dietary menhaden oil enhances mitomycin c antitumor activity toward human mammary carcinoma mx-1

Dietary Menhaden Oil Enhances Mitomycin C Antitumor Activity Toward Human Mammary Carcinoma MX-1

Yu Shao, Lani Pardini and Ronald S. Pardini* Allie M. Lee Laboratory for Cancer Research, Department of Biochemistry, University of Nevada, Reno, Nevada 89557

ABSTRACT: In the present study, we investigated the effects of high levels of dietary fish oil on the growth of MX-1 human mammary carcinoma and its response to mitomycin C (MC) treatment in athymic mice. We found that high levels of dietary fish oil (20% menhaden oil + 5% corn oil, w/w) compared to a control diet (5% corn oil, w/w) not only lowered the tumor growth rate, but also increased the tumor response to MC treatment. We also found that high levels of dietary fish oil significantly increased the activities of tumor xanthine oxidase and DT-diaphorase, which are proposed to be involved in the bioreductive activation of MC. Since menhaden oil is highly unsaturated, its intake caused a significant increase in the degree of fatty acid unsaturation in tumor membrane phospholipids. This alteration in tumor membrane phospholipids made the tumor more susceptible to oxidative stress, as indicated by the increased levels of both endogenous lipid peroxidation and protein oxidation after feeding the host animals the menhaden oil diet. In addition, the tumor antioxidant enzyme activities, catalase (CAT), superoxide dismutase (SOD), glutathione peroxidase (GPOx), and glutathione S-transferase peroxidase (GSTPx), were all significantly enhanced by feeding a diet high in fish oil. MC treatment caused further increases in tumor lipid peroxidation and protein oxidation, as well as in the activities of CAT, SOD, GPOx, and GSTPx, suggesting that MC causes oxidative stress in this tumor model which is exacerbated by feeding a diet high in menhaden oil. Thus, feeding a diet rich in menhaden oil decreased the growth of human mammary carcinoma MX-1, increased its responsiveness to MC, and increased its susceptibility to endogenous and MC-induced oxidative stress, and increased the tumor activities of two enzymes proposed to be involved in the bioactivation of MC, that is, DT-diaphorase and xanthine oxidase. These findings support a role of these two enzymes in the bioactivating of MC and indicate that the type of dietary fat may be important in tumor response to therapy. Lipids 30, 1035-1045 (1995).

*To whom correspondence should be addressed at Department of Biochem- istry, Mail Stop 330, University of Nevada, Reno, NV 89557. Abbreviations: CAT, catalase; CB5R, cytochrome b 5 reductase; CCR, cytochrome C reductase; CO, corn oil; DHA, docosahexaenoic acid; DNPH, 3.4-dinitrophenyl hydrazine; DTD, DT-diaphorase; EPA, eicosapentaenoic acid; GPOx, glutathione peroxidase; GR, glutathione reductase; GSH, glutathione; GSSG, glutathione disulfide; GST, glutathione S-transferase; GSTPx, glutathione S-transferase peroxidase; lINE, 4-hydroxy-nonenal; MC, mitomycin C; MO, menhaden oil; PL, phospholipid; PUFAs, polyunsaturated fatty acids; SOD, superoxide dismutase; TBARS, thiobarbituric reactive substances; T/C, treat/control ratio; XD, xanthine dehydrogenase; XO, xanthine oxidase.

Fish oil, which contains high levels of n-3 long-chain polyunsaturated fatty acids (PUFAs), has been shown to sup- press mammary gland tumorigenesis in rodent models (1-4). Several hypotheses have been proposed to explain how fish oil inhibits mammary tumorigenic processes, including inhibition and/or alteration of eicosanoid metabolism, induction of oxidative stress, and a deficiency in essential fatty acids in fish oil diets (1-4). The n-3 long-chain PUFAs contained in fish oil mainly are eicosapentaenoic acid (20:5, EPA) and docosahexaenoic acid (22:6, DHA), which contain five and six double bonds, respectively. When the PUFAs are presented in the diets, they are readily incorporated into tumor membrane phospholipids (!-4), resulting in an increase in the level of unsaturation of the membrane lipids. This makes the membrane more susceptible to peroxidation in vitro and in vivo, which leads to an increased accumulation of cytostatic and/or cyto- toxic lipid peroxidation products, especially in the tumor (5,6).

Mitomycin C (MC), a quinone-containing antibiotic active in the treatment of solid tumors, undergoes both one-electron and two-electron reduction by the MC bioreductive activating enzymes (7-10). The two-electron reduction of MC forms a hydroquinone structure which can cross-link DNA, especially under hypoxic conditions. The one-electron reduction of MC results in the formation of the semiquinone free radical, which can interact with molecular oxygen to form toxic reactive oxygen species, such as superoxide, hydrogen peroxide, and hydroxy radical (-OH). The free radical dependent process could be critical to the antitumor activity of MC, for either DNA alkylation, DNA oxidation, or peroxidation of cellular membrane phospholipids (7-10).

It has been reported that high levels of dietary fat enhance the activities of a variety of enzymes, including those responsible for metabolizing xenobiotics (11-13). In a previous study (14), we found that MC bioreductive activating enzymes, including DT-diaphorase (DTD) and cytochrome b 5 reductase (CB5R) in human mammary carcinoma MX-I het- erotransplanted in the nude mice, were increased after feeding the host animals a high corn oil (CO) diet (25% CO, w/w). This would result in an increase in the formation of one-electron or two-electron reductive metabolites of MC in the tumor, increasing its antitumor effect through either increasing DNA cross-linking or production of toxic free radicals. In addition, we observed that feeding a high CO diet caused an

Copyright �9 1995 by AOCS Press 1035 Lipids, Vol. 30, no. 11 (1995)

1036 Y. SHAO ETAL.

increased tumor response to MC treatment and increased oxidative stress after MC treatment, as indicated by elevated lipid peroxidation and enhanced antioxidant enzyme activities in the tumor.

In the present study, we determined the effect of feeding a diet high in fish (menhaden) oil [20% menhaden oil (MO) + 5% CO) on the growth, susceptibility to MC treatment, endogenous and MC-induced oxidative stress, and effect on MC-activating enzyme activities in human mammary carcinoma MX-1 grown in athymic mice.

MATERIALS A N D M E T H O D S

Materials. All chemicals were obtained from Sigma Chemi- cal Co. (St. Louis, MO), unless otherwise indicated. The BCA total protein assay kit was obtained from Pierce Co. (Rock- ford, IL). Dietary components were purchased from Dyets Inc. (Bethlehem, PA).

