The composition of alkanes in exhaled air of rats as a result of lipid peroxidation in vivo Effects of dietary fatty acids, vitamin E and selenium

The composition of alkanes in exhaled air of rats as a result of lipid peroxidation in vivo Effects of dietary fatty acids, vitamin E and selenium

Biochimica et Biophysics Acta, 665 (1981) 559-570 Elsevier/North-Holland 559 Biomedical Press BBA 57892 THE COMPOSITION OF ALKANES IN EXHALED AIR...

1MB Sizes 0 Downloads 6 Views

Biochimica et Biophysics Acta, 665 (1981) 559-570

Elsevier/North-Holland

559

Biomedical Press

BBA 57892

THE COMPOSITION OF ALKANES IN EXHALED AIR OF RATS AS A RESULT OF LIPID PEROXIDATION IN VIVO EFFECTS

OF DIETARY FATTY ACIDS. VITAMIN E AND SELENIUM

G.A.A. KIVITS, M.A.C.R. GANGULI-SWARTTOUW and E.J. CHRIST Unilever Research Laboratorium,

3130 AC, Vlaardingen (The Netherlands)

(Received February 13th, 1981) (Revised manuscript received May 26th, 1981) Key words: Lipid peroxidation;

Exhaled air; Alkane; Fatty acid composition;

Dietary effect; (Rat liver)

Alkane production in exhaled air of rats has been studied as an index of lipid peroxidation in vivo in these animals. The effect of feeding essential fatty aciddeficient rats varying levels of n-4, n-6 and n-7 polyunsaturated fatty acids for various periods of time has been studied with regard to the composition of the alkanes produced as well as the fatty acid composition of liver phospholipids and liver and adipose tissue triacylglycerols. It was found that the fatty acid composition of liver lipids depended markedly on the nature and the quantity of polyunsaturated fatty acid in the diet. The composition of the alkanes produced on stimulation of lipid peroxidation in vivo by inhalation of small, non-lethal doses of carbon tetrachloride corresponded closely to the fatty acid composition of the liver phospholipids. The results strongly suggest that the alkanes produced as a result of lipid peroxidation in vivo originate from the methyl end of the fatty acid administered. So ethane is produced from n-3 acid, propane from n-4 acid, pentane from n-6 acid and hexane from n-7 acid. The amounts of a specific alkane produced increase as its corresponding fatty acid, as present in the liver phospholipids, increases. There are indications that relatively more ethane than pentane is produced on stimulation of the in vivo lipid peroxidation although there are considerably more n-6 fatty acids than n-3 fatty acids present in the liver phospholipids.

Introduction Lipid peroxidation in vivo has been implicated as a basic deteriorative reaction that leads to liver injury [ 1,2] and cellular ageing [3]. In vitro peroxidation of membrane lipids is accompanied by extensive damage to structure and function of the membranes [4,5]. Until recently there was considerable controversy about whether lipid peroxidation had any significance in intact cells in vivo. Experiments to find evidence for the presence of lipid peroxides or breakdown products in cells in vivo have failed repeatedly [6]. There is now considerable evidence, however, that lipid peroxidation in vivo is a process which goes on during the entire life of an animal. Pigments such as lipofuscin, which accumulate in OOOS-2760/0000-0000/$02.50

0 1981 ElsevierlNorth-Holland

cells with age, may form by peroxidation of lipids in membranes [7]. Many methods have been used to measure lipid peroxidation [8]. Most of these methods are discontinuous, time-consuming and only applicable to post-mortem tissue. The measurement of gaseous alkanes in expired air seems a rapid, sensitive and nondestructive method for determining the first reaction step in lipid peroxidation. It has been established that in vitamin E and selenium deficiency, significantly increased amounts of alkanes, mainly ethane and pentane, are expired by animals [9-121. However, not only ethane and pentane have been described as products of lipid peroxidation. Ethane and ethylene are formed by plants due to aerobic lipid degradation [ 131. Ethylene formation coupled Biomedical Press

