Download Lipid Hydroperoxide Activation of N-Hydroxy-N

[CANCER RESEARCH 36, 2761-2767, August 1976]
Lipid Hydroperoxide Activation of N-Hydroxy-N
acetylaminofluorene via a Free Radical Route1
Robert A. Floyd,2 Lailing M. Soong, Robert N. Walker, and Melissa Stuart
Oklahoma Medical Research Foundation, Biomembrane Research Laboratory, Oklahoma City, Oklahoma 73104
SUMMARY
The data presented here demonstrate that linoleic acid
hydroperoxide in the presence of methemoglobin or hema
tin activated the carcinogen N-hyd roxy-N-acetyl-2-amino
fluorene via the nitroxyl free radical intermediate into 2nitrosofluorene and N-acetoxy-N-acetyl-2-aminofluorene.
Ascorbate inhibited the activation, in which case the free
radical intermediate was replaced by the ascorbate free
radical. On the basis of optical kinetics,we have established
that the rate of linoleic acid hydroperoxide decrease paral
leled the rate of N-hydroxy-N-acetyl-2-aminofluorene
de
crease and also the rate of 2-nitrosofluorene increase. The
stoichiometry of the reaction was such that, for every 2
linoleic acid hydroperoxide molecules consumed, 2 N-hy
d roxy-N-acetyl-2-aminofluorene
molecules were oxidized
and 1 2-nitrosofluorene and 1 N-acetoxy-N-acetyl-2 amino
fluorenemolecule was formed.
INTRODUCTION
The exact chemical events that result in the activation of
the arylamine carcinogens to the ultimate reactive chemical
form are still an open question. However, much has been
learned in recent years (see Refs. 15, 16, and 21 for re
views). For example, it was discovered in 1960 (5) that the
N-hydroxylation of AAF3 into N-OH-AAF considerably en
hanced the potency of this carcinogen; thus, this step is
now considered an essential step in the activation of the
arylamine carcinogens. N-OH-AAF is apparently not the ulti
mate reaction form because it does not react with proteins
or nucleic acids directly (21); thus, further activation is
apparently required. There are many observations that sug
gest that the sulfate ester of N-OH-AAF is the ultimate reac
tive form in liver (21), but AAF feeding causes tumors of the
ear duct gland and mammary gland of rat. Yet, both of these
tissues are devoid of the enzymes necessary to activate NOH-AAF to the sulfate ester form (12).
One mechanism that may be of importance in N-OH-AAF
1 This
research
was
in
part
supported
by
Grant
1-RO1-CA18591-01
from
the National Cancer Institute.
°Towhom requests for reprints should be addressed, at the Oklahoma
Medical Research Foundation, Biomembrane Research Laboratory, 825
Northeast 13 Street, Oklahoma City, OkIa. 73104.
3 The
abbreviations
used
are:
AAF,
N-acetyl-2-aminofluorene;
N-OH-AAF,
N-hydroxy-N-acetyl-2-aminofluorene; ESR, electron spin resonance; LAHP,
linoleic acid hydroperoxide; NOF, 2-nitrosofluorene; N-OAC-AAF, N-acetoxy
N-acetyl-2-aminofluorene; TLC , thin-layer chromatography.
Received December 9, 1975; accepted April 12, 1976.
AUGUST
activation is the peroxidase or free radical activation route
(7). The initial observations that provided a basis for consid
ering this mechanism were made independently by Bartsch
et a!. (4) and Forrester et a!. (8), first in a purely organic
chemical system and laterextrapolatedto in vitroperoxi
dase enzyme systems by Bartsch et al. (2, 3). The molecular
model for this activation system is discussed at length by
Bartsch and Hecker (2). Briefly, it postulates that N-OH-AAF
acts as a 1-electron donor to the hydrogen peroxide-in
duced cycling peroxidase. The nitroxyl free radicals formed
dismutate to form the nitrosofluorene and N-acetoxyace
tylaminofluorene carcinogens . The horserad ish peroxidase
system has been studied in greater detail, and we (7) have
concluded that the model proposed by Bartsch and Hecker
(2) is correct; however, the peroxidase has some peculiar
properties in the presence of the Carcinogen, cyanide does
not inhibit the reaction (7), as was observed by King et al.
(13).
The free radical activation route is an attractive possibil
ity, and the observation of Stier et al. (20), that the chloro
form/methanol extract of microsomes metabolizing arylam
me carcinogens yields nitroxyl free radical ESR signals,
emphasizes the need to consider it as a possible in vivo
activation mechanism. It is possible that lipid peroxides and
an endogenous peroxidase would act in a manner that
would activate N-OH-AAF. Preliminary work using heme
proteins and various hydroperoxides was encouraging;
therefore, we have tested these ideas in a strictly in vitro
system using LAHP as the peroxide and methemoglobin or
hematin as the catalyst. The results reported here do mdi
cate that this system will activate N-OH-AAF through a ni
troxylfree radicalintermediateto form the carcinogens
nitrosofluorene
and N-acetoxyacetylaminofluorene
. Ascor
bate inhibits
thisactivationmechanism.