Animals and diets. Forty-eight female adult athymic mice (female heterozygous BALB/c nu/+) were used in the present studies. The mice were housed under aseptic conditions (germ-free laming-flow hood, sterilized cages, bedding, and water) at 27~ Autoclaved laboratory mouse chow (Dyets Inc.) was fed ad libitum before the experimental diet was fed. The experimental diet was composed of the casein-based, semipurified AIN-76A diet (15,16) with some modifications according to Birt et al. (17), containing 20% MO + 5% CO (w/w) or 5% CO (w/w) (Table 1).

Tumor implantation and MC treatment. The animals were assigned randomly to the high-MO or low-CO diet, with 24 mice in each of the two experimental groups. Mice were fed their respective experimental diets for 20 d prior to tumor in-

TABLE 1 Composition of Experimental Diets a

Ingredient (g) Control (CO) Menhaden oil (MO)

Corn oil (CO) 5 5 MO 0 20 Casein 20 24.6 DL-methionine 0.3 0.5 Cornstarch 15 9 Sucrose 50 29.1 Cellulose 5 6 AIN vitamin mix 1 1.2 AIN mineral mix 3.5 4.3 Choline bitartrate 0.2 0.3

Total weight 100 100 Caloric density (kcal/g) b 3.85 4.85 % of calories from fat 11.69 46.39

Unsaturation index 141 243 Daily diet consumed (g/mouse) c 4.36 • 0.89 3.54 + 0.80 Daily caloric intake (kcal/mouse) 16.79 • 3.43 17.17 • 3.88

aFormulated in accordance with the American Institute of Nutrition recom- mendation (15,16) and modified according to Birt et aL (17). bCalculated as 4.0 kcal/g for starch, sucrose, and casein, and 9.0 kcal/g for corn oil and menhaden oil. CCalculated daily by subtracting the amount of uneaten food from the amount of the food added the day before.

oculation. A tumor fragment (1-2 mm 3) of human mammary carcinoma MX-1 was implanted subcutaneously on the right foreflank of each nude mouse. Tumor weights were recorded every 3-4 d and were estimated by the calculation based on the formula:

A .B .C tumor weight (mg) = - - [ 1 ]

2

where A, B, and C represent the three perpendicular diame- ters of the tumor in millimeters (18). The percentage T/C ratio was calculated by the mean tumor weight of the treated group divided by the mean tumor weight of the control group (MC untreated) multiplied b y 100 (18). All the mice were assigned to the following two experiments.

Experiment 1. Twelve mice in each dietary group with a total of 24 were used in this experiment. Ten days after tumor implantation, half of the mice in high MO and low CO groups (six animals in each group) were administrated with MC at the dosage of 0.5 mg/kg body weight, i.p., once a week for four consecutive weeks. Thirty days after tumor implantation, the animals in both dietary groups that did not receive MC treatment were sacrificed by cervical dislocation due to the large tumor volume. The remaining animals (MC treated) were kept on the experimental diets for another 30 d, and tumor volume and body weights were recorded as above. (Their tumors were too small to conduct biochemical testing.)

Experiment 2. Another 24 mice with 12 in each dietary group were used. Half (six animals) of the mice in each group received MC treatment 15 d after tumor implantation for three weeks, with the same dosage as in experiment 1. All the mice were sacrificed 24 h after the third MC treatment. Tumors were removed immediately, either homogenized in 3 mL of ice-cold 50 mM phosphate buffer (pH 7.0) for enzyme activities, lipid peroxidation, and protein oxidation assays, or kept in liquid N 2 for glutathione content and membrane phospholipid composition measurement. The homogenate was cen- trifugated at 10,000 x g for 30 min and the supernatant was retained for enzymatic assays.

t~nzyme assays. DTD activity was measured as the dicumarol-sensitive reduction of 2,6-dichlorophenolindophe- nol, according to the method of Ernster (19) and modified by Benson et al. (20). Xanthine oxidase/dehydrogenase activities were assayed with the formation of uric acid from xanthine measured spectrophotometrically at 293 nm (21). Cy- tochrome C reductase (CCR) activity was measured according to the method of Masters et al. (22) and Yasukochi and Masters (23). Superoxide dismutase (SOD) activity was measured using the method described by Oberely and Spitz (24). Catalase (CAT) activity was determined by measuring the sample catalyzed disappearance of H202 spectrometrically at 240 nm (25). Glutathione peroxidase (GPOx) activity was measured in a coupled enzyme reaction with glutathione reductase (GR) (26). Glutathione S-transferase peroxidase (GSTPx) activity was measured similar to GPOx activity (27). GR activity was measured as described by Racker (28).

Lipids, Vol. 30, no. 11 (1995)

MENHADEN OIL AND CHEMOTHERAPY 1037

GST activity was determined by the method according to Mannervik and Guthenberg (29).

Lipid peroxidation. Malondialdehyde (MDA) was measured as an indication of lipid peroxidation by the method of Buege and Aust (30). Thiobarbituric acid [0.375% (wt/vol) with 15% trichloroacetic acid and 0.25 N HCI] was mixed with the test sample and heated in boiling water for 10 min. The precipitate was removed by centrifugation at 1000 • g for 10 min and the supernatant was removed for further analysis. Thiobarbituric reactive substances (TBARS) in the supernatant were measured spectrophotometrically at 535 nm.

Protein oxidation. Protein carbonyl groups were measured as an indication of protein oxidation by the method according to Levine et al. (31). Schiff's bases was also measured by this procedure. DNA was removed from the fraction by strep- tomycin precipitation. The remaining protein was precipi- tated and the pellet mixed with 10 mM 3,4-dinitrophenyl hydrazine (DNPH) for 1 h. Excess DNPH is removed with 1:1 ethanol/ethyl acetate mixture and the pellet is redissolved in 0.6 M guanidine in 20 mM potassium phosphate buffer, pH 6.3. Solution absorbance was then measured at 360-390 nm (e360-390 = 22,000 M-Icm -1) to determine the protein carbonyl content. Control sample absorbance without DNPH was subtracted from the test absorbance.

Tumor glutathione (GSH) levels. Tumors were homogenized in 10% (vol/vol) perchloric acid containing 1 mM bathophenanthrolinedisulfonic acid. This acid extract was then spun at 15,000 x g for 15 min, and the supernatant de- rivatized and analyzed by HPLC by the method of Fariss and Reed (32). HPLC analysis was carried out using a Beckman 338 binary gradient system (Beckman Instruments, Palo Alto, CA) fitted with an a-Aminopropyl-Spherisorb column (Cus- tom LC, Inc., Houston, TX).

Membrane lipid content. Tumor membrane lipid compositions were analyzed as previously described (18). Lipid was extracted from tumor membrane with chloroform/methanol

(2:1, vol/vol). Fatty acid methyl esters were obtained by transesterification with borontrifluoride/methanol reagent. The methyl esters were analyzed by gas chromatography.

Statistical analysis. Data are expressed as means _+ SD with n = 6 per group; differences in means were assessed by analysis of variance using the SAS general linear models pro- gram. Group means were considered to be significantly different at P < 0.05.

