Lipid Peroxyl Radicals from Oxidized Oils and Heme-Iron ...Nippon Oils & Fats Co. (Tokyo, Japan). Olive oils were ob-tamed from Eleourgiki and others in markets in Athens, Greece

Tomohiro Sawa, Takaaki Akaike, Kenji Kida,Yukari Fukushima, Koichi Takagi, and Hiroshi Maeda2

Department of Microbiology, Kumamoto University School of Medicine,

Honjo 2-2-1, Kumamoto 860 [T. S., T. A., H. M.]; Department of AppliedChemistry, Faculty of Engineering, Kumamoto University, Kurokami 2-40-1,

Kumamoto 860 [K. K.]; and Coloproctology Center, Takano Hospital,

Kumamoto 862 [Y. F., K. T.], Japan

Vol. 7, 1007-1012, November 1998 Cancer Epidemiology, Biomarkers & Prevention 1007

Lipid Peroxyl Radicals from Oxidized Oils and Heme-Iron: Implication

of a High-Fat Diet in Colon Carcinogenesis1

Abstract

A diet high in fat and iron is known as a risk factor incancer epidemiology. However, the details of themolecular mechanism remains to be elucidated. Weexamined the possible implication of lipid peroxybradicals generated from fatty acids and heme-iron inDNA damage, and hence in the possibility of coloncancer. F344 female rats were given N-nitroso-N-methyburea six times during a 2-week period and then feddiets containing different amounts of safflower oil andhemoglobin (rich in iron) for 36 weeks; the occurrence of

colon cancer was determined by H&E staining. In thisanimal model, simultaneous feeding of a fat diet andheme-iron produced a significant increase (P < 0.05) inthe incidence of colon cancer compared with a dietwithout hemoglobin. Electron paramagnetic resonance

and chemibuminescence studies revealed that oxidizedrefined vegetable oils, particularly safflower oil, readily

generated lipid peroxyl radicals in the presence of variousheme compounds, and the peroxyl radicals did effectivelycleave DNA. Unpurified native vegetable oils contain ahigh amount of peroxyl radical scavengers, whereasconventional refining processes seem to reduce the levelsof many valuable anti-peroxyl radical compoundsabundant in plant seeds. In conclusion, lipid peroxidesand heme components generate peroxyb radical speciesthat exert DNA-cleaving activity. A plausible explanationis that lipid peroxyl radicals thus generated, whichoriginated from routine dietary components such as fatand red meat, may contribute, at least in part, to thehigh incidence of colon cancer.

Introduction

Epidemiobogical (1-3) and experimental (4, 5) data suggest that

diets high in fat, especially PUFAs3 (2), and red meat are riskfactors contributing to breast, colon, ovarian, prostate, and othercancers. Red meat usually contains a large amount of heme-iron, primarily in myogbobin, together with fat. Although body

stores and intake of iron are also reported to show a positiveassociation with cancer incidence (6-8), the effects of thesimultaneous intake of fatty acids and heme-iron remain to be

fully elucidated at molecular and in vivo levels.It is well known that PUFAs are spontaneously oxidized in

air; they are then called the LOOH forms of PUFAs (9). We

reported previously that a reaction of a water-soluble model

compound of LOOH, i.e.,t-BuOOH, with heme-iron generatedpredominantly alkylperoxyl radicals (ROO) and that only

ROO, not the alkoxyl (RO) or alkyl (R’) radicals, had potentbactericidal activity (10, 1 1). Van der Zee et a!. (12) reportedthe biological effects of ROO, which inactivated several en-zymes and induced potassium leakage from RBCs. From thesefindings, we hypothesized that simultaneous intake of fatty

acids and heme-iron might cause generation of lipid peroxylradicals (LOO) in the intestinal tract and promote cobonic

carcinogenesis. The bactericidal activity of LOO ( 10, 1 1) mayindicate an additional cytotoxic effect on endothebiab cells aswell as damage to DNA, thus enhancing carcinogenic potential

by induction of new cell proliferation as a repair process (13).Because of the longer half-life of LOO, which was estimatedto be >30 mm by means of spin trap EPR spectroscopy (10),compared with �OH or O� radicals, the biological effect of

LOO generation in situ may be more important than we an-ticipated previously. A more recent study showed that thehydroperoxide of diacylglycerol, and perhaps other lipid per-oxides as well, activates polymorphonuclear cells to generate

O� (14) by a mechanism similar to that of phorbob 12-myristate13-acetate, a hallmark compound among the tumor promoters.

To examine the carcinogenic and/or tumor-promoting potential

of LOO, we investigated giving diets simultaneously high inheme-iron (hemoglobin) and fat to rats preconditioned with

N-nitroso-N-methylurea at low doses and assessed the effect oncancer incidence. We also examined the effect on DNA damageby LOO, as well as the LOO�-scavenging and antioxidative

potential of vegetable oils at different refining steps.

Received 4/2/98; revised 8/7/98; accepted 817/98.

The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked advertisement in

accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

I Supported in part by a Grant-in-Aid (Special Category for Promotion of Health

No. 225) from the Ministry of Agriculture, Fishery and Forestry of Japan,

1996-1997, and the Nestle Science Promotion Committee, Tokyo, Japan, 1997-

1998.

