QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Kubo, K. Articles by Maekawa, A. Search for Related Content PubMed PubMed Citation Articles by Kubo, K. Articles by Maekawa, A.

---

Article

Preferential Incorporation of Docosahexaenoic Acid into Nonphosphorus Lipids and Phosphatidylethanolamine Protects Rats from Dietary DHA-Stimulated Lipid Peroxidation¹

Division of Food Science, The National Institute of Health and Nutrition, Shinjuku-ku, Tokyo 162-8636, Japan and ^* Laboratory of Nutritional Biochemistry, Department of Applied Biology and Chemistry, Tokyo University of Agriculture, Setagaya-ku, Tokyo 156-0054, Japan

³To whom correspondence should be addressed.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In a previous study, we found that dietary docosahexaenoic acid (DHA)-stimulated tissue lipid peroxide formation was suppressed to a lesser extent than expected from the peroxidizability index of tissue total lipids. This suppression was presumed to be potentiated by mechanisms other than the lipid peroxide–scavenging system. In this study, we focused primarily on the incorporation of DHA into tissue nonphosphorus lipids and phospholipid species. DHA and different levels of dietary vitamin E (VE; 7.5, 54, 134 and 402 mg/kg of diet) were fed to rats for 32 d. In rats with poor VE status, liver chemiluminescence intensity and kidney and testis thiobarbituric acid (TBA) values correlated with the tissue’s peroxidizability index. In rats with normal VE nutriture, liver lipid peroxide formation was suppressed to a level below that expected from the peroxidizability index, likely because DHA was present in nonphosphorus lipids and utilized preferentially for phosphatidylethanolamine synthesis. In the kidney, differences in the TBA values were associated with differences in the peroxidizability index of total lipids, even in the DHA groups fed VE at higher than normal levels. This may be because the levels of lipid peroxide scavengers were lower than those of liver and because DHA was utilized preferentially for phosphatidylcholine synthesis. In testis, the lipid peroxide levels were not as high as expected from the peroxidizability index, even in rats fed a high DHA diet containing the normal level of VE. This may be because the testis was composed of a high proportion of (n-6) polyunsaturated fatty acids (PUFA), which are low in unsaturation, and thus the proportion of DHA was low. In addition, in testis, VE and ascorbic acid, which act as antioxidants, were retained at higher levels in rats with particularly poor and normal VE nutriture than those of liver and kidney. These results suggest that antioxidant protection against dietary DHA-stimulated lipid peroxidation below the extent expected from the peroxidizability index of tissue total lipids differed from tissue to tissue. The suppression was likely due to not only the lipid peroxide scavenging system but also preferential incorporation of DHA into nonphosphorus lipids and phosphatidylethanolamine, particularly in liver.

KEY WORDS: • DHA • lipid peroxide • peroxidizability index • phospholipids • rats

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The ingestion of docosahexaenoic acid (DHA)⁴ [22:6(n-3)] (Kubo et al. 1997 and 1998 , Saito et al. 1996 ), usually as fish oil (Hammer and Wills 1978 , Hu et al. 1989 , Kobatake et al. 1983 , Mouri et al. 1984 ), enhances the susceptibility of liver and kidney to lipid peroxidation. The enhancement is a function of the dietary DHA level and increases the requirement for vitamin E (VE), a lipophilic membrane antioxidant. This was thought to be attributable to the substitution of membrane fatty acids with DHA, which is very susceptible to lipid peroxidation and thus potentially unstable (Kubo et al. 1998 ).

Lipid peroxide scavengers suppress lipid peroxide formation in tissues, and VE acts as a major lipophilic membrane antioxidant. We showed, however, that the DHA-stimulated lipid peroxidation in the liver and kidney was repressed minimally, even when a high level of VE was ingested (Kubo et al. 1997 ). Accordingly, we proposed that there is a limit to the antioxidative action of VE. This was also suggested in other reports (Farwer et al. 1994 , Kaasgaard et al. 1992 ). Nevertheless, DHA ingestion with a usual level of VE did not promote tissue lipid peroxide formation to the degree expected from the peroxidizability index calculated from the fatty acid composition of tissue total lipids in rats (Kubo et al. 1997 and 1998 ). This phenomenon was especially prominent in the liver. In the brain and testis, the lipid peroxide levels were not increased in rats fed DHA (Kubo et al. 1998 ). Therefore, DHA-stimulated lipid peroxide formation differs from tissue to tissue. Protection against lipid peroxidation is not achieved by antioxidant VE alone; presumably, other factors exist that suppress DHA-stimulated tissue lipid peroxide formation to a level below that expected from the peroxidizability index of the tissue (Kubo et al. 1998 ). The suppressive mechanism is explained in part by increases in the levels of lipid peroxide scavengers such as ascorbic acid and glutathione (GSH) (Kubo et al. 1997 and 1998 ), which augment the antioxidative activity of VE.

The inner and outer aqueous tissue cell microenvironments are separated by cell membranes. In general, DHA, which is highly susceptible to lipid peroxidation, is localized in phospholipids of cell membranes, particularly in phosphatidylethanolamine (PE) and phosphatidylserine (PS) (Holub and Kuksis 1978 ). Polyunsaturated fatty acids (PUFA) dispersed in an aqueous solution and/or micelles have been reported to become further resistant to peroxidation as the number of double bonds in the fatty acids increases (Miyashita et al. 1993 , Yazu et al. 1996 ). Kashima et al. (1991) observed that the oxidative stability of perilla oil in the presence of tocopherols was increased by the addition of PE. The peroxidation of PUFA such as DHA is thought to be influenced by the microenvironment in which the PUFA exist, and the observations above (Kashima et al. 1991 , Miyashita et al. 1993 , Yazu et al. 1996 ) are very interesting when peroxidation of PUFA in vivo is considered. Accordingly, it appears that the location of PUFA, that is, the distribution in different lipid classes as well as in different tissues, should be taken into account when lipid peroxide formation in tissues is investigated.