RESULTS

Tumor growth studies. Dietary intake of high levels of n-3 PUFAs has been reported to inhibit tumor growth and tumorigenesis in rodent models (1-4). In this study, we employed a diet high in n-3 PUFAs to evaluate its effects on tumor growth and responsiveness to MC chemotherapy. As shown in Table 2, the growth of human mammary carcinoma MX- 1 was significantly inhibited in the animals fed a diet high in MO in Experiment 1. At the 30th day after tumor implantation, the average tumor weight in the high-MO group was 2016 _+ 875 mg, whereas in the low-CO group it was 3500 + 1050 mg (43% inhibition by MO). The treatment/control (T/C) ratios resulting from MC treatment were 1 for MO group and 10 for CO group with final mean tumor weights being 15 and 342 mg, respectively. Because 50 and 17% of the animals were clear of tumor at 30 d, the average tumor size of the remaining tumors at day 30 were 29 _+ 10 mg for MO and 386 +_ t02 mg for CO (Table 2). The MO-fed group possessed tenfold better response to MC compared to the CO-fed group. These TIC ratios were calculated by comparing the MC-treated group to its respective dietary control (see Materials and Methods section). If we compared the MO-fed MC-treated tumor to the CO control (dual effects from dietary MO feeding and MC treatment), the combined T/C ratio becomes 0.4. Thus feeding a 25% MO diet resulted in a dramatic reduction in tumor size following MC treatment compared to the 5% CO control. The animals

TABLE 2 Antineoplastic Activity of MC Against MX-1 After Different Diet Feeding

Dietary group Treatment

Net change in body Mean tumor weight Combined weight between day

(mg • SD) T/C ratio T/C ratio % ClearC No. of animals 0 and day 30 (day 30) day 30 a day 30 b Day 30 Day 60 used (N) (g • SD) d

CO (5% corn oil) Control 3500 _+ 1050 0 - - 6 5.1 + 1.2 10 10

MC 342 + 164 e 17 0 6 2.1 +0 .4

MO Control 2016 :~ 875 r 0 - - 6 4.5 • 1.0 (5% corn oil 1 0.43 + 20% menhaden oil) MC 15 • 17 e'g 50 33 6 1.5 • 0.3

aThe percentage treatment/control ratio (TIC ratio) was calculated by taking the mean tumor weight of the mitomycin C (MC)-treated group divided by the mean tumor weight of the respective control (MC untreated) multiplied by 100. Other abbreviations as in Table 1. ~lhe combined T/C ratio accounts for the effect of diet and MC treatment and is calculated by taking the mean tumor weight of the MC-treated groups divided by the mean tumor weight of the CO control group multiplied by 100. Cpercent of animals free of visible or palpable tumor at the days indicated (the days after tumor implantation). dNet change in body weight from day 0 (day of tumor implantation) to day 30 (day of termination of experiment). el'he average size of the tumors in mice that had visible or palpable tumors were 386 _+ 102 mg in CO group and 29 -+ 10 mg in MO group. ~Mean tumor weight is statistically different from the CO control group, P < 0.05. gMean tumor weight is statistically different from the CO+MC group, P< 0.01.

Lipids, Vol. 30, no. 11 (1995)

1038 Y. SHAO ETAL.

were maintained on the experimental diets for an additional 30 d following MC treatment, and observed for reoccurring tumor. In the low CO-fed group, the mouse free of tumor at day 30 experienced reoccurrence by the 60th day following tumor implantation. Of the three tumor-free mice in the MO group, only one appeared to have tumor regrowth, resulting in 2 out of 6 mice free of visible or palpable tumor (33% tumor- free) at the 60th day following tumor implantation. These data clearly indicate that the intake of high levels of MO not only significantly inhibited human mammary carcinoma MX-1 growth, but significantly enhanced the tumor responsiveness to MC treatment.

In Experiment 2, the same number of animals and same experimental diets were used. The only difference between these two experiments was that the time of the first MC treatment was postponed from the 10th day to the 15th day after tumor implantation in order to obtain large tumors to provide sufficient mass to measure tumor lipid peroxidation, protein oxidation, antioxidant enzyme activities, and MC-activating enzyme activities in the tumor. The results of this second study were similar to those from Experiment 1 (Table 2) in that the tumor growth was significantly suppressed by feeding MO diet and the tumor response to MC treatment was enhanced markedly by the MO diet. At the end of the second experiment (26 d after tumor implantation), the T/C ratios were 5 for MO group and 17 for CO group, indicating that the tumors from the MO-fed group were more sensitive to MC than the CO-fed group. There were no differences in host body weights between different dietary groups and MC treatment.

Tumor enzyme activities. Figure 1 shows the alterations of tumor MC bioreductive-activating enzyme activities by the different diets and MC treatment. In the present experiment, DTD activity was increased significantly by MO feeding (P < 0.001) and further increased by MC treatment (P < 0.05; Fig. 1 A) compared to CO feeding. MC treatment had no effect on DTD activity in the CO group (Fig. 1A). CB5R activity was not affected by the diet alone. After MC treatment, CB5R activity was increased in the MO group (MO + MC group, P < 0.01; Fig. IB) compared to the MO, CO + MC or CO groups. CCR activity was not altered by diet alone either, but was increased by MC treatment in both dietary groups (P < 0.01, Fig. IC). A similar pattern was observed for xanthine dehydrogenase (XD) as CCR (Fig. 1D). Xanthine oxidase (XO) was significantly increased by feeding a diet high in MO (P < 0.05) and further increased by MC treatment (Fig. IE) in both dietary groups. These results indicate that MO feeding and/or MC treatment changed tumor drug activating enzyme activities which are probably responsible for the enhanced antineoplastic effect of MC observed after feeding a high-MO diet.

Feeding MO which is rich in n-3 long-chain PUFAs will increase tumor membrane phospholipids (PLs) unsaturation by incorporating these highly unsaturated fatty acids into PLs (1-4). Figure 2 shows that dietary menhaden oil and MC treatment significantly altered tumor antioxidant enzyme activities. GPOx activity was increased by a high-n-3 PUFAs

diet (MO, P < 0.05) and further increased by MC treatment (P < 0.001 in MO and P < 0.05 in CO). Similar changes were observed for GSTPx activity (Fig. 2C). Both tumor CAT and SOD activities were elevated after feeding MO and further elevated by MC treatment (Figs. 2E and 2F). As we observed in a previous study (14), GST activity was not changed by either dietary MO or MC treatment (Fig. 2B). GR activity was inhibited by MC treatment, but increased by feeding the high- MO diet (P < 0.01; Fig. 2D). These findings, along with tumor lipid peroxidation and protein oxidation data (dis- cussed later), suggest that a high intake of n-3 PUFAs (MO in this study) causes an increase in endogenous oxidative stress in the tumor, presumably by changing the unsaturated fatty acid composition of tumor membrane lipids. This could be responsible for the observed suppressed tumor growth and enhanced the response to the prooxidant antitumor agents such as MC.