MATERIALS AND METHODS
Hematin (prepared from bovine blood), methemoglobmn
Grade I (prepared from crystallized bovine blood hemoglo
bin), lipoxygenase Grade I (prepared from soybeans), and
linoleic acid Grade I (99% purity) were purchased from
Sigma Chemical Co. , St. Louis, Mo. AAF and 2-nitrofluo
rene were purchased from Aldrich Chemical Co. , Milwau
kee, Wis. LAHP was prepared from linoleic acid using soy
bean lipoxygenase according to the procedure of Hamberg
and Gotthammar (10). The prepared LAHP was actually the
13-L-hydroperoxy-cis-9,trans-i 1-octadecadienoic acid iso
mer (10). The prepared LAHP was stored in methanol under
1976
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.
2761
A. A. Floyd et a!.
nitrogen at —20°.
An absorbance ofe = 25.25 m@i1 cm1 at
233 nm in 95% ethanol was used to assay the amount of
LAHP present.
N-OH-AAF was synthesized
from 2-nitrofluorene
and re
crystallized from benzene according to the procedure of
Poirier et a!. (17). The melting point of the synthesized
N-OH-AAF was 146°,which is within the desirable range
(146-147°) according
to the results of Poirieret
al. (17). The
absorptivity in buffer was used as a measure of the amount
of N-OH-AAF present as we have described previously (7).
The carcinogen was solubilized in methanol (1 mg/mI) and
was added as such to the reactant solutions. N-Hydroxy-2aminofluorene was prepared according to the procedure of
Lotlikar et a!. (14). NOF was prepared by oxidation of the Nhydroxy-2-aminofluorene with diethyl azodicarboxylate (1).
The reaction mixture was purified by silicic acid chromatog
raphy using N-hexane/benzene (10/3, v/v) as the solvent
system. The 1st green band emerging from the silicic acid
column was crystallized. The absorption spectrum in 95%
ethanol demonstrated a maximum at 362 nm, a peak of 244
nm, a shoulder at 260 nm, and minima at 278 and 224 nm.
This spectrum agrees with that reported for NOF by Lotlikar
et al. (14). The absorptivity at 362 nm as given by Lotlikar
et al. (14) was used as a measure of the amount of NOF
present. N-OAC-AAF was synthesized according to the pro
cedure of Gutmann and Erickson (9). The absorption peaks
and their ratios matched those reported by Gutmann and
Erickson (9). Also, our synthesized N-OAC-AAF had absorp
tivities and TLC patterns identical to those of a sample of
N-OAC-AAF generously supplied by Dr. E. K. Weisburger
(National Cancer Institute, Bethesda, Md.).
Incubation of the horserad ish peroxidase/N-OH-AAF/
H2O2system for TLC of the products
was carried out accord
ing to the method of Bartsch and Hecker (2). Extraction of
the incubation mixture was with cold dichloromethane as
described by Bartsch and Hecker (2). TLC ofthe extract was
carried out using Silica Gel 60 F-254 absorbed on aluminum
sheets (Merck, Darmstadt, Germany) as we have described
previously (7). The solvent system used was dichlorometh
ane/acetone (85/5, v/v). In the hematin/N-OH-AAF/LAHP
system, incubations were the same as those described for
the optical and ESR studies, except that larger volumes
were used. Extraction and TLC of this system were con
ducted as with the peroxidase system (7). Hematin stock
solution was made up by dissolving ox blood hematin in
glass-distilled H20 to which had been added a trace of alkali
as NH4OHor NaOH. The hematin stock solution was stored
at 4°or —20°
until used. The catalytic capacity of the hema
tin decreased slightly with time, apparently due to the p0lymerization of the hematin (11). This affected only the rate
of reaction, not the pattern of the kinetics or the nature of
the products. An absorptivity of 122 mM1 cm1 at 398 nm
was used as a measure of the amount of hematin present
(ii).
Optical studies were carried out with a Gary Model 14
recording spectrophotometer. The reactions (ESR) were
conducted at room temperature, which averaged 25°.ESR
spectra were taken with a Varian E-9 X-band spectrometer
operating at 100 kHz modulation. A 3-port stop-cock di
rectly coupled 2 syringes and directly connected the sy
2762
ringes to the liquid sample cuvet positioned in the ESR
cavity. This system made it possible to inject a sample and
start an ESR sweep in less than 20 sec after adding the
reactants.