560

with peroxidation of lipids was also shown in rat liver microsomes [ 141, Formation of ethylene in rat liver extracts, presumably by conversion of L-methionine and independent of lipid peroxidation processes, was reported recently [ 151. The hydrocarbon gases identified as products of in vivo lipid pero~dation include methane, ethane, ethylene, propane, butane, pentane and hexane [11,16,17]. Although it is assumed generally [lo-121 that ethane and pentane are derived from the terminal ends of the n-3, and the n-6 fatty acid series, respectively, direct evidence has not been provided in in vivo systems. Since also other hydrocarbons have been found as products from in vivo lipid peroxidation or from other processes, it is of interest to demonstrate directly the relationship of the fatty acid composition of cellular lipids and the composition of the alkanes produced. To this end we have studied lipid peroxidation in essential fatty acid~e~cient rats which received various amounts of n-4, n-6 and n-7 fatty acids. Effects of vitamin E and selenium deficiency were studied since these deficiencies are known to stimulate the in vivo lipid peroxidation. For convenience, in other experiments carbon tetrachloride was used as a stimulant [ I] . Methods and Materials Animals and diets Vitamirt E’- and selenium-deficient diets. Male,

Specific Pathogen-Free Wistar rats (3-weeks-old), CPB-WU strain, were obtained from the Central Institute for the Breeding of Laboratory Animals, Zeist, The Netherlands. The rats were housed individually in stainless steel, wire-bottomed cages. The temperature was 24 ? 1°C and the relative humidity 55 + 5%. A day/night cycle of 12 h was installed. The rats were placed for 3 weeks on a lard diet to which 3 I.U. vitamin E/Meal was added, but no selenium, and subsequently on the experimental diets. Diets and drinking water were ad lib. For composition, see Tables I-III. Only one parameter was studied: (i) Vitamin E and selenium content. During the test period, all animals received a soy protein/ sunflower oil diet deficient in vitamin E and selenium, which for control animals was made adequate by addition of these factors. After the animals had been on the lard diet for

3 weeks, they were divided into dietary treatment groups according to body weight (mean body weight: 149-151 g). These conditions led to groups of animals of at least 8 animals each. Before measurements, rats were fasted for 20 h (overnight). Essential fatty acid-de~cie~ t artd F~lyuFlsa~ratgd fatty acids-enriched diets. Male Wistar rats, from

already partly essential fatty acid-deficient mothers, were fed an essential fatty acid-free diet for about 12 weeks until constant body weights were reached. All animals showed essential fatty acid-deficiency signs (e.g. scafy skin, see also Table V for the presence of 20 : 3 (n-9), ‘Mead acid’, as indicative of essential fatty acid deficiency). The animals were then divided into groups of 4-8 animals which received the following treatments: (1) 700 mg oleic acid methyl ester per stomach tube; (2) 700 mg sunflower oil per stomach tube; (3) 3 energy % sunflower oil/day for 7 days in diet; (4) 700 mg (Z,Z,Z)-8,11,14-heneicosatrienoic acid methyl ester (21 : 3 n-7) per stomach tube; (5) 3 energy % 21 : 3 n-7 methyl ester/day for 7 days in diet; (6) 700 mg (Z,Z,Z,)-8,11,14-octadecatrienoic acid methyl ester (18 : 3 n-4) per stomach tube. The alkane composition of exhaled air of these animals was measured 24 h after feeding the diets. Lipid peroxidation was stimulated by injection of a non-lethal dose of carbon tetrachloride into the animal holding chamber (initial concentration 0.23 pmol/ml). After decapitation of the animals, liver and fat pads were quickly removed for fatty acid analysis of phospholipids and triacylglycerols. Fatty acid analysis

Liver and adipose tissue lipids were isolated by conventional techniques [ 181. Lipid classes were separated by thin-layer chromatography with the solvent system heptane/diethyl ether/acetic acid (80 : 20 : 1, v/v~v). Methyl esters of fatty acids were prepared by transesteri~~tion with methanolic hydrochloric acid (1.35 mol/l). Gas liquid chromatography with a Hewlett Packard 5 711 B gas chromatograph with flame ionisation detection was carried out on 5% DEGS (diethyleneglycol succinate) on HPCHROM, WAW, DMCS 80 mesh, glass columns (2 m; diameter, 2.0 mm) at a nitrogen flow rate of 10 mllmin. The detector temperature was 250°C and the injector temperature 200°C. The column temperature

was programmed as follows: 4 min isotherm at 126”C, followed by a rise of 4”Cjmin to a temperature of 19O”C, which was maintained for 30 min. The cornposition was calculated from the proportionalities of peak areas by an automated digital integrator; peaks were identified by comparison of retention times with standards. Description

ofanimal

chamber

The animal holding chamber consisted of a 2.5 1 vacuum desiccator equipped with a magnetically stirred fan and a septum-sealed opening for gas sampling (Fig. 1). The animals were placed in an air environment on a metal wire gauze beneath which was a layer of sodalime granules to remove respiratory CO*. Absorbed COz was replaced with hydrocarbonscrubbed oxygen from an outside supply. Prior to measurements of hydrocarbon release the animaI holding chamber was swept with hydrocarbon free air for 2 h. Hy~o~arbon production was calculated in pmol per 100 g rat weight per hour. From the net chamber volume without animal (2 680 ml) the rat’s weight in g was subtracted to yield a close approximation of the hydrocarbon distribution volume.