2 To whom requests for reprints should be addressed, at Department of Micro-

biology, Kumamoto University School of Medicine, Honjo 2-2-1, Kumamoto

860, Japan. Phone: 81-96-373-5098; Fax: 81-96-362-8362; E-mail: msmaedah@

gpo.kumamoto-u.ac.jp.

Materials and Methods

Chemicals. N-Nitroso-N-methylurea, hemoglobin, myogbobin,

hematin, t-BuOOH, Cu:Zn-superoxide dismutase, and deferox-

3 The abbreviations used are: PUFA, polyunsaturated fatty acid: LOOH, lipid

hydroperoxide; t-BuOOH, tert-butylhydroperoxide; ROO’, alkylperoxyl radical;

RO’, alkoxyl radical; R’, alkyl radical; LOO’, lipid peroxyl radical: EPR. electron

paramagnetic resonance; DMPO. 5,5-dimethyl-l-pyrroline-N-oxide; DTPA, di-

ethylenetriaminepentaacetic acid; SSB, single-strand breakage.

Association for Cancer Research. by guest on October 6, 2020. Copyright 1998 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

https://bloodcancerdiscov.aacrjournals.org

1008 LipId Peroxyl Radicals in Carcinogenesis

Table 1 Effect of dietary fats and hemoglobin on colon cancer

F344 rats (female. 7 weeks of age) were administered 2 mg of N-nitroso-N-methylurea intrarectally every other day six times during the initial 2 weeks. The rats were

fed a diet containing the indicated amounts of safflower oil and hemoglobin for 36 weeks. At the end of experiments, the occurrence of colon cancer was determined by

conventional H&E staining. The protein content in the feed of the control and no iron groups was adjusted by addition of soy protein to match the protein content of the

hemoglobin in the diets of the high iron groups.

G Safflower oil Hemoglobin No. of No. of tumor- No. of carcinoma- Total no. of Total no. ofroup � % rats/group positive rats (%) positive rats (%) adenomas adenocarcinomas

Control 5 26 7 (26.9) 4 (15.4) 3 4

High fat + no iron 12 25 8 (32.0) 7 (28.0) 1 9

Fat + high iron 5 3 26 14 (53.8)” 10 (38.5)” 5 16

Fat + high iron” 5 3 25 1 1 (44.0) 8 (32.0) 9 4

“ Significant difference from control group at P < 0.05.S Rooibos tea (1%) was given as drinking water replacement.

amine were obtained from Sigma Chemical Co. (St. Louis,

MO). Ferrous sulfate, mannitol, and protocatechuic acid werepurchased from Wako Pure Chemicals Industries (Osaka,Japan). Linobeic acid and palmitobeic acid were obtained from

Nippon Oils & Fats Co. (Tokyo, Japan). Olive oils were ob-

tamed from Eleourgiki and others in markets in Athens, Greece.Rapeseed oils and safflower oils from different refining pro-

cesses were obtained from Showa Sangyo, Ltd. (Funabashi,Chiba, Japan). The rooibos tea (Aspa!athus !inea!is) was fromSouth Africa, and the five-star grade was found to contain ahigh lipid radical-scavenging capacity in the t-BuOOHIheme-

iron system (1 1). DMPO, a spin trap agent in EPR measure-ment, and the metal chelator DTPA were from Dojindo Labo-ratories (Kumamoto, Japan). Luminob was obtained fromAldrich Chemical Co. (Milwaukee, WI). Supercoiled plasmid

DNA (pUCl9) was prepared from Escherichia co!i ATCC25922 strain bearing the plasmid DNA according to the biter-ature (15).

Tumor Study. All animal experiments were carried out ac-cording to the guidelines of the Laboratory Protocol of AnimalHandling, Kumamoto University School of Medicine. F344 rats

(female, 7 weeks of age) were administered 2 mg of N-nitroso-N-methylurea intrarectabby every other day six times during the

initial 2 weeks (total, 12 mg). MB-l (Funabashi Farm, Inc.) wasused as a basal diet. The rats were fed diets containing theindicated amounts of safflower oil and hemoglobin for 36

weeks. The total cabories/g of food in each diet group wasadjusted to be the same by addition of corn starch to the feed ofthe control group as described in Table 1. Antioxidant compo-

nents present in the feed were vitamin C (30 mg/lOO-g diet),vitamin E (15 mg/b00-g diet), and vitamin A (1200 IU/l00-g

diet) in each group. The basal diet contained -20% of protein.

In addition to the basal protein, 3% of hemoglobin was addedin the group for the heme-iron diet. To adjust protein content,soy protein was added to the non-heme-iron diet. Safflower oilfrom a commercial source, which contained -9.4% saturatedfatty acids, 12.7% monounsaturated fatty acid (oleic acid),70.5% binobeic acid, and 2.0% other PUFAs, was used as the fatdiet. The oil contained an undetectable amount of peroxidizedlipid at the beginning of the study, which was determined byconventional iodometry of the oil (16). The basal diet contained5% (w/w) safflower oil. Any uneaten food was replaced every

day with food that was stored vacuum-sealed at -20#{176}Cbefore

use. At the end of the experiment at the 38th week, the occur-rence of colon cancer was determined by conventional H&Estaining.