In this study, an excess of DHA was given to rats (Kubo et al. 1998 , Saito et al. 1996 ) to study dietary DHA-stimulated tissue lipid peroxide formation. The influence of different levels of dietary VE on the lipid peroxide formation in certain tissues over an extended feeding period was also examined. In addition, the relation between tissue lipid peroxide formation and incorporation of DHA into tissue nonphosphorus lipids (NPL) and phospholipid species was analyzed to clarify the mechanisms involved in suppressing dietary DHA-stimulated lipid peroxide formation in tissues.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Animals and diets.

The experimental procedures used in this study met the guidelines of animal committee of The National Institute of Health and Nutrition (Tokyo, Japan).

Male Sprague-Dawley rats (Japan SLC, Hamamatsu, Japan), 5 wk of age and weighing 101–122 g, were housed individually in stainless steel wire-bottomed cages at a constant temperature of 22 ± 1°C and humidity of 50–60% with a 12-h light:dark cycle. The composition of the experimental diets, based on the AIN-76 purified diet for rats (AIN 1977 and 1980 ), is shown in Table 1 . DHA ethyl esters (83% pure) prepared from sardine oil were donated by Maruha Corporation (Tsukuba, Japan). To prevent the autoxidation of DHA in the diets, they were prepared beforehand without adding DHA and stored at -20°C. DHA was stored at -80°C and was mixed with the diet every day immediately before feeding. The VE concentration as RRR--tocopherol equivalent of the control diet [linoleic acid (LA) group] was 54 mg/kg, and those of the test diets (DHA groups) were 7.5, 54, 134 and 402 mg/kg, respectively. The relative biological activity of RRR--, RRR-ß-, RRR-- and RRR--tocopherols were assumed to be 100, 25, 5 and 0.1, respectively, in the calculation (Mino et al. 1988 ). All-rac--tocopheryl acetate (>99% pure), purchased from Kawai Pharmaceutical (Tokyo, Japan), was used to adjust the VE concentration in the diets. The fatty acid composition (g/100 g fatty acids) of dietary lipids is also indicated in Table 1 . The control lipid, devoid of DHA, contained 41.4% LA [18:2(n-6)], which is comparable to the DHA level of 40.2% in the 8.7 energy % (en%) diet. In addition, test lipids in the 8.7 en% diet were prepared to contain 2.1 en% of LA as an essential fatty acid of the (n-6) series in which the proportion of LA was 9.7%. The degree of unsaturation of dietary lipids is presented as the double-bond index (DBI) (Pietrangelo et al. 1990 ) and peroxidizability index (Hu et al. 1989 ). It has been reported that the relative reaction rate of peroxidation was 1, 2, 3, 4 and 5 against PUFA in which the number of the methylene groups among double bonds was 1, 2, 3, 4 and 5, respectively (Cosgrove et al. 1987 ). The peroxidizability index of lipids, therefore, is calculated according to the following equation (Hu et al. 1989 ):

After rats were fed the basal diet containing 5 g olive oil/100 g diet for 3 d, 6–7 rats of each group were fed the experimental diets for 32 d. Food and water were consumed ad libitum. Each diet was made available to the rats in the evening and removed the next morning. After being deprived of food overnight, the rats were killed by cardiac puncture. The tissues were promptly excised, washed with isotonic saline and weighed. The liver was then perfused with ice-cold isotonic saline via the portal vein. The liver, kidney and testis samples were stored at -80°C until needed for analysis. Serum was separated by centrifugation at 2700 x g for 15 min at 4°C.

The liver conjugated dienes were determined by the method of Hu et al. (1989) and Rao and Recknagel (1968) . The conjugated dienes were measured at 233 nm using an extinction coefficient of 27,000 (mol/L)^-1 · cm^-1.

Chemiluminescence intensity analysis.

The liver, kidney and testis chemiluminescence intensities of the homogenates were determined according to the method of Miyazawa et al. (1984) . The light emitted from the homogenates is due mainly to singlet molecular oxygen and/or excited carbonyl compounds resulting from the breakdown of lipid peroxyradicals (Boveris et al. 1981 , Miyazawa et al. 1981 ), which are produced in the early stage of peroxidation. The chemiluminescence intensity is indicated in terms of the average counts per 30 s for 7-min measurements and is corrected for the background count.

Thiobarbituric acid (TBA) value analysis.

The serum TBA value was determined by the method of Yagi (1976) . BHT as an antioxidant was added to the reaction mixture at a final concentration of 0.36 mmol/L. The liver, kidney and testis TBA values were measured according to the method of Ohkawa et al. (1979) with a minor modification, in which BHT was added to the reaction mixture at a final concentration of 0.45 mmol/L. TBA values are expressed in terms of the malondialdehyde equivalent.

Fluorescent substance analysis.

Serum water-soluble fluorescent substances were analyzed by the method of Tsuchida et al. (1985) . Liver microsomes were prepared (Saito and Yamaguchi 1988 ), and the microsomal lipofuscin content was determined by the method of Fletcher et al. (1973) . The microsomal protein content was measured by the method of Lowry et al. (1951) .

Vitamin E analysis.

The vitamin E (-tocopherol) levels in the test lipids, serum and tissues were analyzed by HPLC as described (Saito et al. 1992 ).

Total ascorbic acid and nonprotein sulfhydryl (SH) assays.

Total ascorbic acid (Roe et al. 1948 ) and nonprotein SH (Beutler et al. 1963 ) levels in liver, kidney and testis were measured. The nonprotein SH component consists mostly of GSH.

