts marked differences
between Escherichia coli and simian kidney cells. Proc. Natl.
Acad. Sci. U.S.A. 91, 11899-11903.
(49) Benamira, M., Singh, U., and Marnett, L. J . (1992) Site-specific
frameshift mutagenesis by
a propanodeoxyguanosine adduct
(41) Wurm, F., Polastri, G., Hilfenhaus, J ., Harth, H., and Zankl, H.
(1985) Long term cultivation of Namalva cells for interferon
production: stable cytogenetic markers for identification of cells
in spite of drastic chromosomal variation. Dev. Biol. Stand. 60,
393-403.
(42) Burlinson, B. (1989) An in vivo unscheduled DNA synthesis (UDS)
assay in the rat gastric mucosa: preliminary development.
Carcinogenesis 10, 1425-1428.
(43) Pool-Zobel, B. L., Lotzmann, N., Knoll, M., Kuchenmeister, F.,
Lambertz, R., Leucht, U., Schro¨der, H. G., and Schmezer, P.
(1994) Detection of genotoxic effects in human gastric and nasal
mucosa cells isolated from biopsy samples. Environ. Mol. Mu-
tagen. 24, 23-45.
(44) Singh, N. P., McCoy, M. T., Tice, R. R., and Schneider, E. L. (1988)
A simple technique for quantitation of low levels of DNA damage
in individual cells. Exp. Cell Res. 175, 184-191.
positioned in the (CpG)4 hot-spot of Salmonella typhimurium his
D3052 carried on an M13 vector. J . Biol. Chem. 267, 22392-
22400.
(50) J anzowski, C., J acob, D., Henn, I., Zankl, H., Pool-Zobel, B. L.,
Wiessler, M., and Eisenbrand, G. (1989) Investigation on organ-
specific metabolism and genotoxic effects of the urinary bladder
carcinogen N-nitrosobutyl-3-carboxypropylamine (BCPN) and its
analogs N-nitrosodibutylamine (NDBA) and N-nitrosobutyl-4-
hydroxybutylamine (4-OH-NDBA). Toxicology 59, 195-209.
(51) Pfeifer, E., and Metzler, M. (1992) Inhibitors of microtubule
assembly: covalently and non-covalently binding substances have
different no-effect-level characteristics in vitro. Mutagenesis 7,
165-166.
(52) Nath, R. G., and Chung, F.-L. (1993) Detection of 1,N2-propan-
odeoxyguanosine adducts in rodent and human liver DNA by 32P-
postlabeling. Proc. Am. Assoc. Cancer Res. 34, 137.
(45) Betti, C., Barale, R., and Pool-Zobel, B. L. (1993) Comparative
studies on cytotoxic and genotoxic effects of two organic mercury
compounds in lymphocytes and gastric mucosa cells of Sprague-
Dawley rats. Environ. Mol. Mutagen. 22, 172-180.
(46) Grollman, A. P. (1989) Molecular mechanisms of chemical mu-
tagenesis. Proc. Am. Assoc. Cancer Res. 30, 682.
(53) Benedetti, A., Pompella, A., Fulceri, R., Romani, A., and Comporti
M. (1986) Detection of 4-hydroxynonenal and other lipid peroxi-
dation products in the liver of bromobenzene-poisoned mice.
Biochim. Biophys. Acta 876, 658-666.
TX9600107