Statistical Analysis. The Student’s t test, � test, and Kruskal-Wallis test were used to compare the difference in numbers of

adenomas and adenocarcinomas between rats fed a 5% saf-

flower oil diet and rats fed other diets. The difference was

considered statistically significant when P was �0.05.

EPR Spectroscopy. Generation of the peroxyl radical fromLOOH in the presence of hemoglobin was determined by EPR

spectroscopy using DMPO as the spin trap agent. Linoleic and

palmitoleic acids were air-oxidized at 60#{176}Cfor 3 days in the

dark. EPR spectra were recorded with a JES-RE1X spectrom-

eter (JEOL, Tokyo, Japan), using quartz flat cells (inner dimen-sions, 60 X 10 x 0.31 mm) with an effective sample volume of

1 80 �l, at room temperature under the following conditions:modulation frequently, 100 kHz; modulation amplitude, 0.079mT; scanning field, 336.1 ± S mT; receiver gain, 1000; re-

sponse time, 0.3 5; sweep time, 2 mm; microwave power, 40

mV; and microwave frequency, 9.421 GHz. Each reaction

mixture contained in 0.2 ml of 0.01 M phosphate buffer (pH7.4), 10 mM LOOH, 0. 1 mg/mi hemoglobin, 1 80 mn DMPO,

and 0.5 mM DTPA, respectively. All spectra were recorded at 1

mm after mixing and incubation at room temperature.

Chemiluminescence Study. Generation of peroxyl radicalsfrom edible oils was quantitated by means of a buminol-

enhanced chemiluminescence assay (17). Oils were air-oxi-dized at 37#{176}Cfor 30 days in the dark. Peroxyl radical-mediated

chemiluminescence was then measured for those oils. The

assay mixture contained 0.3 ml of 10 mivi phosphate-buffered0.15 M saline (pH 7.4), 50 �l of 10 msi DTPA, and 50 �l of

ethanol-containing oil (3 mg/mb); 50 �b of 0.1 msi luminol wereadded and mixed well. The chemiluminescence assay was

started after addition of 50 �l of hemoglobin (1 mg/mb). The

rate, peak intensity, and peak area of chemiluminescence were

measured by using a chemiluminescence multichannel analyzer

(Berthold Model LB 9505 AT, Wildbad, Germany).

DNA Damage Assay. SSB of DNA was determined by mon-

itoring the morphological transition of supercoiled plasmid

DNA to open circular DNA by agarose gel electrophoresis. The

assay mixture contained 2 �l of pUCl9 plasmid DNA (50p.g/ml), 2 �l of 2 mM DTPA, 2 �l of LOOH, and 10 �l of 50mM phosphate buffer (pH 7.4), with or without various inhib-

itors, and was mixed using a vortex mixer. The reaction wasinitiated by addition of 2 pA of heme- or non-heme-iron and

continued for 30 mm at 37#{176}C.Rapeseed and safflower oils were

air-oxidized for >6 days at 60#{176}Cbefore use. Those oils were

then suspended in 0. 1 M phosphate buffer at 30 mg/ml in the

presence of 0.05% Brij 58, a nonionic surfactant to keep the

reaction mixture homogeneous. An aliquot of oxidized oilssimilar to t-BuOOH was added to the reaction mixture in place

of LOOH. An aliquot of reaction mixture (10 �d) was applied

for agarose gel electrophoresis at room temperature for 30 mmat 100 V. After ebecirophoresis, plasmid DNA was stained with



A�

�

c�

D�A/�’W\j�

E

F

Fig. 1. Generation of peroxyl radical from LOOHs in the presence of hemo-

globin determined by EPR spectroscopy using DMPO as a spin trap agent.Linoleic acid and palmitoleic acid were air-oxidized at 60”C for 3 days in the

dark. The redox state of heme-iron used was Fe(III) in all experiments. Concen-

trations of samples in 0.1 M phosphate buffer (pH 7.4) were 10 mM, 100 jsg/ml,

and 180 mM for LOOH, hemoglobin, and DMPO, respectively. Reaction mixtures

contained t-BuOOH and DMPO (A): hemoglobin and DMPO (B): t-BuOOH,

hemoglobin, and DMPO (C); linoleic acid and DMPO (D); linoleic acid, hemo-

globin, and DMPO (F); and palmitoleic acid, hemoglobin, and DMPO (F).

Arrows, peroxyl radical signals. Many peaks generated after the addition of

hemoglobin in E and F may be attributed to other unidentified radical species

including alkyl and alkoxyl. See text for details.

Cancer Epidemiology, Biomarkers & Prevention 1009

ethidium bromide, and the bands were visualized using a UVilluminator (Model TES-20, UVP, Inc., San Gabriel, CA). To

quantify the SSB, the image of the electrophoresis was photo-

graphed, and then densitometric processing was performed.