Selenium-dependent glutathione peroxidase assay.

Selenium-dependent glutathione peroxidase (EC 1.11.1.9; GSHPx) activity was determined according to the method of Noguchi et al. (1973) with a minor modification as described (Saito 1990 ). One unit is equivalent to the disappearance of 1% of the substrate (GSH) per min.

Serum aspartate aminotransferase and alanine aminotransferase assays.

The activities of aspartate aminotransferase (EC 2.6.1.1; AST) and alanine aminotransferase (EC 2.6.1.2; ALT) in the serum were determined with a clinical enzyme assay kit (Wako Pure Chemical, Osaka, Japan) by the method of Reitman and Frankel (1957) .

Lipid analysis.

Liver, kidney and testis from 2 -3 rats were pooled and three samples per group were prepared for every tissue. Tissue total lipids were extracted from each tissue according to the method of Folch et al. (1957) . Total lipids from the tissues were separated into phospholipids and NPL in silica Sep-Pak columns (500 mg) obtained from Waters Associates (Milford, MA) (Juaneda and Rocquelin 1985 ). The NPL concentrations were measured gravimetrically, and the molar concentration was calculated using triolein’s molecular weight of 885.45. The phospholipids were separated into phospholipid classes, i.e., PE, phosphatidylcholine (PC) and a mixture of PS and phosphatidylinositol (PS + PI) by TLC on Merck type 60 silica gel plates (20 x20 cm) with a 0.5-mm layer (Merck A.G., Darmstad, Germany). L--PE (dioleoyl, synthetic), L--PC (dioleoyl, synthetic), L--PI (ammonium salt, from bovine liver) and L--PS (from bovine brain) were purchased from Sigma Chemical (St. Louis, MO) and were used as the reference standards for TLC. After the plates had been developed in chloroform/methanol/ammonium hydroxide (280 g/L) solution (65:35:4.5), the phospholipids were detected under UV light after spraying with rhodamine 6G solution. Each phospholipid was scraped off and extracted with chloroform/methanol (1:2 v/v) and then with methanol solutions. Each phospholipid class was determined according to the method of Stewart (1980) . That is, each phospholipid was measured colorimetrically by forming a complex with ammonium ferrothiocyanate, and calibration graphs were prepared for all of the phospholipids by using individual phospholipid standards obtained from Sigma Chemical. Only the PS standard was used for the measurement of a mixture of PS + PI; the molar concentration was calculated using a molecular weight of 774.0. The fatty acid compositions of dietary lipids, tissue total lipids, phospholipid species and NPL were analyzed by gas-liquid chromatography (GLC) (Saito et al. 1990 ).

Fatty acid methyl esters (FAME) of dietary lipids and tissue lipids were prepared as follows: lipids were saponified with 0.5 mol/L sodium hydroxide/methanol solution and the resultant free fatty acids were converted into methyl esters by using boron trifluoride/methanol solution (140 g/L). The methyl esters were extracted with n-hexane and analyzed by GLC with dual-flame ionization detectors (Hitachi 263–50 gas-liquid chromatograph, Tokyo, Japan) by using a 3 m x 2.5 mm i.d. glass column containing 5% Advance DS on 80–100 mesh Chromosorb W. The column temperature was 195°C. Injector and detector temperatures were 260°C. Nitrogen gas was employed as the carrier gas. Standard mixtures of FAME, lauric acid, myristic acid, (n-7) myristoleic acid, palmitic acid, (n-7) palmitoleic acid, stearic acid, (n-9) oleic acid, (n-6) linoleic acid, (n-3) linolenic acid, (n-6) eicosadienoic acid, (n-3) eicosatrienoic acid, (n-6) arachidonic acid, (n-6) docosadienoic acid and (n-3) DHA (Nihon Chromato Works, Tokyo, Japan) and (n-3) eicosapentaenoic acid (EPA), (n-6) docosatetraenoic acid and (n-3) docosapentaenoic acid (Supelco, Tokyo, Japan) were used for identification of peaks.

Statistical analysis.

After confirming the normality of data and the homogeneity of variance of data for the treatment groups (the latter being evaluated by the Bartlett test), the significance of differences between mean values was assessed by ANOVA coupled with Duncan’s multiple-range test at the 5% level of significance (Duncan 1957 ).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The rats consumed 13.8–16.4 g food/d and gained 4.4–5.7 g/d over the 32-d experiment. Food intake was significantly lower in the 7.5 mg/kg VE group than the LA group (control group) (Table 2 ). Body weight gain was also significantly lower in the 7.5 mg/kg VE group than in the LA group. Relative liver weight of the 7.5 mg/kg VE group was higher than that of the LA group; relative kidney weights of the 7.5 and 134 mg/kg VE groups were slightly but significantly higher than those of the control group, and the relative testis weight of the 402 mg/kg VE group was slightly but significantly lower than that of the control (Table 2) .

View this table: [in this window] [in a new window]   Table 3. Influence of the level of dietary vitamin E (VE) on thiobarbituric acid (TBA) value, water-soluble fluorescent substance level, -tocopherol level and aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activities in serum of rats fed docosahexaenoic acid (DHA)12

The testis -tocopherol levels of the DHA groups were significantly lower than the control level, even at the highest dose of VE (Table 6) . However, the levels of -tocopherol in the 7.5 and 54 mg/kg VE groups were higher than those of liver and kidney. The ascorbic acid and nonprotein SH concentrations and GSHPx activity in testis did not differ in the DHA groups compared with the LA group.