RESULTS
Chart 1 demonstrates the free radical signal obtained with
methemoglobin- and hematin-catalyzed oxidation of N-OH
AAF by LAHP. We compute a g value of 2.0063 forthe free
radical observed in both reactions. This value is the same as
the one we have obtained in the horseradish peroxidase/N
OH-AAF/H202
system
and the same as that
reported
by
Bartsch and Hecker (2) for the latter system. The g value of
the lower spectrum in Chart 1 appears to be slightly lower,
but this is due to a slight shift in frequency between the 2
samples. Many separate experiments in which potassium
nitrosodisülfonate was used as a standard have shown that
the g values of the free radical intermediates in both reac
tions are the same. The 14Nhyperfine splitting is 7.7 gauss.
This value agrees with that reported by Bartsch and Hecker
(2) in the horseradish
peroxidase
system.
Thus,
for the
above reasons, we conclude that the free radical signal we
obtain in either the methemoglobin- or hematmn-catalyzed
oxidation of N-OH-AAF by LAHP is the nitroxyl free radical
of N-OH-AAF. The peak heights decrease from low field to
high field (left to right) because the amount of free radical is
decreasing with time (see below). Ifeitherthe carcinogen or
the LAHP is left out of the reaction mixture, no signal is
obtained. The LAHP must be added last in order to obtain a
nitroxyl free radical signal.
Chart 2 shows the optical difference spectrum of methe
moglobin and of hematin plus N-OH-AAF before and after
Chart 1. ESR spectra of the hematin- and methemoglobin-catalyzed oxi
dation of N-OH-AAF by LAHP. Top spectrum, methemoglobin-catalyzed reac
tion, middle spectrum, hematin-catalyzed reaction. The concentrations of
methemoglobin, N-OH-AAF, and LAHP were 3.6, 81, and 45 MM, respectively,
and the concentrations of hematin, N-OH-AAF, and LAHP were 4.0, 41, and
25 @M,
respectively. The temperature was approximately 25°.The spectra
were recorded at 25 gauss/mm with a filter constant of 3 sec. The microwave
frequency was 9.529 GHz and the modulation amplitude was 5 gauss at a
frequency of 100 KHz. Bottom spectrum, potassium nitrosodisulfonate
standard.
CANCER
RESEARCH
VOL. 36
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.
@
@-
Lipid Peroxide Activation of Carcinogens
ment are shown in Chart 3. There is a large and rapid
monotonic absorbance decrease at 233 nm which is accom
panied by a monotonic 368 nm absorbance increase. As the
reaction proceeds, there is an absorbance decrease at 301
nm, but this decrease has 2 phases. The results of another
kinetic experiment are presented in Chart 4. The semiloga
rithmic plot of the absorbance changes illustrates these
results clearly. The 233 nm absorbance decrease and 368
A
368nm
5.01
260
@Ju
@
juu
4W
560
360
4@lOmn
ôôo
lóOnm
Chart 2. The optical spectra of methemoglobin or hematin plus N-OH-AAF
before and after LAHP addition. A. Trace I, difference spectrum of 3.6 @zM
of
methemoglobin in 0.05 M potassium phosphate buffer, pH 7.4, to which NOH-AAF was added to the sample cuvet to a final concentration of 40 @M;
Trace II, spectrum after LAHP to a final concentration of 42 .tMwas added to
the sample cuvet. B. Trace I, difference spectrum of [email protected]
hematin in potas
sium phosphate buffer to which N-OH-AAF was added to the sample cuvet to
a final concentration of 34
@;
Trace II, spectrum after LAHP (30 @M)
was
added to the sample cuvet.
LAHP addition. Carcinogen addition to the solutions con
taming the heme compounds yields a spectrum similar to
that obtained after carcinogen addition to buffer solution
only. There is a sharp 301 nm peak, a shoulder at 289 nm, a
slight shoulder at 281 nm, and a maximum at 273 nm. This
spectrum is very similar to that obtained after N-OH-AAF
addition to methanol, exceptthatthe maximum in methanol
is 281 nm (7). In separate experiments we have found that
the phosphate buffer shifts the maximum to 273 nm. LAHP
addition to the solutions containing the heme compounds
plus carcinogen caused a decrease in absorbance at 301,
289, and 273 nm, but there is a concomitant
30@nm
0
4
;3
2
1
0
mm
Chart 3. The kinetics of the 233, 301, and 368 nm absorbance changes of
LAHP added to a solution containing hematin plus N-OH-AAF. The hematin
and N-OH-AAF were added to potassium phosphate buffer (0.05 M, pH 7.4) to
a final concentration of 5 and 35 @M,
respectively. The carcinogen was added
to the sample cuvet only, but hematin was in both sample and reference
cuvets. LAHP to a final concentration of 30 MMwas then added to the sample
cuvet only, and the time course was recorded at the appropriate wavelength
starting 20 sec after LAHP addition. The slit width was no more than 1 nm in
any of the measurements, and the temperature was approximately 27*.
increase of a
broad absorbance centered at 368 nm. This absorbance
increase is due to nitrosofluorene accumulation (see below)
which absorbs maximally in this region (4). LAHP addition
also causes absorbance alterations at about 410 and 420 nm
as seen in the methemoglobin solutions (Chart 2A). How
ever, in separate experiments (not shown here), these ab
sorbance changes can be accounted for by the action of
LAHP only on the heme compound.