Results

Colon Cancer Promotion by Diets Containing Fat andHeme-Iron. The body weights and food intake of animalswere identical in each diet group (data not shown). It is note-worthy that the number of rats with tumors and carcinoma wasincreased in rats fed safflower oil (5%) and hemoglobin simul-taneously compared with rats fed without hemoglobin (Table1). The incidence of carcinoma (percentage of rats with carci-

noma) was slightly but not significantly greater in rats fed thediet containing high safflower oil (12%) compared with those

fed the diet with 5% safflower oil. Rats that were given rooibos

tea, a tea from Africa that possesses LOO-scavenging activity,

as drinking water showed a tendency toward decreased mci-dence of colon cancer. Thus, iron (as heme) seems to increase

the incidence of colon carcinogenesis. These findings stronglysupport the above hypothesis concerning the carcinogenic orcancer-promoting potential of LOO and are consistent withepidemiological studies.

Generation of LOO from Oxidized Oils and Heme-Iron.We examined the mechanism of L00 generation from autoxi-

dized fatty acids (i.e.,LOOH) in the presence of various heme

compounds. In this study, we used t-BuOOH as well as linoleicand other unsaturated fatty acids as model compounds forLOOH. EPR spectroscopy using a spin trapping agent (DMPO)

confirmed the generation of LOO. Spontaneously air-oxidizedlinoleic acid, which is the major fatty acid in safflower oil,

readily generated LOO and other radicals by catalysis ofhemoglobin (Fig. 1).

The potential of edible vegetable oils to generate LOOwas also determined by use of luminol-enhanced chemilumi-

nescence (11, 17). We examined various commercial vegetableoils and oils produced at different refining steps during the

manufacturing of commercial oils, i.e.,unpurified native oil,deacidified oil (by alkaline treatment), decolorized oil (by acidclay adsorption of colored compounds), and deodorized oil (bysteaming at >200#{176}Cfor 1 h). Fig. 2 indicates that the potency

of edible oils in generating LOO varied to a great extent, whichwas dependent on the purification steps and which increased

upon air exposure in a time-dependent manner. For rapeseed

oil, deacidification and decolorization greatly increased the

LOO-generating potential compared with that of the native oil.Safflower oil showed the strongest LOO�-generating potential

independent of the purification process among the various oilsat different purification steps. On the other hand, unpurifiedvirgin and extra virgin olive oils and the sesame oils, among thecommercially available oils tested, showed no apparent gener-ation of LOO , even in the presence of hemoglobin, althoughthey were exposed to air at 37#{176}Cfor 30 days. In conclusion,purification processes tend to remove the components thatsuppress LOO generation and/or lipid oxidation in these veg-

etable oils.

LOO-scavenging Potential of Vegetable Oils. We also in-vestigated the LOO-scavenging potential of these oils by using

a system generating t-BuOO , based on chemiluminescenceinduced by t-BuOOH plus hemoglobin (1 1, 17). Inhibitory

potentials against this t-BuOO generation were expressed as

Trolox-equivalent values. Trobox is a water-soluble derivativeof a-tocopherol that effectively scavenges t-BuOO (1 1). Fig. 3shows that the extra virgin and virgin olive oils showed the

highest t-BuOO-scavenging potential. Sesame oil and rapeseedoil (manufactured by traditional/classic preparation methodswithout acid/base wash and steaming) showed weaker scaveng-ing activities than virgin olive oils; all other refined commercialoils including safflower, corn, rapeseed (modern manufactur-ing), and rice bran oils showed no detectable t-BuOO -scav-

enging potential. All purification processes decreased remark-

ably the t-BuOO-scavenging potency compared with thepotency of the unpurified native oils, which had possessed

radical-scavenging capacity.

DNA-cleaving Potential of Lipid-derived Radicals. It nowbecame more important to clarify whether the LOO woulddirectly injure DNA, the most crucial step for cancer develop-ment. We thus examined the possibility of SSB of DNA using

circular plasmid DNA (pUC19), which changes from a super-coiled form to an open circular form and is readily identifiable

with 1% agarose gel electrophoresis.As shown in Fig. 4A, neither t-BuOOH nor hematin alone

could induce SSB, whereas simultaneous addition of t-BuOOH

and hematin, which generates the t-BuOOH-derived radical

[t-BuOO}, caused significant SSB that depended on the con-centration of t-BuOOH. Substitution of heme-iron by nonheme

ferrous iron (DTPA-Fe2�) is known to generate the primaryalkyl radical (t-Bu), or C-centered radical, and the alkoxyl

radical (t-BuO), but not t-Bu00 (10). Both t-Bu and t-BuOfailed to induce SSB. In other words, these results confirm thatneither t-Bu nor t-Bu0 is not the operating species for SSB.



Rapeseed oilE0.(5

‘!; 1000

1

II

4.

5.

6’

7.