Tumor GSH. Tumor levels of GSH were measured after feeding the host animals with different diets, and the results are shown in Table 3. Tumor GSH levels were decreased by feeding the PUFA-rich MO diet (P < 0.05). Although the glutathione disulfide (GSSG) levels were not changed in the two different dietary groups, the GSH/GSSG ratio was significantly decreased in the MO group. These data indicate that feeding a diet rich in high levels of unsaturated PUFAs (such as MO) causes a decrease of GSH level in the tumor, which is consistent with increased tumor oxidative stress, and increased response to chemotherapy.

Lipid peroxidation and protein oxidation. The data in Fig- ure 3 demonstrate that after feeding an MO-rich diet, a significant increase in tumor lipid peroxidation and protein oxidation occurred. MC treatment further increased both tumor lipid peroxidation and protein oxidation, similar to the re- sponses of the antioxidant enzymes. Again, these results indicate that both dietary menhaden oil and MC treatment caused oxidative stress in this tumor model. These results are consistent with the observed decreased tumor growth and increased responsiveness to MC treatment.

Tumor fatty acid composition. As shown in Figure 4, tumor membrane PL fatty acid composition was changed by the different diets. There was a significant elevation of 18: ln- 9, 18:2n-6, and 20:4n-6 fatty acids in the tumor membranes from the CO group, whereas in the tumor membranes from the MO group, there were significant increases of n-3 PUFAs such as 18:3n-3, 20:4n-3, 20:5n-3, 22:5n-3, and 22:6n-3. This change in the fatty acid pattern reflects an increased degree of unsaturation, with an unsaturation index of 208 in the MO- fed groups compared to 147 in the CO-fed groups, and, therefore, susceptibility to oxidative stress in the tumor lipid from the MO-fed group compared to the CO-fed group (Fig. 4).

DISCUSSION

Fish oil has been reported to inhibit mammary tumorigenesis in a variety of animal models, including carcinogen-induced rat and mouse models, transplantable mammary tumor mod-

Lipids, Vol. 30, no. 11 (1995)


FIG. 1. Effects of feeding a diet high in menhaden oil (MO), or low in corn oil (CO) and mitomycin C (MC) treatment on tumor (A) DT-diaphorase (DTD), (B) cytochrome b s reductase (CB5R), (C) cytochrome C reductase (CCR), (D) xanthine dehydrogenase (XD), and (E) xanthine oxidase (XO) activities. Statistically significant differences were in (A) DTD activity between MO + MC and MO groups (a: P < 0.05), MO + MC and CO + MC groups (b: P < 0.001 ), and between MO and CO groups (c: P < 0.001 ); in (B) CB5R activity between MO + MC and MO groups (a: P < 0.01 ), between MO + MC and CO + MC groups (b: P< 0.01); (C) CCR activity between MO + MC and MO groups (a: P< 0.01) and between CO + MC and CO groups (b: P < 0.01 ); (D) in xanthine dehydrogenase activity between MO + MC and MO groups (a: P < 0.01) and between CO + MC and CO groups (b: P< 0.01); and (E) in xanthine oxidase activity between MO + MC and MO groups (a: P< 0.01), between MO + MC and CO + MC groups (b: P< 0.01), between MO and CO groups (c: P< 0.05), and between CO + MC and CO groups (d: P< 0.05).

els, or spontaneous mouse mammary tumor models (for review, see Refs. 1-4). Because fish oils are often deficient or marginal in the essential fatty acids, such as linoleic acid (18:2n-6), it is possible that the suppression of mammary tumorigenesis exerted by a high-fish oil diet may be due to in- sufficient amounts of linoleic acid. In fact, supplementation of fish oil diet with amounts of linoleic acid has been reported to block or partially block the inhibitory effects on mammary tumor development by a high-fish oil diet (33-36). In this study, we supplemented the high-fish oil diet with 5% corn oil, to provide essential fatty acid as recommended by the American Institute of Nutrition (15,16). We also observed a significant inhibitory effect on the growth of human mammary carcinoma by feeding a diet high in MO compared to a

low-CO diet (Table 2). Thus, the inhibitory effect of MO on human mammary carcinoma growth in this study should not be due to the lack of sufficient essential fatty acids in the MO diet.

Previous studies have shown that dietary fat influences membrane structure and function (1-4) and have also demonstrated that membranes with a high unsaturation index are more sensitive to peroxidation (37-40). In this study, MO feeding resulted in a higher amount of n-3 PUFAs being incorporated into tumor membrane PLs with a decreased incorporation of arachidonic acid (20:4n-6; Fig. 4). These long- chain PUFAs significantly increased the degree of tumor membrane lipid unsaturation compared to the CO group, with a calculated unsaturation index of 208 and 147, respectively.

Lipids, Vol. 30, no. 11 (1995)

1040 Y. SHAO ETAL.

FIG. 2. Effects of feeding a diet high in MO, or low in CO and MC treatment on tumor (A) glutathione peroxidase (GPOx), (B) glutathione S-transferase (GST), (C) glutathione S-transferase peroxidase (GSTPx), (D) glutathione reductase (GR), (E) catalase (CAT), and (F) superoxide dismutase (SOD) activities. Statistically significant differences were (A) GPOx activity between MO + MC and MO groups (a: P < 0.001 ), between MO + MC and CO + MC groups (b: P< 0.05), between MO and CO groups (c: P< 0.05), and between CO + MC and CO groups (d: P< 0.05); (C) in GSTPx activity between MO + MC and MO groups (a: P< 0.001), between MO + MC and CO + MC groups (b: P< 0.01), between MO and CO groups (c: P < 0.05), and between CO + MC and CO groups (d: P < 0.01 ); (D) in GR activity between MO + MC and MO groups (a: P < 0.01 ), between MO and CO groups (b: P< 0.01), and between CO + MC and CO groups (c: P< 0.01); (E) in CAT activity between MO + MC and MO groups (a: P < 0.001), between MO + MC and CO + MC groups (b: P< 0.01), between MO and CO groups (c: P< 0.05), and between CO + MC and CO groups (d: P< 0.01); and (F) in SOD activity between MO + MC and MO groups (a: P< 0.001), between MO + MC and CO + MC groups (b: P< 0.01), between MO and CO groups (c: P< 0.05), and between CO + MC and CO groups (d: P< 0.01). Abbreviation as in Figure 1.