In liver, the level of NPL was significantly lower than that of controls in DHA-fed rats given >54 mg/kg VE (Table 7 ). In contrast, the liver PE level was greater or generally greater in the DHA-fed groups than in the LA-fed group. The PS + PI levels in the DHA-fed groups also were generally higher than those in the LA-fed group, whereas no significant difference in the PC level was noted among all of the groups. When the level of each phospholipid was compared, those of PE and PS + PI were almost the same and that of PC was ~2 times that of PE or PS + PI. In kidney and testis, no significant influences of DHA or dietary VE were found on the levels of any of the lipid classes (data not shown). Therefore, only the results for the LA group and DHA group fed 54 mg/kg VE are shown. The PE and PS + PI levels in the kidney were similar to those in the liver, but the PC level was less than two thirds of the PE and PS + PI levels. In testis, the levels of NPL and PE, PC and PS + PI were almost the same in all the groups. When the lipid levels of each tissue were compared, the levels of the phospholipid classes and NPL were greatest in the liver and smallest in the testis.

In testis (Table 9) , the proportions of (n-6) docosapentaenoic acid (DPA) [22:5(n-6)] were characteristically higher than in other tissues. The total proportions of (n-6) PUFA were also higher in testis than in other tissues. When the LA and DHA groups were compared, DHA was higher in each phospholipid class of the DHA groups, and AA and (n-6) DPA were lower in the same groups. The DBI and peroxidizability index were slightly higher in PE than those of other phospholipid classes, but DHA intake had a negligible influence on them in each phospholipid class. DHA levels were comparatively high in PE of the DHA group, whereas AA and (n-6) DPA levels were high in PE of the LA group.

The relation between peroxidizability indices and lipid peroxide levels in the liver, kidney and testis expressed relative to the control values is shown in Figure 1 . In liver, the peroxidizability indices of total lipids (fatty acid composition not shown) were 2.3–2.4 times the control value when rats were fed DHA. The liver chemiluminescence intensity for the 7.5 mg/kg VE group was 2.3 times the control value and coincided with the peroxidizability index ratio for liver total lipids (Fig. 1A ). On the other hand, values of TBA and conjugated dienes in the DHA groups were 1.7–1.9 and 1.3–1.4 times control, respectively, regardless of the dietary VE level. Accordingly, they were lower than the levels expected from the peroxidizability indices of liver total lipids. The peroxidizability index ratios of each phospholipid class in the 54 mg/kg VE group seemed to coincide with those of conjugated diene and chemiluminescence intensities (Fig. 1A ).

View larger version (34K): [in this window] [in a new window]   Figure 1. Lipid peroxide levels and peroxidizability indices of total lipids (TL) and phospholipid classes expressed relative to control values [linoleic acid (LA) group] in liver, kidney and testis of rats fed docosahexaenoic acid (DHA) and various levels of vitamin E. Peroxidizability index (P-index) was used in this study as an indicator to determine the vulnerability of tissue lipids to oxidation. The P-index of lipids is calculated according to the following equation: P-index = (dienoic % x 1) + (trienoic % x 2) + (tetraenoic % x 3) + (pentaenoic % x 4) + (hexaenoic % x 5). Values for P-index are the mean for three pooled samples from seven rats for 7.5, 54 and 134 mg/kg VE groups and six rats for 402 mg/kg VE group. Conjugated diene levels, chemiluminescence intensities, thiobarbituric acid (TBA) values and microsomal lipofuscin levels are represented by shaded, open, closed and slanting bars, respectively. Values are means ± SD. Means for the same analytical item in each tissue that are not followed by a common letter are different, P < 0.05.

In testis, peroxidizability indices of total lipids (fatty acid composition not shown) were 1.4–1.6 times control in the DHA groups. The TBA value for the 7.5 mg/kg VE group was 1.5 times control, and the level seemed to coincide with that of the peroxidizability index ratio for total lipids. On the other hand, the TBA values were 0.9–1.2 times control in the DHA groups fed >54 mg/kg VE, and were lower than the level expected from the peroxidizability index of testis total lipids (Fig. 1C ). The testis chemiluminescence intensity was lower than the level expected from the peroxidizability index of total lipids, regardless of the dietary VE level (Fig. 1C ). However, in each phospholipid class for DHA-fed groups, the peroxidizability indices were 0.9–1.0 times control and comparable to those of the LA group. These levels seemed to be particularly associated with the levels of TBA values in the DHA groups fed >134 mg VE (Fig. 1C ).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In a previous short-term (15 d) feeding study (Kubo et al. 1998 ), we found that dietary DHA-stimulated tissue lipid peroxide formation was suppressed below the extent expected from the peroxidizability index of tissue total lipids. This suppression was presumed to be potentiated by some mechanisms other than the lipid peroxide-scavenging system, because the lipid peroxide levels in the tissue did not change significantly in response to increasing levels of dietary VE and tissue ascorbic acid and GSH (Kubo et al. 1997 and 1998 ).

In this study, therefore, we examined the relationship between tissue lipid peroxide formation and incorporation of DHA into tissue NPL and phospholipid species to clarify further the mechanisms mentioned above.

We employed DHA that was high in purity (83%) in this study; the remaining 17% (as impurities) was composed of PUFA such as EPA and (n-3) DPA as indicated in Table 1 . Changes in lipid peroxides and their scavengers are assessed as general influences caused by PUFA, including DHA and its impurities. However, dietary levels of PUFA from the impurities were very low compared with DHA (Table 1) ; in addition, EPA and (n-3) DPA as major impurities are converted to DHA in vivo. Therefore, we assessed primarily the influences of dietary DHA in this study.

The results in this study showed that the serum TBA concentration decreased with increasing dietary VE levels (Table 3) , returning to the control level in rats fed 402 mg/kgVE. This was not observed in the short-term (15 d) experiments (Kubo et al. 1997 and 1998 ). Therefore, an antioxidative effect of high dietary VE on lipid peroxide formation became apparent in the serum when the ingestion period was prolonged to 32 d.