8

9,

10’

11’

12

13

14

E0.(5

0

5<

C0

0asC00)

a0-I

E0.(5

0

5<

C

0

(aasC0

1500’

0 10 20 30 40

(c) Commercial oils

10 100

1000’

500

0 10 20 30 40

Days of oxidation

Typical scavengers for O� (superoxide dismutase) and

#{149}OH(mannitol and DMSO) were examined; they failed to

suppress SSB induction by t-BuOO in the t-BuOOH/hemoglo-bin system (Fig. 4D). Therefore, the common reactive oxygen

species such as O� and 0H may not be major contributors toSSB induction in the present LOO-generating system. On the

other hand, in the same system in the presence of sodiumcyanide, which was reported to inhibit the generation of t-

BuOO in the t-BuOOH/heme-iron system (18), no SSB wasobserved (Fig. 4D, Lane 7).

We then examined the LOO -scavenging activity of com-mon free radical scavengers, many of which are known to havepreventive effects on carcinogenesis in animal models (19-21).

For example, protocatechuic acid, which suppresses chemicalcarcinogenesis (19 -2 1), also effectively suppressed the gener-

ation of t-BuOO as determined by EPR (data not shown) as

webb as SSB of plasmid DNA, as shown in Fig. 4D. Taking allthese data together, it appears that t-BuOO and LOO are thusprimary molecular species responsible for SSB.

Discussion

The present study demonstrates that the simultaneous uptake offat and heme-iron significantly increased the incidence of colon

This result is also consistent with the dominance of the bacte-ricidab action of LOO over LO or L (10, 11). Similar DNA

cleavage was also evidenced in an experiment using oxidizedrapeseed oils and safflower oils (Fig. 4C). Furthermore, theincreasing DNA-cleaving potential was correlated to the ad-vances of the refining process and was parallel to the decreased

capacity for scavenging the LOO radical.

1010 I�i� Peroxyl Radicals in Carcinogenesis

LOO. scavenging potential(Trolox equiv., mg/g)

Fig. 3. LOO’-scavenging potentials of various vegetable oils. Chemilumines-cence was generated by the reaction oft-BuOOH (10 mM) with hemoglobin (100

jsg/ml). After addition of various edible oils, the LOO’-scavenging potential was

quantified against the model radical component generated (t-BuOO’). Results are

expressed as equivalent values to the well-known antioxidant, Trolox. Results for

extra virgin olive oil, virgin olive oil, and sesame oil are the means of threesamples each from different companies; bars, SD (n - 3). Oils examined: 1, extra

virgin olive oil (commercial); 2, virgin olive oil (commercial); 3, sesame oil

(commercial); 4, rapeseed oil (traditional/classic refining, commercial); 5, rape-seed oil (unpurified native oil); 6, com oil (com germ/unpurified native oil); 7,

safflower oil (unpurified native oil); 8, com oil (deacidified oil); 9, com oil

oil 0 (decolorized oil); 10, com oil (deodorized oil); 11, com oil (commercial); 12,safflower oil (commercial); 13, rice bran oil (commercial); and 14, rapeseed oil

(commercial). See text for details.

Fig. 2. The LOO’-generating potentials of various vegetable oils determined by

luminol-enhanced chemiluminescence. Oils were air-oxidized at 37”C for 30 days

in the dark. Peroxyl radical-mediated chemiluminescence was measured for those

oils (3 mg/mI) in the presence of hemoglobin (100 �sg/ml) and luminol (10 LM)

in 0. 1 M phosphate buffer. a, rapeseed oils from different refining processes:

unpurified native oil (fl); deacidified oil (#{149});decolorized oil (0); and deodorized

oil (s). b, safflower oils from different refining processes: unpurified native oil

(LI); deacidified oil (U); decolorized oil (0); and deodorized oil (s). c, commer-cial oils: safflower oil (U); com oil (#{149});rapeseed oil (traditional/classic prepa-

ration; 0); sesame oil (s); olive oil (A); virgin olive oil (U); and extra virgin olive

oil (0).



A

hematin ( jiM) 0 200 0 0 0 200 200 200

t-BuOOH ( mM) o 0 1 10 100 1 10 100

OC

SC

B

OC

SC

C

OC

SC

D

OC

SC

Fig. 4. SSB on plasmid DNA (pUC19) caused by t-BuOOH or oxidized oils

plus heme-iron. A, concentrations of t-BuOOH and hematin were varied asindicated. B, effect of different types of heme compounds on t-BuOOH-induced

DNA damage. Concentration of t-BuOOH was 0 mrsi (Lane 1) or 20 inst (Lanes

2-6). The concentration of iron was fixed at 100 .sM, where hemoglobin (Lane

3), myoglobin (Lane 4), hematin (Lane 5), or DTPA-Fe2� (Lane 6) were used. C,

DNA damage by oxidized edible oils with heme-iron. Plasmid DNA was incu-

bated with oxidized edible oils as follows in the presence of hematin (200 �.sM):

Lane 1, unpurified native rapeseed oil; Lane 2, deacidified rapeseed oil; Lane 3,

decolorized rapeseed oil; Lane 4, deodorized rapeseed oil; Lanes 5 and 6,

commercial safflower oils from different companies. Neither oxidized oil alone

nor hematin induced DNA cleavage (data not shown). D, effect of various free

radical scavengers on DNA damage induced by t-BuOOH (3 mM)/hemoglobin [25

�LM in Fe(III)] at 37#{176}Cfor 30 mm. Lane 1, positive control without scavenger;

Lane 2, control plus deferoxamine (I mM); Lane 3, control plus sodium cyanide

(5 mM); Lane 4, control plus Cu:Zn-superoxide dismutase (1000 units/mi); Lane

5, control plus mannitol (10 mM); Lane 6, control plus DMSO (10 vol%); Lane

7, control plus protocatechuic acid (10 mM). SC, supercoiled form of DNA; OC,

open circular form of DNA. See text for details.