Similar changes of
absorbance were observed aftercumene hydroperoxidead
dition to microsomes (6).
LAHP absorbs maximally
at 233 nm, and this occurs at a
position such that absorbance from either the heme com
pounds or N-OH-AAF does not interfere greatly (see Chart
2). We have conducted
kinetic
experiments
on the LAHP/
hematin/N-OH-AAF system where LAHP disappearance and
NOF appearance were determined by monitoring
absorb
ance changes at233 and 368 nm, respectively.
Absorbance
changes at 301 nm were also determined as the reaction
proceeded, but the interpretation of these absorbance
changes is compounded in the sense that, in addition to NOH-AAF,
N-OAC-AAF
also absorbs
strongly
at this wave
length (9). We have conducted these kinetic experiments
and the remainder of the studies reported herein using
hematin as a catalyst. The results of a typical kinetic experi
AUGUST 1976
tim.(s.c)
Chart 4. The semilogarithmic plot of the kinetics of the relative absorb
ance changes at 233, 301, and 368 nm following LAHP addition to a solution
containing hematin plus N-OH-AAF. The experiments were carried out ex
actly as described in Chart 3 except that the concentrations of hematin, NOH-AAF, and LAHP were 5, 37. and 30 @M,
respectively. The scale for the 301
nm change is 10 times more sensitive than that for the 233 and 368 nm
absorbance
changes.
2763
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.
A. A. Floyd et al.
nm absorbance increase are kinetically equivalent, with a
half-time in this particular reaction of 25 and 26 sec, respec
tively. The 30i nm decrease can be broken into 2 phases.
The fast phase is kinetically equivalent to the 233 and 368
nm absorbance changes, with a half-time of 24 sec. How
ever, there is a slow phase that continues after the fast
phase is complete and that has a half-time of 186 sec in the
experiment presented in Chart 4. We have conducted many
similar experiments, and they all have exhibited this same
behavior kinetically. Thus, it is clear that the 233 nm absorb
ance decrease corresponds kinetically both to the 368 nm
absorbance increase and to the fast phase of the 301 nm
absorbance decrease. The total absorbance changes at
I
.1
233, 301, and 368 nm in the 1st 2 mm of the reaction
presented in Chart 3 are 0.13, 0.023, and 0.052, respec
tively. After 2 mm of reaction, there was a slight 233 nm
absorbance increase at the same time that there was a slow
but steady 301 nm decrease which continued for as long as
15 to 20 mm. These slow absorbance
changes
were ex
cluded from the calculations presented above and will be
discussed later.
Chart 5 shows that ascorbate inhibits the LAHP-induced
301 and 368 nm absorbance
changes
in solutions
contain
ing hematin plus N-OH-AAF. The lowest trace is the spec
trum of hematin plus N-OH-AAF versus hematin. The upper
trace is the spectrum after ascorbate addition to the hema
tin plus N-OH-AAF solution only. Ascorbate addition re
suIted in an increased absorbance at 265 nm as would be
expected. After LAHP addition (middle trace) there was a
large decrease in ascorbate, but N-OH-AAF was not con
sumed and, in addition, nitrosofluorene was not formed
(see below). There was a disproportionate amount of ascor
bate consumed to LAHP added (approximately 12/i). In
separate experiments (not shown) in which we monitored
the kinetics of these components, ascorbate oxidation did
not follow kinetically LAHP consumption except for only a
small initial component. After the 1st 2 mm, ascorbate oxi
dation proceeded in a somewhat linear fashion for 8 to 10
mm after LAHP consumption had ceased. It appears as if
the catalytic decomposition of LAHP triggered ascorbate
oxidation.
The ascorbate radical was present ratherthan the nitroxyl
radical during ascorbate prevention of N-OH-AAF oxidation.