In liver, a tendency similar to that for DHA ingestion over 15 d (Kubo et al. 1998 ) was found at one dose of VE (54 mg/kg diet), i.e., the conjugated diene level, chemiluminescence intensity and TBA values were generally higher (Table 4) , and the -tocopherol level lower (Table 6) when rats were fed DHA. In DHA-fed rats given >54 mg VE/kg diet, however, activities of ALT and AST in the serum were not higher but rather lower with the increase in dietary VE (Table 3) . Hence, it was thought that the lipid peroxidation was not so stimulated so as to induce tissue parenchymal cell damage. On the other hand, in the rats given as little as 7.5 mg VE/kg diet, the lipid peroxide levels in the serum and tissues were remarkably higher, and the liver microsomal lipofuscin level was generally higher (Table 4) . The serum AST activity was also significantly higher in the low VE group (Table 3) , and yellow fat in the perirenal adipose tissue was observed on autopsy (data not shown). Charnock et al. (1986) reported that when fish oil was administered to rats, yellow fat was found in the storage fat in which lipofuscin pigments had accumulated, but not in the liver or kidney. Lipofuscin is defined as an end product of lipid peroxidation and accumulates in the body as a physiologically nonactive component. When rats are fed a diet containing highly unsaturated fatty acids such as DHA but devoid of VE or very low in VE, tissue cell injury is promoted by enhanced lipid peroxidation and the end products of lipid peroxidation accumulate in storage fat.

As shown in Figure 1A , the liver TBA value, even in the lowest dietary VE group, did not increase to the extent expected from the peroxidizability index of liver total lipids, whereas the chemiluminescence intensity in the lowest VE group seemed to be closely associated with the change in the peroxidizability index. In chemiluminescence analysis, the light emitted is due mainly to singlet molecular oxygens and/or excited carbonyl compounds resulting from the breakdown of lipid peroxy radicals (Boveris et al. 1981 , Miyazawa et al. 1981 ). At low dietary VE, which approximates VE deficiency, the chain reaction of lipid peroxidation to form peroxy radicals of fatty acids such as highly unsaturated DHA may not be suppressed easily.

-Tocopherol level in the liver increased greatly with increasing dietary VE level, and the levels of both ascorbic acid and GSH in the liver also increased with dietary VE (Table 6) . A reductive reaction by ascorbic acid for the -tocopheroxyl radical is known (Meister 1992 , Tappel 1962 , Wells et al. 1995 ). Biosynthesis of ascorbic acid and GSH may be promoted as the lipid peroxidation–induced increase in the requirement for VE is enhanced in the liver. Moreover, the TBA values and conjugated diene levels were not significantly lower in rats with higher intakes of VE (Table 4) . Therefore, the insufficient antioxidative function of the lipid peroxide–scavenging system in the liver was reconfirmed (Kubo et al. 1997 ). Farwer et al. (1994) observed similar results in rats fed diets high in fish oil.

The cytosolic, selenium-dependent GSHPx activity in the DHA-fed rats given >54 mg VE/kg diet did not change (Table 6) , regardless of the fact that the tissue lipid peroxide levels were higher in the DHA-fed rats (Table 4) . In addition, in the low VE group in which the lipid peroxides were formed, GSHPx activity decreased (Table 6) ; the explanation is not clear.

The TBA values and conjugated dienes in the liver were not suppressed significantly, even after the higher intake of VE. Furthermore, the liver lipid peroxide levels in the rats fed more VE than usual were, as in our former report (Kubo et al. 1998 ), lower than the extent expected from the peroxidizability index calculated from the fatty acid composition of liver total lipids (Fig. 1A ). Accordingly, the relationship cannot be explained merely by the levels of VE, ascorbic acid and GSH, and GSHPx activity. Mechanisms protecting against dietary DHA-induced lipid peroxidation other than the lipid peroxide–scavenging system may exist. To clarify this, as was mentioned in the introduction (Kashima et al. 1991 , Miyashita et al. 1993 , Yazu et al. 1996 ), changes in PUFA profiles in liver lipid classes were examined, focusing in particular on the proportion of DHA.

The level of NPL in liver, most of which are triacylglycerols, was much greater than in other tissues (Table 7) . In liver, triacylglycerol is accumulated in the cytoplasm of cells, mainly as temporary storage fat (Cook 1958 ). Therefore, the DHA incorporated into storage fat is supposed to be in an environment resistant to lipid peroxidation. This is very likely to be associated with one of the mechanisms by which dietary DHA fails to promote lipid peroxide formation in the liver to levels expected from the peroxidizability index of liver total lipids. In addition, it has been reported that the energy state is more stable when the stearic acid is combined at the sn-1 position and DHA at the sn-2 position than when AA or (n-6) eicosatrienoic acid is combined at the sn-2 position (Applegate and Glomset 1991 ). This may also be a mechanism of the suppression.