This increased level of unsaturation in the MO group increased the tumor lipid susceptibility to lipid peroxidation and subsequent endogeneous oxidative stress, as demonstrated by the following indicators of oxidative stress. (i) A significant elevation in the amounts of lipid peroxidation products

(TBARS) and protein oxidation were observed in the tumor from the MO-fed group (Fig. 3). (ii) Antioxidant enzymes which participate in cellular protection against lipid peroxidation such as SOD, CAT, GPOx, and GSTPx were also activated by MO feeding (Fig. 2). (iii) Tumor GSH levels were

TABLE 3 Glutathione and Glutathione Disulfide Status in Human Mammary Carcinoma MX-la

Dietary group n b GSH GSSG GSH/GSSG

CO (5% CO) 6 2.20 • 0.13 0.10 + 0.01 22.0 _+ 0.5

MO (5% CO + 20% MO) 6 1.40 • 0.02 c 0.10 + 0.01 14.0 • 0.8 c

aValues are means + SD in pmol/g wet tumor wt for glutathione (GSH) and glutathione disulfide (GSSG). Other abbreviations as in Table 1. bNumber of mice per group. cp < 0.05, significant differences between CO and MO groups.

Lipids, Vol. 30, no. 11 (1995)


FIG. 3. Effects of feeding a diet high in MO, or low in CO and MC treatment on tumor (A) lipid peroxidation and (B) protein oxidation. Statistically significant differences were (A) in lipid peroxidation between MO + MC and MO groups (a: P< 0.01), between MO + MC and CO + MC groups (b: P < 0.01 ), between MO and CO groups (c: P < 0.01 ), and between CO + MC and CO groups (d: P < 0.01) and in (B) protein oxidation between MO + MC and MO groups (a: P< 0.00t), between MO + MC and CO + MC groups (b: P< 0.01), between MO and CO groups (c: P< 0.0l), and between CO + MC and CO groups (d: P< 0.01). Abbreviations as in Figure 1.

significantly decreased in the MO-fed group, indicating that high levels of dietary menhaden oil resulted in predisposing the tumors to oxidative stress. Although the mechanism by which oxidative stress inhibits tumor growth is not clear, lipid peroxidation products are reported to be capable of decreas- ing cell proliferation by damaging cell membranes, changing cellular structures, and inactivating cellular macromolecules such as key enzymes and other protein factors (41-44). How- ever, under extreme oxidative conditions, lipid peroxidation may lead to cell death (45-48). Therefore, the observed inhibition of tumor growth by MO in this study may in part be at- tributed to the increased susceptibility of tumor cells to lipid peroxidation by altering the level of unsaturation of their membranes. This conclusion is consistent with the findings of Gonzalez et al. (5) who related the inhibition of tumor growth in animals fed fish oil to lipid peroxidation by preventing the fish oil inhibition of tumor growth by feeding high levels of o~-tocopherol.

We previously reported that feeding a 25% CO diet also increased tumor oxidative stress as measured by increased tumor antioxidant enzyme activities and lipid peroxidation. However, contrary to findings reported here, in comparison to the 5% CO diet, the 25% CO diet enhanced human mam-

mary carcinoma growth (14). The reason for this difference in tumor growth between the 25% CO (14) and 25% MO (Table 4) dietary regimes may be the difference in lipid peroxidation found in the tumors fed the different high-fat diets. Under certain conditions, metabolites of PUFA peroxidation actually activate cellular proliferation (49), whereas the cyto- toxic effects of lipid peroxidation occur only under extreme oxidative conditions (4548). The data in Table 4 indicate that lipid peroxidation in the tumors from animals fed 25% CO increased 25% over the 5% CO controls, whereas in the 5% CO + 20% MO groups, lipid peroxidation was increased 300%. Therefore, we speculate that the differences between stimulating tumor growth in the 25% CO diet and inhibiting tumor growth in the 5% CO + 20% MO diet is the extent of lipid peroxidation and production of toxic peroxidation by- products. As reviewed by Esterbauer et al. (50), o)6 PUFAs, such as linoleic acid (C18:2) and arachidonic acid (C2o:4), yield 4-hydroxynonenal (HNE), 4-hydroperoxynonenal, and 4,5- dihydroxydecenal, whereas the m3 PUFAs, like docosahexaenoic acid (C22:6), yield 4-hydroxyhexenal, 4-hydroper- oxy hexenal, and 4,5-dihydroxy heptenal. The 4-hydroxy alkenals have widespread activity, including formation of cellular membrane bleb leading to cell death in Ehrlich ascites,

Lipids, Vol. 30, no. 11 (1995)

1042 Y. SHAO ETAL.

0 �9 r,,',,,I

r.r 0

O r._)

�9 e " ' l

r

30

25

20

15

10

5

0

m co

NM0

U n s a t u r a t i o n

1 4 7

208

I n d e x

C

a

C

la C

a

C

~.~,~.~,~.-, .~ ~ ,~ ~ , ~ ,~ ~, ~ u ~ ~ .~ ,u _,~ ,~ ,,~- ,,~ ,,~ ~ ~ r , , ~ _.~,~" ~.~," ~," _.~," ,~,-_.~r .~" ~.~," ~..~

I:lG. 4. Effects of feeding a diet high in MO and low in CO on tumor fatty acid compositions. Statistical differences between MO and CO groups were: a, P< 0.05; b, P< 0.01; and c, P< 0.001. The unsaturation index is calculated by multiplying the fatty acid percentage with the number of the double bond it has. Abbreviations as in Figure 1.

inhibition of glycolysis, respiration, depletion of GSH and protein thiois, protein syntheses, and DNA synthesis (51-56). Thus, the production of these toxic lipid peroxidation by- product could account for the depressed tumor growth observed by MO feeding.

Based on the different fatty acid distribution in the tumor from the CO- and MO-fed groups (Fig. 4), we would expect different lipid peroxidation products to be formed in the tumors from each group. The extent to which these different products are formed in each tumor has not yet been determined, and therefore their contribution to the different growth patterns of tumor from the two groups is at present unknown. On the other hand, CO contains high amounts of n-6 PUFAs, which can enhance the synthesis of prostaglandins after incorporation into the cell membranes; and prostaglandins have been proved to stimulate tumor proliferation (1--4). In fact, the enhanced tumor growth by high-n-6 PUFA diets could be partially reversed by the treatment of indomethacin; a prosta-

glandin synthase inhibitor. However, n-3 PUFAs (such as MO) have been shown to inhibit prostaglandin synthesis by either directly inhibiting prostaglandin synthase or competing for the substrate binding site with arachidonic acid (which is the natural substrate for prostaglandin synthase to produce prostaglandins) (1-4). In this study, MO feeding resulted in a threefold decrease of arachidonic acid levels in the tumor cell membrane compared to the 5% CO (Fig. 4). The decreased arachidonic acid and increased incorporation of n-3 PUFAs in the tumor membranes of the MO-fed animals may affect the production of prostaglandins in the tumor cells.