This is shown clearly in Chart 6. Trace A shows the presence
of the nitroxyl free radical of N-OH-AAF during the hematin
calalyzed, LAHP-induced oxidation of N-OH-AAF. When
ascorbate was present, the nitroxyl radical was not present,
but there was a radical present with a g value of 2.005 (Chart
6, Trace C) which
split into a doublet
(Chart 6, Trace B)
when the modulation amplitude was decreased in order to
resolve the hyperfine structure. The g value and splitting
constant (1.7 to 2.0 gauss) fit the properties of the ascor
bate radical (18). The nitroxyl radical of N-OH-AAF was
monitored on a kinetic basis (Chart 6, Trace D) in the LAHP/
hematin/N-OH-AAF systems. Analysis of the time course of
the nitroxyl radical disappearance in these reactions re
vealed that in most cases the decay curves were usually
biphasic, the kinetics of which did not correlate rigidly with
the optical absorption changes discussed previously
(Charts 3 and 4). Chart 6, Traces E and F illustrates that,
2764
2@t
360
4ô0
5ô0
6ôOnm
Chart 5. Optical difference spectra of the effect of LAHP addition to a
solution of hematin plus N-OH-AAF containing ascorbate. The sample and
reference cuvets contained 5 MM hematin in potassium phosphate buffer.
Bottom trace, spectrum after N-OH-AAF was added to a final concentration of
36 MMto the sample cuvet. Top trace, spectrum after ascorbate addition (61
@.tM);middle
trace,
spectrum
after LAHP
(5
@M)was then added
to the sample
cuvet.
when ascorbate was present in the system, the ascorbate
radical was present in the early phases of the reaction at
which time the nitroxyl radical was absent. However, after
the ascorbate radical disappeared, the nitroxyl radical ap
peared. Trace G shows that a small amount of nitroxyl
radical was present when as small a concentration of LAHP
as 1.8 j.@M
was added to the system.
Fig. 1 presents the results of TLC of the hematin-cata
lyzed LAHP oxidation of N-OH-AAF. The results show that
NOF and N-OAC-AAF are produced
in the hematin-catalyzed
reaction as is true for the horseradish peroxidase/H2O2catalyzed oxidation of N-OH-AAF. The horseradish peroxi
dase reaction system yielded 2 major products correspond
ing to N-OAC-AAF and NOF (Fig. 1A, Lane 5). This result
agrees with that observed by Bartsch and Hecker (2) and
ourselves (7). NOF and N-OAC-AAF are also the products of
the hematin-catalyzed reaction (Fig. 1A, Lane 6). Ascorbate
prevents the formation of these prod ucts as is shown in Fig.
iB , Lane 2. Thus, the TLC experiments corroborate what we
deduced from the optical studies presented earlier (Chart
5). In the hematin-catalyzed reaction, there was some N-OH
AAF
remaining
at
the
termination
of
the
reaction,
but
this
was not true of the peroxidase-catalyzed reaction. There
was a small amount of AAF formed in both the hematin- and
peroxidase-catalyzed reactions. On the basis of visual com
parison of spot intensity there is less N-OAC-AAF than NOF
formed in the hematin-catalyzed reaction. The reaction
shown in Fig. 1 was stopped after 9 mm, but if the reaction
CANCER RESEARCH VOL. 36
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.
Lipid Peroxide Activation of Carcinogens
A
was stopped after only 2.5 mm (experiments not shown),
then the N-OAC-AAF spot was nearly as intense as the NOF
spot. Thus, we conclude that the continued slow decrease
at 301 nm (see Chart 4) is due to a decrease in the amount of
N-OAC-AAF present. The products
B
.
of N-OAC-AAF decompo
sition are not known under our conditions.
r
DISCUSSION
The data presented in this paper demonstrate that a lipid
hydroperoxide in the presence of a heme compound acti
vated N-OH-AAF via the nitroxyl free radical intermediate of
N-OH-AAF into the products
nitrosofluorene
F.
and N-acetoxy
acetylaminofluorene. The reaction was inhibited by ascor
bate, in which case the free radical of N-OH-AAF was re
placed by the ascorbate free radical.
We do not know the detailed mechanism of the reaction
p.
i
..
and, in fact,we are studying itingreaterdetailnow. How
@
ever, there are certain characteristics that the present study
does make clear, namely, that the rate of LAHP disappear
ance parallels the rate of NOF accumulation and that the
2
3456
123
Fig. 1. TLC of the hematin/LAHP-catalyzed N-OH-AAF oxidation reaction
as effected by ascorbate. A. The conditions for each ofthe lanes are: Lane 1,
AAF standard; Lane 2, N-OH-AAF standard; Lane 3, NOF standard; Lane 4, NOAC-AAF standard (the large spot corresponds to N-OAC-AAF, the smaller
spots are breakdown products as described in Ref. 7; Lane 5, horseradish
peroxidase/H202/N-OH-AAF system as described previously (7); Lane 6, hem
atin/LAHP/N-OH-AAF system similar to that described in Chart 3. B. Lane 1
corresponds to AAF standard, Lane 3 corresponds to the NOF, and Lane 2 is
as described in Lane 6 of A , but ascorbate was present as described in Chart
5.