Feeding lipids different in fatty acid composition generally does not significantly alter the distribution or amount of phospholipids as a component of tissue cell membranes (Charnock et al. 1986 ). However, dietary DHA increased the liver phospholipid levels, particularly that of PE, in this study. Because the proportions of DHA in tissue phospholipids were higher in the order of liver > kidney > testis (Table 9) , the response of DHA to transacylation to the sn-2 position in the process of phospholipid biosynthesis differed from tissue to tissue. From the results of phospholipid levels (Table 7) and the fatty acid compositions (Table 9) in the liver, DHA was present in PE and PC in almost the same amount. But the PE level was lower than that of PC; thus DHA was utilized preferentially for PE synthesis. This utilization of DHA for PE synthesis occurred also in the testis. In addition, antioxidant synergism between VE and amino-containing phospholipids such as PE and PS has been reported in in vitro studies (Dziedzic and Hudson 1984 , Kashima et al. 1991 , Totani 1997 ), suggesting an action of amino-containing phospholipids to suppress the oxidative decomposition of VE and enhance the antioxidative activity of VE in vitro. If the antioxidant synergism exists between VE and these amino-containing phospholipids in biomembranes as well, incorporation into PE and PS of DHA might lead to the protection of DHA itself from peroxidation in cellular membranes of the liver, and even to an antioxidation in tissues to protect PUFA from peroxidation. However, as far as the peroxidizability index ratios of phospholipid species are concerned, no particular phospholipid species was related to the changes in liver lipid peroxide levels in this study (Fig. 1A ). This observation is different from that found in a previous study with a shorter feeding period (15 d) (Kubo et al. 1998 ) in which the lipid peroxide levels, as assessed by TBA values, seemed to be associated with changes in the peroxidizability index of PC (+ cardiolipin) in the tissues. Over the 32 d of this study, fatty acid metabolism may be in a state of dynamic equilibrium in tissues such as liver, kidney and testis; thus the difference among peroxidizability index ratios of each corresponding phospholipid species might appear to be very small.

In kidney, peroxidizability indices of each phospholipid were lower in the LA group than those of the liver and testis (Table 9) ; conversely, the TBA value was higher in the same group (Tables 4 and 5) . This seemed to have been due to the lower levels of -tocopherol, ascorbic acid and GSH in the kidney than in the liver and testis (Table 6) . However, when rats were fed the highest amount of VE, the lipid peroxide levels were not different from the control level (Table 5) , as was the serum TBA value. We suggested previously that the antioxidative action of VE in the kidney is insufficient over a feeding period of 15 d (Kubo et al. 1998 ). Thus, when the feeding period was extended, the antioxidative activity of VE became intensified, suggesting an adaptive response.

In the kidney of low VE-fed rats, the lipid peroxides, particularly the chemiluminescence intensity, were higher than that expected from the peroxidizability index of the total lipids (Fig. 1B ). At a nutritional status close to VE deficiency, it would be difficult to suppress the formation of peroxyradicals; thus, the chemiluminescence emitted in the initial stage of lipid peroxidation may be increased more easily than the TBA reactive substances formed in the later stage of lipid peroxidation. The composition of the kidney includes more (n-6) PUFA low in unsaturation compared with (n-3) PUFA high in unsaturation in the total lipids (Kubo et al. 1998 ), likely leading to a higher chemiluminescence emission in the initial stage. In the groups fed >54 mg/kg VE, changes in the lipid peroxides, particularly the TBA values, seemed to be associated with changes in the peroxidizability index of total lipids (Fig. 1B ). This appeared to be explained, at least in part, by the lower levels of lipid peroxide scavengers and higher (n-6) PUFA low in unsaturation. Accordingly, variations in the PUFA profiles of the kidney lipid species were investigated further, focusing in particular on the DHA profile.

The NPL level in the kidney was rather low and almost half that of the liver in the DHA groups (Table 7) . The composition of PUFA, including DHA, was also low in the DHA groups (Table 8) . Thus, even if DHA present in NPL plays a role in suppressing lipid peroxide formation in the kidney as in liver NPL, the contribution may be small compared with that in the liver. This may also be one of the reasons that the changes in kidney lipid peroxide levels, except chemiluminescence intensity in the low VE group, nearly coincided with the changes in peroxidizability index for kidney total lipids.

For the phospholipid level in kidney (Table 7) , less PC was detected but the DHA was distributed in almost equal amounts in each phospholipid, i.e., PE, PC and PS + PI, because the proportion of DHA in DHA groups was a little higher in PC (Table 9) . Thus, DHA was utilized preferentially for PC biosynthesis in the kidney. This phenomenon may also be related to the concomitant changes in peroxidizability index and lipid peroxide levels in DHA groups fed VE at more than the usual level (Fig. 1) . However, in the kidney as well, no specific phospholipid species related directly with the changes in lipid peroxide levels.

The lipid peroxide levels in testis did not increase, even when a high DHA diet containing the usual level of VE was fed to rats (Table 5) . Accordingly, the susceptibility to lipid peroxidation is lower in the testis than in the liver and kidney. Actually, the TBA value for the testis was extremely low compared with those for other tissues. This is supported by the higher -tocopherol level, even in the low 7.5 mg/kg VE group (Table 6) . That the testis is insensitive to lipid peroxidation was recognized even in our shorter feeding period study (Kubo et al. 1998 ). Moreover, the ascorbic acid level of the testis was retained at a higher level, ~2 times that of the kidney (Table 6) , suggesting effective -tocopherol recycling mediated by ascorbic acid. Therefore, the testis likely contains enough lipid peroxide scavengers to suppress lipid peroxide formation even after high DHA intake.

The testis was high in (n-6) PUFA, and thus the proportion of DHA was lower than that in the liver and kidney, resulting in smaller variations in the fatty acid profiles of the NPL and phospholipid species. This appears to lead to a relatively low VE requirement, and also explains why the testis is insensitive to lipid peroxidation. In the groups given dietary VE at more than the usual level, the lipid peroxide levels, as in the liver, were lower than those expected from the peroxidizability index of testis total lipids (Fig. 1C ). This result may be explained by the lipid peroxide scavengers and the less altered fatty acid profiles in the lipid species as just mentioned.

The results obtained herein suggest that the antioxidative effect of VE on lipid peroxidation differed from tissue to tissue, and the period of VE intake and the differences in PUFA incorporation, particularly DHA, into tissue lipids were closely associated with the antioxidative activity of VE in tissues, in addition to dietary VE levels and tissue ascorbic acid and GSH levels. Furthermore the amount of NPL and phospholipid classes, and the incorporation of PUFA, particularly of DHA, into those classes were also thought to be associated closely with the tissue lipid peroxide formation. Further investigations are required to elucidate whether specific phospholipid species relate to the changes in tissue lipid peroxide formation.