The most interesting finding in this study is the relationship between dietary fat and tumor chemotherapy. In our ear- lier study (14), a high-CO diet (25% CO) significantly increased the tumor response to MC chemotherapy. We believed that this was due to activation of the MC metabolizing system by the high-fat diet, because some well-known MC bioreductive-activating enzyme activities such as DTD, CB5R, and

Lipids, Vol. 30, no. 11 (1995)

MENHADEN OIL AND CHEMOTHERAPY

TABLE 4 Comparison Between the Effects of High-Corn Oil Diet (25%) and High-Menhaden Oil Die~

1043

HCO HCO + MC MO MO + MC

Tumor weight (mg) at day 30 T/C ratio b Combined T/C ratio c % of TBARS increase d DT-diaphorase activity (nmol/min/mg) Xanthine oxidase activity (pmol/min/mg)

4100 + 1950 200 + 69 2016 + 875 15 + 17 5 1 5 0.37

25 160 300 600 16.2 + 2.0 19.2 + 2.9 30.9 • 3.6 38.0 • 4.3 130 • 13 190 • 14 143 • 23 289 • 52

aData were collected from previous study (Ref. 14) and present study. HCO was the high-corn oil diet containing 25% corn oil in the diet; MO was the menhaden oil diet containing 20% menhaden oil + 5% corn oil. bThe percentage treatment/control ratio (T/C ratio) was calculated by the mean tumor weight of the MOtreated group divided by the mean tumor weight of the respective control group (MC untreated) multiplied by 100. C]-he combined T/C ratio was calculated by the mean tumor weight of the MC-treated groups divided by the mean tumor weight of the HCO control group multiplied by 100. dThe data were obtained by the mean thiobarbituric acid reactive substances (TBARS) value in the respective dietary groups (HCO or MO) divided by the mean TBARS value from the low-fat (5% corn oil) control group mutiplied by 100.

XO/XD activities were enhanced by the high-CO diet. In the present study, human mammary carcinoma MX- 1 was more sensitive to MC in the MO-fed group than in the control group (CO) (Table 2). MC treatment caused a significant depression in tumor growth compared to CO (Table 2) or 25% CO group (14) (Table 4). Tumor DTD and XO were also activated by dietary MO, but CB5R and XD activities were not increased (Fig. 1). These results indicate that feeding a high amount of dietary fat could alter the tumor susceptibility to MC treatment by enhancing certain drug activating enzyme systems directly in the tumor. A diet containing different fatty acids may have a different effect on various tumor enzymes. For example, we previously reported that feeding a high-CO diet (25% CO, w/w) increased the activities of CB5R and XD in addition to DTD and XO in human mammary carcinoma MX-1 (14); whereas in this study, the activities of CB5R and XD were not increased by high dietary MO. Since MO is associated with increased responsiveness to MC therapy, these findings tend to rule out the role of intratumor CB5R and XD in the MO-induced increased response to MC, although both enzyme activities were increased by treatment with MC. In this regard, the highest increase in MC-activating enzymes was XD, where treatment resulted in a threefold increase in tumor XD activity in MO-fed mice.

DTD catalyzes a two-electron reduction of MC resulting in the formation of the corresponding hydroquinone structure (10). Although it has been controversial, the role of DTD in the bioreductive activation of MC (57,58) has been clarified by using purified enzymes in cell-free systems. Metabolism of MC to DNA-reactive species by DTD is pH dependent in that it has been reported that the DNA cross-linking induced by MC increases as the intracellular pH is lowered. In addition, the bioreductive-activation of MC by DTD seems more efficient under aerobic conditions than under hypoxic conditions (for review, see Ref. 59). Both high levels of dietary CO (14) and MO increased endogeneous DTD activity in human mammary carcinoma MX-l, while increasing sensitivity to MC. Both diets support increased DTD activities following MC treatment, which support a major role for DTD in the

bioreductive activation of MC in this model. XO activity was increased in tumors from the MO-fed group, and MC treatment further increased XO activity. These findings indicate that XO also plays a role in MC activation. We cannot rule out XD as a possible activation enzyme of MC, in that it was always activated following MC treatment on either the 25% CO diet (14) or the MO diet (Fig. 1). Under aerobic conditions, hydroquinone can be reoxidized to the semiquinone free radical or quinone structure with subsequent formation of superoxide, hydrogen peroxide, and hydroxy radical (7). This results in increased oxidative stress in the tumor which could contribute to the antineoplastic activity of MC. In fact, Pritsos et al. (60) reported that the toxicity of MC toward EMT-6 cells was decreased in a dose-dependent manner with increasing activity of SOD, a well-known free radical scav- enge enzyme. In the present study, MC treatment caused significant levels of oxidative stress in the tumor from both MO and CO groups, as indicated by the increased antioxidant enzyme activities (SOD, CAT, GPOx, and GSTPx) and lipid peroxidation and protein oxidation: In addition, tumor GR was inhibited by MC treatment (Fig. 2D), which would be expected to increase tumor oxidative stress. This finding is consistent with the findings of others (61) that quinones can inhibit the activities of thiol containing enzymes. The role of inhibition of GR in the antitumor activity of MC requires further investigation, but is consistent with the decreased GSH levels in the tumors (Table 3). Collectively, these results indicate clearly that MC treatment caused oxidative stress in the tumor. As expected, the tumors from the MO-fed group were more susceptible to MC-induced oxidative stress because their membrane PLs were more unsaturated.

In summary, high intake of fats rich in long-chain n-3 PUFAs such as MO will increase the level of unsaturation in tumor membrane PLs. This results in tumor lipids possessing a high degree of unsaturation, which makes them more susceptible to endogenous and exogenous lipid peroxidation. When exogenous oxidative stress is induced by prooxidant drugs such as MC treatment, increased lipid peroxidation, protein oxidation, and increased responsiveness to MC

Lipids, Vo[. 30, no. 11 (1995)

1044 Y. SHAO ETAL.

chemotherapy is observed. In addition, increased tumor activities of DTD and XO, two of the proposed MC bioreductive activating enzymes, resulted from feeding a high-MO diet, indicating that nutritional intervention involving this type of dietary fat may increase the therapeutic response to prooxidant therapy.

REFERENCES

1. Welsch, C.W. (1992) Relationship Between Dietary Fat and Ex- perimental Mammary Tumorigenesis: A Review and Critique, Cancer Res. 52, 2040s-2048s.

2. Carroll, K.K. (1992) Dietary Fat and Breast Cancer, Lipids 27, 793-797.

3. Welsch, C.W. (1987) Enhancement of Mammary Tumorigene- sis by Dietary Fat: Review of Potential Mechanisms, Am. J. Clin. Nutr. 45, 192-202.