Cancer Epidemiology, Biomarkers & Prevention 1011

- -

-

�

-

I 2 3 4 5 6

1 2 3 4 5 6

-� � �*m = �: � -.� - -

1 2 3 4 5 6 7� ,‘ �. � �.

‘a. � ‘�. � �, ;_ �

cancer in rats, compared with a diet without heme-iron (Table1). Fat as hydroperoxides and heme-iron in the diet were found

to generate LOO (Figs. 1 and 2), which may be cytotoxic to the

endothelial cells in the digestive tract, similar to the observationof its bactericidal action (10, 1 1 ). This cytotoxic effect of LOOin the digestive tract may induce proliferation of new cells as a

repair process, thus conferring greater carcinogenic potential( 13). In addition to the cytotoxic effect, the radicals derivedfrom fat and heme-iron were found to cleave DNA in vitro. Itis interesting to note that protocatechuic acid, a vegetablecomponent, completely suppresses DNA cleavage by L00

(Fig. 4D, Lane 7). This result suggests that such DNA damagemay augment carcinogenic potential, particularly if it occurs intumor suppressor genes such as the p53 (22) or APC (23) genes,

and highlights the importance of dietary vegetable components

for scavenging LOO.

Although oxidized oils were known to damage DNA (24,

25), there have been few studies to examine the mechanism in

detail, including the identification of reactive molecular spe-cies. The present study using the LOO -specific generationsystem and several inhibition assays indicates that the principalactive component for SSB is LOO (Figs. I and 4), which isalso generated readily from oxidized edible oils by heme-iron

catalyzed decomposition. These results seem to be important

for two reasons: (a) the half-life of LOO is much longer [>30

mm (10)] than those of OH, O�, or RO’; LOO thus can travellonger distances and thus allow larger target areas in vito; and

(b) the lipophilic characteristics may facilitate crossing theintestinal wall via particles such as chybomicrons, although this

notion needs to be verified.L00 (LOOH) may also be a source of O� or 0H through

a variety of pathways. For instance, electron transfer betweenLOO and ubiquinol-lO could lead to generation of O� (26). O�

is also generated by enzymatic decomposition of LOO ,asreported for the lipoxygenase system (27). Very recently, poly-morphonuclear cells were found to generate O� from hydroper-

oxides of diacylglycerol ( 14). Intracellular O� could also affect

cell proliferation by modulating ras-mediated mitogenic sig-

naling (28). These LOO radicals and other reactive oxygenspecies, including singlet oxygen (‘02), could oxidativeby mod-ify DNA bases, as found in the formation of 8-hydroxydeox-

yguanosine as a representative event, which would result in

genetic alteration. All of these effects may contribute to theage-related increase of pathogenesis of diseases such as cancerand arteriosclerosis (29).

In our earlier studies using Raji cells harboring EBV, thetransformation of Raji cells to the early antigen-positive cellsby phorbol 12-myristate 13-acetate was suppressed effectivelyby hot water extracts of various green vegetables. This sup-pression parallels the t-Bu00 -scavenging activity data ob-tamed for > 100 vegetables (r = 0.82; Ref. 17). The suppres-

sion of Raji cell transformation to early antigen-positive cell by

a t-BuOO scavenger suggests a possibility that tumor promo-tion is caused by LOO ; thus, L00, and hence LOOH, could

themselves exert a tumor-promoting effect or do so via freeradical generation, including O�, involving polymorphonuclear

cells (14). These findings further support the involvement offree radical species in colon cancer related to dietary fats.

The effect of lipids, particularly LOO, may thus be animportant factor in cancer promotion. Native vegetable oilscontain a large amount of peroxyl radical scavengers as webb asnatural antioxidants, in which two crucial steps are involved:prevention of oxidation of PUFAs to form LOOH and scav-

enging of lipid radicals (LOO’) thus generated from LOOH.As commented in the “Introduction,” a diet high in lipids

is a risk factor for cancer incidence. A concordant result in this

connection was reported (30); a supplementation of lipids infeed was found to stimulate tumor growth in animal model, andits restriction suppressed transplanted tumor growth (30). Al-though this result is not directly related to the present study, the

excessive dietary fat content, especially oxidized fat, might

affect the cancer-promoting effect in one way or the other.It should also be noted from the present data (especially