rate of LAHP disappearance parallels the rate of the fast
phase of the 301 nm absorbance decrease, which in large
part is due to N-OH-AAF disappearance. An examination of
the stoichiometry of the reaction also allows certain conclu
sions. In Chart 3 it was noted that the absorbance changes
D
at233, 301 ,and 368 nm were 0.13,0.023,and 0.052 absorb
Chart 6. ESR spectra and kinetics of the hematin plus N-OH-AAF solution
with or without ascorbate present after LAHP addition. Trace A, ESR spec
trum started 20 sec after the addition of LAHP (10 @.tM)
to a phosphate buffer
solution containing 5 @M
hematin and 40 @M
N-OH-AAF. The modulation
amplitude was 5 gauss, microwave frequency 9.529 GHz, scan rate 20 gauss/
mm, filter constant 3 sec. and the receiver gain 4 x 10g. Trace B, ESR
spectrum of a similar solution as in Trace A except that 170 @M
ascorbate was
also present. The conditions are the same as in Trace A except that the
modulation amplitude was 2 gauss. Trace C, ESR spectrum of a solution
exactly as in Trace B except that the modulation amplitude was 5 gauss.
Trace 0, time course of the decay of the nitroxyl free radical monitored by
setting the field at the maximum of the low field transition. The solution and
the conditions used were as in Trace A. Trace E, time course of the decay of
the ascorbate free radical monitored by setting the field at the maximum of
the over-modulated ascorbate radical transition (Trace C). The conditions
otherwise are as in Trace A. The solutions used are as in Traces B and C.
Trace F, time course of the nitroxyl radical in the solutions to which ascor
bate was present. The nitroxyl radical was monitored at the field position as
described in Trace D. The solutions used were exactly as in Traces B, C, and
E. Trace G, time course of the nitroxyl free radical in a phosphate buffer
solution containing 10 @M
hematin and 40 @M
N-OH-AAF to which LAHP was
added to a final concentration of 1.8 @M.
The free radical was monitored as
described in Trace 0; otherwise, the conditions were as described for Trace
A. Downward arrows of Traces 0, E, F, and G, time zero in the reaction;
upward arrows, time when reaction mixture was injected into the cavity.
AUGUST 1976
ance units, respectively. On the basis of the known absorb
ance of the compounds, we calculated that for every 2 LAHP
molecules consumed 1 NOF molecule accumulated. The
absorbance change at 301 nm is more complicated since NOAC-AAF absorbs stronger than N-OH-AAF at this wave
length and since NOF would contribute some absorbance at
this wavelength. If we presume that for every 2 LAHP mole
cules decomposed, 2 N-OH-AAF molecules are oxidized
which yield 1 N-OAC-AAF and 1 NOF molecule, then using
the known absorptivities of N-OH-AAF, N-OAC-AAF, and
NOF at 301 nm we can account for0.0189 ofthe observed
0.023 (or 82%) of the absorbance change at 301 nm, by the
above mechanism. As was noted previously, N-OAC-AAF is
decreasing with time and, since the absorbance changes
were calculated at 2 mm, we would have overestimated the
N-OAC-AAF contribution;
thus, this would increase the cal
culated 82% value closer to the expected 100%. On the
basis of the above considerations, we postulate that the
heme compound causes LAHP to decompose; either the
decomposition products or the decomposition process per
se oxidizes N-OH-AAF into the nitroxyl free radical of N-OH
AAF.
Two
of
these
radicals
then
form
NOF
and
N-OAC-AAF,
perhaps by a dismutation process. The postulated reaction
is presented in Chart 7. The reaction as written is not com
plete, for another product, perhaps a lipid alcohol, should
2765
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.
A. A. Floyd et al.
@-H H H
c5@'
C—C—'C—(CH2)7—COOH
‘I
2 CH3—(CH2)4—C—C
H
@
@
CH@1
1
I
NOF
LAHP
+
@—
hematin ,. 2L@J@J3@.
_J —@
+
0
I
1
2(@i―@T'@T'@
—C—CH3
0
@@@yN—C—CH3
O—C—CH3
0
N—OH—AAF
N —OAC—AAF
Chart 7. Proposed reaction scheme for the LAHP/hematin-catalyzed activation of N-OH-AAF into N-OAC-AAF and NOF.
also be represented. We are currently investigating the re
action mechanism in greaterdetail. The important point, as
far as carciiogenesis is concerned, is that more active
carcinogens are formed in the reaction.