	ACKNOWLEDGMENTS

 
We thank Maruha Corporation of Japan for the generous donation of DHA.

	FOOTNOTES

 
¹ Supported by a Health Science Research Grant from the Ministry of Health and Welfare, by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture and by the Skylark Food Science Institute Academic Research Grant Program, Japan.

² Current address; Laboratory of Nutritional Biochemistry, Department of Applied Biology and Chemistry, Tokyo University of Agriculture, 1–1-1, Sakuragaoka, Setagaya-ku, Tokyo 156-0054, Japan.

⁴ Abbreviations used: AA, arachidonic acid; ALT, alanine aminotransferase; AST, aspartate aminotransferase; DBI, double-bond index; DHA, docosahexaenoic acid; DPA, docosapentaenoic acid; en%, energy %; EPA, eicosapentaenoic acid; FAME, fatty acid methyl esters; GLC, gas-liquid chromatography; GSH, glutathione; GSHPx, Selenium-dependent glutathione peroxidase; LA, linoleic acid; NPL, nonphosphorus lipids; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PS, phosphatidylserine; PUFA, polyunsaturated fatty acids; SH, sulfhydryl; TBA, thiobarbituric acid; VE, vitamin E.

Manuscript received September 1, 1999. Initial review completed October 20, 1999. Revision accepted March 6, 2000.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. American Institute of Nutrition Report of the American Institute of Nutrition ad hoc committee on standards for nutritional studies. J. Nutr. 1977;107:1340-1348

2. American Institute of Nutrition Second report of the ad hoc committee on standards for nutritional studies. J. Nutr. 1980;110:1726

3. Applegate K. R., Glomset J. A. Effect of acyl chain unsaturation on the conformation of model diacylglycerols: a computer modeling study. J. Lipid Res. 1991;32:1635-1644[Abstract]

4. Atwater W. O. Principles of nutrition and nutritive value of food. U. S. Dep. Agric. Farmers’ Bull. 1910;142:48

5. Beutler E., Duron O., Kelly B. M. Improved method for the determination of blood glutathione. J. Lab. Clin. Med. 1963;61:882-888[Medline]

6. Boveris A., Cadenas E., Chance B. Ultraweak chemiluminescence: a sensitive assay for oxidative radical reactions. Fed. Proc. 1981;40:195-198[Medline]

7. Charnock J. S., Abeywardena M. Y., McLennan P. L. Comparative changes in the fatty-acid composition of rat cardiac phospholipids after long-term feeding of sunflower seed oil- or tuna fish oil-supplemented diets. Ann. Nutr. Metab. 1986;30:393-406[Medline]

8. Cook R. P. Chemistry, biochemistry, and pathology. Cook R. P. eds. Cholesterol 1st ed. 1958:145-180 Academic Press New York, NY.

9. Cosgrove J. P., Church D. F., Pryor W. A. The kinetics of the autoxidation of polyunsaturated fatty acids. Lipids 1987;22:299-304[Medline]

10. Duncan D. B. Multiple range tests for correlated and heteroscedastic means. Biometrics 1957;13:164-176

11. Dziedzic S. Z., Hudson B.J.F. Phosphatidyl ethanolamine as a synergist for primary antioxidants in edible oils. J. Am. Oil Chem. Soc. 1984;61:1042-1045

12. Farwer S. R., Boer B.C.J.D., Haddeman E., Kivits G.A.A., Wiersma A., Danse B.H.J.C. The vitamin E nutritional status of rats fed on diets high in fish oil, linseed oil or sunflower seed oil. Br. J. Nutr. 1994;72:127-145[Medline]

13. Fletcher B. L., Dillard C. J., Tappel A. L. Measurement of fluorescent lipid peroxidation products in biological systems and tissues. Anal. Biochem. 1973;52:1-9[Medline]

14. Folch J., Lees M., Sloane-Stanley G. H. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 1957;226:497-507[Free Full Text]

15. Hammer C. T., Wills E. D. The role of lipid components of the diet in the regulation of the fatty acid composition of the rat liver endoplasmic reticulum and lipid peroxidation. Biochem. J. 1978;174:585-593[Medline]

16. Holub B. J., Kuksis A. Metabolism of molecular species of diacylglycerophospholipids. Adv. Lipid Res. 1978;16:1-125[Medline]

17. Hu M.-L., Frankel E. N., Leibovitz B. E., Tappel A. L. Effect of dietary lipids and vitamin E on in vitro lipid peroxidation in rat liver and kidney homogenates. J. Nutr. 1989;119:1574-1582

18. Juaneda P., Rocquelin G. Rapid and convenient separation of phospholipids and non phosphorus lipids from rats heart using silica cartridges. Lipids 1985;20:40-41[Medline]

19. Kaasgaard S. G., Hølmer G., Høy C.-E., Behrens W. A., Beare-Rogers J. L. Effects of dietary linseed oil and marine oil on lipid peroxidation in monkey liver in vivo and in vitro. Lipids 1992;27:740-745[Medline]

20. Kashima M., Cha G.-S., Isoda Y., Hirano J., Miyazawa T. The antioxidant effects of phospholipids on perilla oil. J. Am. Oil Chem. Soc. 1991;68:119-122

21. Kobatake Y., Hirahara F., Innami S., Nishide E. Dietary effect of -3 type polyunsaturated fatty acids on serum and liver lipid levels in rats. J. Nutr. Sci. Vitaminol. 1983;29:11-21

22. Kubo K., Saito M., Tadokoro T., Maekawa A. Changes in susceptibility of tissues to lipid peroxidation after ingestion of various levels of docosahexaenoic acid and vitamin E. Br. J. Nutr. 1997;78:655-669[Medline]