4. Pariza, M.W. (1988) Dietary Fat and Cancer Risk: Evidence and Research Needs, Ann. Rev. Nutr. 8,167-183.

5. Gonzalez, M.J., Schemmel, R.A., Dugan, L., Jr., Gray, J.I., and Welsch, C.W. (1993) Dietary Fish Oil Inhibits Human Breast Carcinoma Growth: A Function of Increased Lipid Peroxida- tion, Lipids 28, 827-832.

6. Kaasgaard, S.G., Holmer, G., Hoy, C-E, Behrens, W.A., and Beare-Rogers, J.L. (1992) Effects of Dietary Linseed Oil Ma- rine Oil on Lipid Peroxidation in Monkey Liver in vivo and in vitro, Lipids 27, 740-745.

7. Powis, G. (1989) Free Radical Formation by Antitumor Quinones, Free Radical Biology & Medicine 6, 63-101.

8. Sinha, B.K. (1989) Free Radicals in Antitumor Drug Pharma- cology, Chem. Biol. Interactions 69, 293-3 i 7.

9. Sinha, B.K., and Mimnaugh, E.G. (1990) Free Radicals and An- ticancer Drug Resistance by Certain Tumors, Free Radical Bi- ology & Medicine 8, 567-58 I.

10. Riley, R.J., and Workman, P. (1992) DT-Diaphorase and Can- cer Chemotherapy, Biochem. Pharmacol. 43, 1657-1669.

11. Yang, C.S., Brady, J.F., and Hong, J.Y. (1992) Dietary Effects on Cytochrome P450, Xenobiotic Metabolism, and Toxicity, FASEB J. 6, 737-744.

12. Yoo, J-S.H., Hong, J.Y., Ning, S.M., and Yang, C.S. (1990) Roles of Dietary Corn Oil in the Regulation of Cytochrome P450 and Glutathione S-Transferase, J. Nutr. 120, 17 i 8-1726.

13. Yoo, J-S.H., Ning, S.M., Pantuck, C.B., Pantuck, E.J., and Yang C.S. (i 991) Regulation of Hepatic Microsomal Cytochrome P450IIEI Level by Dietary Lipids and Carbohydrates in Rats, J. Nutr. 121,959-965.

14. Shao, Y., Pardini, L., and Pardini, R.S. (1994) Enhancement of the Antineoplastic Effect of Mitomycin C by Dietary Fat, Can- cer Res. 54, 6452-6457.

15. American Institute of Nutrition (1977) Report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nu- tritional Studies, J. Nutr. 107, ! 340-1348.

16. American Institute of Nutrition (1980) Second Report of the Ad Hoc Committee on Standards for Nutritional Studies, J. Nutr. 110, 1726.

17. Birt, D.F., White, L.T., Choi, B., and Pelling, J.C. (1989) Di- etary Fat Effects on the Initiation and Promotion of Two-Stage Skin Tumorigenesis in the SENCAR Mouse, Cancer Res. 49, 4170--4174.

18. Borgeson, C.E., Pardini, L.,. Pardini, R.S., and Reitz, R.C. (1989) Effects of Dietary Fish Oil on Human Mammary Carci- noma and on Lipid-Metabolizing Enzymes, Lipids 24, 290-295.

19. Ernster, L. (1967) DT-Diaphorase, Methods in Enzymology 10, 309-317.

20. Benson, A.M., Hunkler, M.J., and Talalay, P. (1980) Increase of NAD(P)H:Quinone Reductase by Dietary Antioxidants: Possi-

ble Role in Protection Against Carcinogenesis and Toxicity, Proc. Natl. Acad. Sci. USA. 77, 5216-5220.

21. Sukiman, S.A., and Stevens, J.B. (1976) Purification of Xan- thine Dehydrogenase: A Rapid Procedure with High Enzyme Yield, Arch. Biochem. Biophys. 258, 2040s-2048s.

22. Masters, B.S.S., Williams, C.H., and Kamin, H. (1967) The Preparation and Properties of Microsomal TPNH-Cytochrome C Reductase from Pig Liver, Methods in Enzymology 10, 565-573.

23. Yasukochi, Y., and Masters, B.S.S. (1976) Some Properties of a Detergent Solubilized NADPH-Cytochrome C (Cytochrome P450) Reductase Purified by Biospecific Affinity Chromatogra- phy, J. Biol. Chem. 251, 5337-5344.

24. Oberely, L.W., and Spitz, D.R. (1984) Assay of Superoxide Dis- mutase in Tumor Tissue, Methods in Enzymology 105, 457-464.

25. Aebi, H. (1984) Catalase in vitro, Methods in Enzymology 105, 121.

26. Strauss, R.G., Snyder, E.L., Wallace, P.D., and Rosenburg, T.G. (1980) Oxygen Detoxification Enzymes in Neutrophils of In- fants and Their Mothers, J. Lab. Clin. Med. 95, 897.

27. Prohaska, J.R., Oh, S-H., Hockstra, W.G., and Ganther, H.E. (1977) Glutathione Peroxidase: Inhibition by Cyanide and Re- lease of Selenium, Biochem. Biophys. Res. Comm. 74, 64-71.

28. Racker, E. (1955) Glutathione Reductase, Methods in Enzymol- ogy 2, 722.

29. Mannervick, B., and Guthenberg, C. (1981) Glutathione Trans- ferase: Human Placenta, Methods in Enzymology 77, 231-235.

30. Buege, J.A., and Aust, S.D. (1978) Microsomal Lipid Peroxida- tion, Methods in Enzymology LII, 302-310.

31. Levine, R.L., Garland, D., Oliver, C.N., Amici, A., Climent, I., Lenz, A-G., Ahn, B., Shalteil, S., and Stadtman, E.R. (1990) De- termination of Carbonyl Content in Oxidatively Modified Pro- teins, Methods in Enzymology 77, 231-235.

32. Fariss, M.W., and Reed, D.J. (1987) High-Performance Liquid Chromatography if Thiols and Disulfides: Dinitrophenol Deriv- atives, Methods in Enzymology 143, 101-109.

33. Aksoy, M., Berger, M. R., and Schmahl, D. (1987) The Influ- ence of Different Levels of Dietary Fat on the Incidence and Growth of MNU-Induced Mammary Carcinoma in Rats, Nutr. and Cancer 9, 227-235.

34. Gabor, H. and Abraham, S. (1986) Effect of Dietary Menhaden Oil on Tumor Cell Loss and the Accumulation of Mass of a Transplantable Mammary Adenocarcinoma in BALB/c Mice, J. Natl. Cancer lnst. 76, 1223-1229.

35. Hudson, E.A., Beck, S.A., and Tisdale, M.J. (1993) Kinetics of the Inhibition of Tumor Growth in Mice by Eicosapentaenoic Acid Reversed by Linoleic Acid, Biochem. Pharmac. 45, 2189-2194.