It cannot be stated unequivocally on the basis of the
present results that the reaction is proceeding via an obliga
tory nitroxyl free radical intermediate; however, all the ob
servations presented here certainly indicate strongly that
this is true. The lack of rigid correlation between the rate of
nitroxyl radical decrease and the rate of N-OH-AAF disap
pearance and NOF appearance certainly does not rule out
the obligatory free radical intermediate concept. The
amount of the nitroxyl free radical present at any time would
depend on the ratio of its rate of formation to its rate of
decay. It would be expected that the ratio of these 2 rates
would not necessarily remain constant as the reaction pro
ceeded and the substrates became depleted. Also, it is
possible that N-OAC-AAF may decompose via the nitroxyl
free radical form.
Bartsch and Hecker (2) discovered that horseradish per
oxidase plus H2O2would activate N-OH-AAF into nitrosoflu
orene and N-acetoxyacetylaminofluorene.
However, in an
attempt to extrapolate their horseradish peroxidase obser
vations, Bartsch and Hecker (2) were unable to recover Nacetoxyacetylaminofluorene or were unable to observe the
nitroxyl free radical in lipid peroxidizing microsomes; yet
Stier et al. (20) did obtain nitroxyl free radical spectra in
chloroform/methanol extracts of microsomes metabolizing
arylamine carcinogens. The fact that we have observed the
nitroxyl free radical with as small a concentration of lipid
hydroperoxide as 1.8 @M
does force a consideration of the
activation of N-OH-AAF by lipid peroxidation products when
a heme protein such as hematin or methemoglobin is pres
ent. We have also observed that ascorbate inhibits the lipid
hydroperoxide plus heme compound activation of N-OH
AAF.
This
was
also
true
of
the
horseradish
peroxidase/H202
activation of N-OH-AAF (7). From our results alone, it is not
possible to determine the action of ascorbate in the system.
We speculate that ascorbate is reducing the nitroxyl free
radical back to N-OH-AAF, since N-OH-AAF was not oxi
dized when ascorbate was present (see Chart 5). However,
ascorbate involvement in the reduction state of the heme
cannot be ruled out. Ascorbate radical was not present
when ascorbate was added to hematin solution per se. This
observation would tend to exclude the oxygen scavenger
action of ascorbate here.
On the basis of our results, a protective effect for ascor
2766
bate against AAF carcinogenesis is implicated. Perhaps
other free radical scavengers such as vitamin E and other
antioxidants would inhibit the reaction described herein.
Many observations indicate that antioxidants do exert an
inhibiting influence on carcinogenesis; conversely, lipid
peroxidation may enhance carcinogenesis (see Ref. 19 and
references therein). The reaction we describe here does
provide a theoretical model to support these observations,
especially in AAF carcinogenesis. More detailed studies,
now underway, are necessary before generalizations can be
drawn.
ACKNOWLEDGMENTS
I would like to thank Dr. Helmut Bartsch for suggesting the use of diethyl
azodicarboxylate
as an oxidizing
agent to form nitrosofluorene
and Dr. E. K.
Weisburger for sending us a sample of N-OAC-AAF.
REFERENCES
1. Ames, B. N., Gurney, E. G. , Miller, J. A., and Bartsch, H. Carcinogens as
Frameshift Mutagens: Metabolites and Derivatives of 2-Acetylaminofluo
rene and Other Aromatic Amine Carcinogens. Proc. NatI. Acad. Sci. U..
S., 69: 3128—3132,
1972.
2. Bartsch,H., and Hecker,E. On the MetabolicActivationof the Carcino
gen N-Hydroxy-N-2-acetylaminofluorene. Ill. Oxidation with Horseradish
Peroxidase to Yield 2-Nitrosofluorene and N-Acetoxy-N-2-acetylamino
fluorene. Biochim. Biophys. Acta, 237: 567-578, 1971.
3. Bartsch,H., Miller,J. A., and Miller, E. C. N-Acetoxy-N-acetylaminofluo
rene and Nitrosoarenes, One-Electron Non-Enzymatic and Enzymatic
Oxidation Products of Various Carcinogenic Aromatic Acethyroxamic
Acids. Biochim. Biophys. Acta, 273: 40—51
, 1972.
4. Bartsch, H., Traut, M. , and Hecker, E. On the Metabolic Activation of NHydroxy-N-2-acetylaminofluorene. I. Simultaneous Formations of 2-Ni
trosofluorene and N-Acetoxy-N-2-acetylaminofluorene from N-Hydroxy
N-2-acetylaminofluorene via a Free Radical Intermediate. Biochim. Bio
phys. Acta, 237: 556—566,
1971.
5. Cramer, J. W., Miller, J. A., and Miller, E. C. N-Hydroxylation: A New
Metabolic Reaction Observed in the Rat with the Carcinogen 2-Acetyl
aminofluorene. J. Biol. Chem., 235: 885-888, 1960.