Figs. 2 and 3) that conventional purification processes usedduring the manufacturing of vegetable oils may reduce suchvaluable anti-LOO activities and antioxidant components, both

of which are abundant in the plant seeds originally. Conse-quently, oils such as virgin olive oil that are rich in L00scavengers are preferable not only for an anticarcinogenic po-

tential but also for prevention of reactive oxygen-rebated dis-eases, including cardiac disease and arteriosclerosis. Conse-

quently, one might rationalize that Mediterranean diets rich in



1012 LIpId Peroxyl Radicals In Carclnogenesis

olive oils may thus be valuable for their potent L00 -scaveng-ing activity (3 1, 32), because the virgin and extra virgin oils are

strongest in this activity (Figs. 2 and 3).In conclusion, the present investigation shed light on

L00 formation via heme plus oxidized oil and its effect onDNA cleavage. Furthermore, native oils are generally richer in

antioxidant as well as L00-scavenging activity than pro-cessed/purified oils. This indicates that important radical-scav-enging as well as antioxidant components in vegetable oilsmight be removed by manufacturing processes.

Acknowledgments

We acknowledge Toni Yamada, who provided the most diligent and excellent

technical assistance, and Dr. Masahiro Takano, who provided support with great

dedication for the prevention of colon cancer. Without their contribution, it would

have been impossible to complete this work. We also thank Judith Gandy for

editing and Rie Yoshimoto for typing.

References

1. Cohen, L. A. Diet and cancer. Sci. Am., 257: 42-48, 1987.

2. Giovannucci, E., Stampfer, M. J., Colditz, G., Rimm, E. B., and Willett, W. C.

Relationship of diet to risk of colorectal adenoma in men. J. Natl. Cancer Inst., 84:

91-98, 1992.

3. Risch, H. A., Jam, M., Marrett, L. D., and Howe, G. R. Dietary fat intake andrisk of epithelial ovarian cancer. J. NatI. Cancer Inst., 86: 1409-1415, 1994.

4. Reddy, B. S., and Maeura, Y. Tumor promotion by dietary fat in azoxymeth-

ane-induced colon carcinogenesis in female F344 rats: influence of amount andsource of dietary fat. J. NatI. Cancer Inst., 72: 745-750, 1984.

5. Takeshita, M., Ueda, H., Shirabe, K., Higuchi, Y., and Yoshida, S. Lack of

promotion of colon carcinogenesis by high-oleic safflower oil. Cancer (Phila.),

79: 1487-1493, 1997.

6. Stevens, R. G., Jones, D. Y., Micozzi, M. S., and Taylor, P. R. Body iron stores

and the risk of cancer. N. EngI. J. Med., 319: 1047-1052, 1988.

7. Nelson, R. L., Davis, F. G., Sutter, E., Sobin, L. H., Kikendall, J. W., andBowen, P. Body iron stores and risk ofcolonic neoplasia. J. NatI. Cancer Inst., 86:

455-460, 1994.

8. Giovannucci, E., Rimm, E. B., Stampfer, M. J.. Colditz, G. A., Ascherio, A.,

and Willett, W. C. Intake of fat, meat, and fiber in relation to risk of colon cancer

in men. Cancer Res., 54: 2390-2397, 1994.

9. Halliwell, B., and Guueridge, J. M. C. Lipid peroxidation: a radical chain

reaction. In: B. Halliwell and J. M. C. Gutteridge (eds.), Free Radicals in Biologyand Medicine, pp. 188-276. Oxford, United Kingdom: Oxford University Press,

1989.

10. Akaike. T., Sato, K., Ijiri, S., Miyamoto, Y., Kohno, M.. Ando, M., and

Maeda, H. Bactericidal activity of alkyl peroxyl radicals generated by heme-iron-catalyzed decomposition of organic peroxides. Arch. Biochem. Biophys., 294:

55-63. 1992.

I I . Akaike. T., Ijiri, S., Sato, K., Katsuki, T., and Maeda, H. Determination of

peroxyl radical-scavenging activity in food by using bactericidal action of alkyl

peroxyl radical. J. Agric. Food Chem., 43: 1864-1870, 1995.

12. Van der Zee, J., Van Steveninck, J., Koster, J. F., and Dubbelman, T. M.

Inhibition of enzymes and oxidative damage of red blood cells induced by

t-butylhydroperoxide-derived radicals. Biochim. Biophys. Acta, 980: 175-180,

1989.

13. Weinstein, I. B. Mitogenesis is only one factor in carcinogenesis. Science

(Washington DC). 251: 387-388. 1991.

14. Yamamoto, Y., Kambayashi, Y., Ito, T., Watanabe, K., and Nakano, M.

I .2-Diacylglycerol hydroperoxides induce the generation and release of super-

oxide anion from human polymorphonuclear leukocytes. FEBS Lets., 412: 461-464, 1997.

15. Heilig, J. S., Lech, K., and Brent, R. Large-scale preparation of plasmid

DNA. In: F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman,J. A. Smith, and K. Struhl (eds.). Current Protocols in Molecular Biology, Vol. 1.,

pp. 1.7.1-1.7.1 1. New York: John Wiley & Sons, 1987.

16. Pryor, W. A., and Castle, L. Chemical methods for the detection of lipid

hydroperoxides. Methods Enzymol., 105: 293-299, 1984.