36. Ip, C., Carter, C.A., and Ip, M.M. (1985) Requirement of Essen- tial Fatty Acid for Mammary Tumorigenesis in the Rat, Cancer Res. 45, 1997-2001.

37. Burns, A., Lin, Y-G., Gibson, R., and Jamieson, D. (1991) The Effect of a Fish Oil Enriched Diet on Oxygen Toxicity and Lipid Peroxidation in Mice, Biochem. Pharmacol. 42, 1353-1360.

38. Javouhey-Donzel, A., Guenot, L., Naupoil, V., Rochette, L., and Rocquelin, G. (1993) Rat Vitamin E Status and Heart Lipid Per- oxidation: Effect of Dietary ~-Linolenic Acid and Marine n-3 Fatty Acids, Lipids 28, 651--655.

39. Chen, L-C., Boissonneault, G., Hayek, M.G., and Chow, C.K. (1993) Dietary Fat Effects on HepatieqSipid Peroxidation and Enzymes of H202 Metabolism and NADPH Generation, Lipids 28, 657-662.

40. Cho, S-H., and Choi, Y-S. (1994) Lipid Peroxidation and An- tioxidant Status Is Affected by Different Vitamin E Levels When Feeding Fish Oil, Lipids 29, 47-52.

41. Gonzalez, M.J. (1992) Lipid Peroxidation and Tumor Growth: An Inverse Relationship, Med. Hypotheses 38, 106-110.

Lipids, Vol. 30, no. 11 (1995)

MENHADEN OIL AND CHFMOTHERAPY 1045

42. Spector, A.A., and Yorek, M.A. (1985) Membrane Lipid Com- position and Cellular Function, J. Lipid Res. 26, I 015-1035.

43. Spector, A.A., and Bums, C.P. (1987) Biological and Therapeu- tic Potential of Membrane Lipid Modification in Tumors, Can- cer Res. 47, 45294-4537.

44. Farber, J.L., Kyle, M.E., and Cleman, J.B. (1990) Mechanisms of Cell Injury by Activated Oxygen Species, Lab. Invest. 62, 670-679.

45. Roubal, W.T. and Tappel, A.L. (1966) Damage to Proteins, En- zymes, and Amino Acids by Peroxidizing Lipids, Arch. Biochem. Biophys. 113, 5--8.

46. Reiss, U., and Tappel, A.L. (1973) Fluorescent Product Forma- tion and Changes in Structure of DNA Reacted with Peroxidiz- ing Arachidonic Acid, Lipids 8, 199-202.

47. Dianzani, M.U. (1982) in Free Radicals, Lipid Peroxidation and Cancer (McBrien, D.C., and Slater, T.F., eds.) pp. 129-158, Academic Press, London.

48. Bull, A.W., Nigro, N.D., and Marnett, J.J. (1988) Structural Re- quirements for Stimulation of Colonic Cell Proliferation by Ox- idized Fatty Acids, Cancer Res. 48, 1771-1776.

49. King, M.M., Bailey, D.M., Gibson, D.D., Pitha, J.V., and McCay, P.B. (1979) Incidence and Growth of Mammary Tu- mors Induced by 7,12-Dimethyl-benz[a]anthracene as Related to the Dietary Content of Fat and Antioxidant, J. Natl. Cancer Inst. 63, 657--663.

50. Esterbauer, H., Schaur, R.J., and Zollner, H. (1991) Chemistry and Biochemistry of 4-Hydroxynonenal, Malonaldehyde, and Related Aldehydes, Free Rad. Biol. Med. 11, 81-129.

51. Schauenstein, E. (1982) Effects of Low Concentrations of Alde- hydes on Tumor Cells and Tumor Growth, in Free Radicals, Lipid Peroxidation and Cancer (McBrien, D.C., and Slater T.F., eds.) pp. 159-171, Academic Press, London.

52. Khoschsorur, G., Schaur, R., Schauenstein, E., Tiilian, H.M., and Reiter, M. (I 98 I) Intra-Ceilular Effects of Hydroxy Alke- nals on Animal Tumors, Z. Naturforsch 36, 572-578.

53. Haupthlorenz, S., Esterbauer, H., Moll, W., Pumpel, R., Schauenstein, E, and Puschendorf, B. (1985) Effects of the Lipid

Peroxidation Product 4-Hydroxy Nonenal and Related Aldehy- des on Proliferation and Viability of Cultured Ehrlich Ascites "rumor Cells, Biochem. Pharmac. 34, 3803-3809.

54. Bickis, I. J., Schauenstein, E., and Taufer, M. (1968) Effects of Hydroxypentenal on the Metabolism of Tumor Cells and Nor- mal Cells, Monatsa. Chem. 100, 1077-1104.

55. Cadenas, E., Muller, A., Brigelius, R., Esterbauer, A., and Sies, H. (1983) Effects of 4-Hydroxy Nonenal on Isolated Hepato- cytes, Biochem. J. 214, 479487.

56. tlaenen, G.R.M.M., Tai Tin Tsoi, J.N.L., Vermeulen, N.P.E., Timmermann, A., and Bast, A. (1987) Hydroxy-2,3-trans-nonenal Stimulates Microsomal Lipid Peroxidation by Reducing the Glutathione-Dependent Protection, Arch. Biochem. Biophys. 259, 4494-459.

57. Workman, P., Walton, M.I., Powis,, G., and Schlager, J.J. (1989) DT-Diaphorase: Questionable Role in Mitomycin C Re- sistance, But a Target for Novel Bioreductive drugs? Br. J. Can- cer 60, 800-802.

58. Marshall, R.S., Rauth, A.M., and Paterson, M. (1989) Reply to the Letter from Workman et al., Br. J. Cancer 60, 803.

59. Ross, D., Siegel, D., Beall, H., Prakash, A.S., Mulcahy, R.T., and Gibson, N.W. (1993) DT-Diaphorase in Activation and Detoxification of Quinones. Bioreductive Activation of Mito- mycin C, Cancer Metastasis Rev. 12, 83-101.

60. Pritsos, C.A., Keyes, S.R., and Sartorelli, A.C. (1989) Effect of the Superoxide Dismutase Inhibitor, Diethyldithiocarbamate, on the Cytotoxicity of Mitomycin Antibiotics, Cancer Biochem. Biophys. 10, 289-298.

61. Takahashi, N., Schreiber, J., Fischer, V., and Mason, R.P. (1987) Fromation of Glutathione-Conjugated Semiquinones by the Reaction of Quinones with Glutathione: An ESR Study, Arch. Biochem. Biophys. 252, 414-48.

[Received May 1, 1995, and in final revised form August 28, 1995; Revision accepted September 5, 1995]

Lipids, Vol. 30, no. 11 (1995)

dietary menhaden oil enhances mitomycin c antitumor activity toward human mammary carcinoma mx-1

Documents