6. Floyd,R. A. MicrosomeCatalyzedConversionof N-Hydroxy-N-Acetyl-2Aminofluorene by Cumene-Hydroperoxide. Life Sci., 18: 189-196, 1976.
7. Floyd, R. A., Soong. L. M., and Culver, P. L. Horseradish Peroxidase/
Hydrogen Peroxide-catalyzed Oxidation of the Carcinogen N-Hydroxy-N
acetyl-2-aminofluorene as Effected by Cyanide and Ascorbate. Cancer
Res.,
36:1510-1519,
1976.
8. Forrester, A. A., Ogilvy, M. M. , and Thomson, R. H. Mode of Action of
Carcinogenic Amines. I. Oxidation ofN-Aryl-hydroxamic Acids. J. Chem.
Soc., (C): 1081-1083, 1970.
9. Gutmann, H. R., and Erickson, R. R. The Conversion of the Carcinogen
N-Hydroxy-2-fluorenylacetamide to o-Amidophenols by Rat Liver in Vi
tro. An Inducible Enzymatic Reaction. J. Biol. Chem., 244: 1729-1740,
1969.
10. Hamberg, M., and Gotthammar, B. A New Reaction of Unsaturated Fatty
Acid Hydroperoxides: Formation of 11-Hydroxy-1 2 ,13-epoxy-9-octade
cenoic Acid from 13-Hydroperoxy-9.11-octadecadienoic Acid. Lipids, 8:
737-744, 1973.
CANCER
RESEARCH
VOL. 36
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.
Lipid Peroxide Activation of Carcinogens
11. Inada, Y. , and Shibata, K. The Soret Band of Monomeric Hematin and Its
Changes on Polymerization. Biochem. Biophys. Res. Commun. , 9: 323327, 1962.
12. Irving, C. C., Janss, D. H., and Russell, L. T. Lack of N-Hydroxy-2acetylaminofluorene Sulfotransferase Activity in the Mammary Gland
and zymbal's Gland of the Rat. Cancer Res., 31: 387-391 . 1971.
13. King. C. M., Bednar, T. W., and Linsmaier-Bednar, E. M. Activation of
the Carcinogen N-Hydroxy-2-fluorenylacetamide: Insensitivity to Cya
nide and Sulfide of the Peroxidase-H2O2Induced Formation of Nucleic
Acid Adducts. Chem. Biol. Interactions, 7: 185-188, 1973.
14. Lotlikar, P. D. . Miller, E. C., Miller, J. A. . and Margreth. A. The Enzymatic
Reduction of the N-Hydroxy Derivatives of 2-Acetylaminofluorene and
Related Carcinogens by Tissue Preparations. Cancer Res., 25: 17431752, 1965.
15. Miller, J. A. Carcinogenesis by Chemicals: An Overview —
G. H. A. Clowes
Memorial Lecture. Cancer Res.. 30: 559-576, 1970.
16. Miller, J. A., and Miller, E. C. Guest Editoiral—Chemical Carcinogenesis:
AUGUST 1976
Mechanisms and Approaches to Its Control. J. NatI. Cancer Inst. , 47: 514, 1971.
17. Poirier, L. A., Miller, J. A., and Miller, E. C. The N- and Ring-Hydroxyl
ation of 2-Aminofluorene in the Dog. Cancer Res.. 23: 790-800, 1963.
18. Ruf, H. E., and Weis, W. Hyperfeinstruktur des ESR-Spectrums von
Semidehyd ro-L(+)-Ascorbins@ure. Biochim . Biophys. Acta , 261: 339340, 1972.
19. Shamberger. R. J.. Andreone, T. L., and Willis, C. E. Antioxidants and
Cancer. IV. Initiating Activity of Malonaldehyde as a Carcinogen. J. NatI.
Cancer Inst., 53: 1771-1773, 1974.
20. Stier, A. . Reitz, I., and Sackmann, E. Radical Accumulation in Liver
Microsomal Membranes during Biotransformation of Aromatic Amines
and Nitro Compounds. Naunyn-Schmiedebergs Arch. Exptl. Pathol.
Pharmakol., 274: 189-191, 1972.
21. Weisburger, J. A., and Weisburger, E. K. Biochemical Formation and
Pharmacological. Toxicological. and Pathological Properties of Hydrox
ylamines and Hydroxamic Acids. Pharmacol. Rev., 25: 1-66, 1973.
2767
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.
Lipid Hydroperoxide Activation of N-Hydroxy-N
-acetylaminofluorene via a Free Radical Route
Robert A. Floyd, Lailing M. Soong, Robert N. Walker, et al.
Cancer Res 1976;36:2761-2767.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/36/8/2761
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1976 American Association for Cancer Research.