17. Maeda, H., Katsuki, T., Akaike, T., and Yasutake, R. High correlation

between lipid peroxide radical and tumor-promoter effect: suppression of tumor

promotion in the Epstein-Barr virus/B-lymphocyte system and scavenging ofalkyl peroxide radicals by various vegetable extracts. Jpn. J. Cancer Res., 83:

923-928, 1992.

18. Davies, M. J. Detection of peroxyl and alkoxyl radicals produced by reaction

of hydroperoxides with heme-proteins by electron spin resonance spectroscopy.

Biochim. Biophys. Acta, 964: 28-35, 1988.

19. Yang, C. S., and Wang, Z. Y. Tea and cancer. J. NatI. Cancer Inst., 85:

1038-1049, 1993.

20. Kawamori, T., Tanaka, T., Kojima, T., Suzui, M., Ohnishi, M., and Moo, H.

Suppression of azoxymethane-induced rat colon aberrant crypt foci by dietary

protocatechuic acid. Jpn. J. Cancer Res., 85: 686-691, 1994.

21. Yamane, T., Takahashi, T., Kuwata, K., Oya, K., Inagake, M., Kitao, Y.,

Suganuma, M., and Fujiki, H. Inhibition of N-methyl-N’-nitro-N-nitrosoguani-

dine-induced careinogenesis by ( - )-epigallocatechin gallate in the rat glandular

stomach. Cancer Res., 55: 2081-2084, 1995.

22. Greenblan, M. S., Bennett, W. P., Hollstein, M., and Harris, C. C. Mutations

in the p53 tumor suppressor gene: clues to cancer etiology and molecular patho-

genesis. Cancer Res., 54: 4855-4878, 1994.

23. Morin, P. J., Vogelstein, B.. and Kinzler, K. W. Apoptosis and APC in

colorectal tumorigenesis. Proc. Nail. Acad. Sci. USA, 93: 7950-7954, 1996.

24. Vaca, C. E., Wilhelm, J.. and Harms-Ringdahl, M. Interaction of lipidperoxidation products with DNA. Mutat. Res., 195: 137-149, 1988.

25. Yang, M. H., and Schaich, K. M. Factors affecting DNA damage caused by

lipid hydroperoxides and aldehydes. Free Radical Biol. Med., 20: 225-236, 1996.

26. Frei, B., Kim, M. C., and Ames, B. N. Ubiquinol-lO is an effective lipid-

soluble antioxidant at physiological concentrations. Proc. NatI. Acad. Sci. USA,

87: 4879-4883, 1990.

27. Chamulitrat, W., Hughes, M. F., Eling, T. E., and Mason, R. P. Superoxide

and peroxyl radical generation from the reduction of polyunsaturated fatty acid

hydroperoxides by soybean lipoxygenase. Arch. Biochem. Biophys.. 290: 153-

159, 1991.

28. Irani, K., Xia, Y., Zweier, J. L., Sollott, S. J., Der, C. J., Fearon, E. R.,

Sundaresan, M., Finkel, T., and Goldschmidt-Clermont, P. J. Mitogenic signaling

mediated by oxidants in Ras-transformed fibroblasts. Science (Washington DC),

275: 1649-1652, 1997.

29. Fraga, C. G., Shigenaga, M. K., Park, J. W., Degan. P., and Ames, B. N.

Oxidative damage to DNA during aging: 8-hydroxy-2’-deoxyguanosine in rat

organ DNA and urine. Proc. Natl. Acad. Sci. USA, 87: 4533-4537, 1990.

30. Wang, Y., Corr. J. G., Thaler, H. T.. Tao, Y., Fair, W. R., and Heston, W. D.

Decreased growth of established human prostate LNCaP tumors in nude mice feda low-fat diet. J. Natl. Cancer Inst., 87: 1456-1462, 1995.

31. de Lorgeril, M., Renaud, S., Mamelle, N., Salen, P., Martin, J. L., Monjaud,

I., Guidollet, J., Touboul, P., and Delaye, J. Mediterranean a-linolenic acid-rich

diet in secondary prevention of coronary heart disease. Lancet, 343: 1454-1459,

1994.

32. Trichopoulou, A., Katsouyanni, K., Stuver, S., Tzala, L., Gnardellis, C.,Rimm, E., and Trichopoulos, D. Consumption of olive oil and specific food

groups in relation to breast cancer risk in Greece. J. NatI. Cancer Inst., 87:

110-116, 1995.



1998;7:1007-1012. Cancer Epidemiol Biomarkers Prev T Sawa, T Akaike, K Kida, et al. implication of a high-fat diet in colon carcinogenesis.Lipid peroxyl radicals from oxidized oils and heme-iron:

Updated version

http://cebp.aacrjournals.org/content/7/11/1007

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cebp.aacrjournals.org/content/7/11/1007To request permission to re-use all or part of this article, use this link



https://bloodcancerdiscov.aacrjournals.orgcgi/alerts

mailto:[email protected]



Documents

Lipid Peroxyl Radicals from Oxidized Oils and Heme-Iron ...Nippon Oils & Fats Co. (Tokyo, Japan). Olive oils were ob-tamed from Eleourgiki and others in markets in Athens, Greece