23. Kubo K., Saito M., Tadokoro T., Maekawa A. Dietary docosahexaenoic acid does not promote lipid peroxidation in rat tissue to the extent expected from peroxidizability index of the lipid. Biosci. Biotechnol. Biochem. 1998;62:1698-1706[Medline]

24. Lowry O. H., Rosebrough N. J., Farr A. L., Randall R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 1951;193:265-275[Free Full Text]

25. Meister A. On the antioxidant effects of ascorbic acid and glutathione. Biochem. Pharmacol. 1992;44:1905-1915[Medline]

26. Mino M., Tamai H., Yasuda C., Igarashi O., Hayashi M., Hirahara F., Katsui G., Kijima S. Biopotencies of tocopherol analogues as determined by dialuric acid-induced hemolysis in rats. Vitamins (Japan) 1988;62:241-246

27. Miyashita K., Nara E., Ota T. Oxidative stability of polyunsaturated fatty acids in an aqueous solution. Biosci. Biotechnol. Biochem. 1993;57:1638-1640

28. Miyazawa T., Kaneda T., Takyu C., Yamaguchi A., Inaba H. Generation of singlet molecular oxygen in rat liver homogenate on adding autoxidized linseed oil. Agric. Biol. Chem. 1981;45:1597-1601

29. Miyazawa T., Tsuchida K., Kaneda T. Riboflavin tetrabutyrate: an antioxidative synergist of alfa-tocopherol as estimated by hepatic chemiluminescence. Nutr. Rep. Int. 1984;29:157-165

30. Mouri K., Ikesu H., Esaka T., Igarashi O. The influence of marine oil intake upon levels of lipids, -tocopherol and lipid peroxidation in serum and liver of rats. J. Nutr. Sci. Vitaminol. 1984;30:307-318

31. Noguchi T., Cantor A. H., Scott M. T. Mode of action of selenium and vitamin E in prevention of exudative diathesis in chicks. J. Nutr. 1973;103:1502-1511

32. Ohkawa H., Ohishi N., Yagi K. Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal. Biochem. 1979;95:351-358[Medline]

33. Pietrangelo A., Grandi R., Tripodi A., Tomasi A., Ceccarelli D., Ventura E., Masini A. Lipid composition and fluidity of liver mitochondria, microsomes and plasma membrane of rats with chronic dietary iron overload. Biochem. Pharmacol. 1990;39:123-128[Medline]

34. Rao K. S., Recknagel R. O. Early onset of lipoperoxidation in rat liver after carbon tetrachloride administration. Exp. Mol. Pathol. 1968;9:271-278[Medline]

35. Reitman S., Frankel S. A colorimetric method for the determination of serum glutamic oxalacetic and glutamic pyruvic transaminase. Am. J. Clin. Pathol. 1957;28:56-63[Medline]

36. Roe J. H., Mills M. B., Oesterling M. J., Damron C. M. The determination of diketo-1-gulonic acid, dehydro-1-ascorbic acid, and 1-ascorbic acid in the same tissue extract by the 2,4-dinitrophenyl-hydrazine method. J. Biol. Chem. 1948;174:201-208[Free Full Text]

37. Saito M. Polychlorinated biphenyls-induced lipid peroxidation as measured by thiobarbituric acid-reactive substances in liver subcellular fractions of rats. Biochim. Biophys. Acta 1990;1046:301-308[Medline]

38. Saito M., Kubo K., Ikegami S. An assessment of docosahexaenoic acid (DHA) intake with special reference to lipid metabolism in rats. J. Nutr. Sci. Vitaminol. 1996;42:195-207

39. Saito M., Nakatsugawa K., Oh-hashi A., Nishimuta M., Kodama N. Comparison of vitamin E levels in human plasma, red blood cells, and platelets following varying intakes of vitamin E. J. Clin. Biochem. Nutr. 1992;12:59-68

40. Saito M., Oh-hashi A., Kubota M., Nishide E., Yamaguchi M. Mixed function oxidases in response to different types of dietary lipids in rats. Br. J. Nutr. 1990;63:249-257[Medline]

41. Saito M., Yamaguchi M. Influence of excessive ascorbic acid dose on liver microsomal mixed function oxidase system in guinea pigs. J. Clin. Biochem. Nutr. 1988;4:123-137

42. Stewart J.C.M. Colorimetric determination of phospholipids with ammonium ferrothiocyanate. Anal. Biochem. 1980;104:10-14[Medline]

43. Tappel A. L. Vitamin E as the biological lipid antioxidant. Vitam. Horm. 1962;20:493-510

44. Totani Y. Antioxidative effect of nitrogen-containing phospholipids on autoxidation of polyunsaturated oil. J. Jpn. Oil Chem. Soc. (Yukagaku) 1997;46:3-15

45. Tsuchida M., Miura T., Mizutani K., Aibara K. Fluorescent substances in mouse and human sera as a parameter of in vivo lipid peroxidation. Biochim. Biophys. Acta 1985;834:196-204[Medline]

46. Wells W. W., Xu D. P., Washburn M. P. Glutathione: dehydroascorbate oxidoreductases. Methods Enzymol 1995;252:30-38[Medline]

47. Yagi K. A simple fluorometric assay for lipoperoxide in blood plasma. Biochem. Med. 1976;15:212-216[Medline]

48. Yazu K., Yamamoto Y., Ukegawa K., Niki E. Mechanism of lower oxidizability of eicosapentaenoate than linoleate in aqueous micelles. Lipids 1996;31:337-340[Medline]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Kubo, K. Articles by Maekawa, A. Search for Related Content PubMed PubMed Citation Articles by Kubo, K. Articles by Maekawa, A.