A Hewlett Packard 5 700 A gas ~~omato~aph equipped with a Valco 6-port gas sample device and a dual flame ionisation detector was used. The gas sample loop was connected to the animal holding chamber with Teflon rubing via a proportioning pump.

molecular sieve “42

---cm

M

absorbent

Fig. 1. Schematic diagram of rat chamber.

Nitrogen was used as the carrier gas at a precoiumn pressure of 270 kPa and was passed successively through an oxygen trap, a hydrocarbon filter (charcoal) and a molecular sieve (13X) before entering the column. Hydrocarbons were removed from air by molecular sieves, SA and 13X, the latter immersed in a -78°C bath. ~onven~onal glass columns (length, 1.80 m; inner diameter, 2.0 mm) were packed with either Porapak P or Carbosieve B for determination of ethane and propane. A column of Porasil D coated with 6% of hexadecane was used for analysis of pentane and hexane. Flow rate of both columns kept at 30 m&in, Column temperature was room temperature (22~24°C); injector temperature was 100°C and detector temperature 250°C. For analysis of pentane and hexane an airtight syringe provided with a pressure lock (sample volume 2.0 ml) was used. For analysis of ethane and propane the gas sampling device was used as it permits introduction of larger volumes. Calibration was done with diluted gases in pressurized cylinders. The calibration curves for both ethane and pentane were linear in the range to be measured. Since the peaks were very narrow, peak heights rather than areas were used for calculations. The standard deviation was between 15 and 25%. Gktathione

peroxidase

activity

The glutathione peroxidase activity was determined in the transformed hemolysate of rat’s tail blood (0.35 ml) and in the supernatant of rat liver, following the principles of Gfinzler et al. [19]. The final reaction mixture contained: f unit/ml ~utat~one reductase, 0.05 M potassium phosphate, pH 7.0,0.5 mM EDTA, 1 mM GSH, 2 mM NaNa and enzyme. Total volume: 3 ml. The enzyme was preincubated at 25°C. The peroxide-independent NADPH consump tion was measured by the addition of 0.25 mM NADPH and 0.01% NaHCOa (final concentration). The ~ro~dedependent NADPH ~on~mpt~on was then measured by the addition of J .2 mM (CH&COOH or OS mM Hz02. The reaction mixture was run at 25°C in 3ml cuvettes and followed by the continuous change in absorbance at 340 nm. For preparation of rat liver supernatant, liver was perfused with 09% NaCl at a flow rate of 30 ml/r&n to remove erythrocytes. 2 g of tissue was homogenized in 20 ml of 0.25 M sucrose with a Potter Elvehjem homo-

562

genizer with Teflon pestle. The homogenate was centrifuged for 5 min at 900 X g in a Sorvall refrigerated centrifuge (Rotor SS34). The pellet was rehomogenized in 10 ml of 0.25 M sucrose and centrifuged again for 5 min at 900 Xg. The combined supernatants were centrifuged for 10 min at 27 000 Xg and the resulting supernatant was again centrifuged for 60 min at 100 000 X g in a Beckmann ultracentrifuge (Rotor 50 Ti). Glutathione peroxidase activity was determined in the final supernatant. Protein content was measured according to Jacobs et al. [20]. The concentration of hemoglobin was determined with a Biochemica Test combination of Boehringer (Mannheim).

Hydrocarbon

(nmoles)

60 I 0

5.0 -

0 0

3.0 &‘O_

0

0

lo-

//

0 //

0

Results Fig. 2 shows the amounts of hydrocarbons released in iron-catalyzed peroxidation of fresh, non-activated fatty acids, ethane is evolved from n-3 acid, pentane from n-6 acid and hexane from n-7 acid. In contrast to data from the literature [23,24] which state that substantial peroxidation could only be achieved with activated (either by light exposure during several days or lipoxidase-induced) fatty acids, peroxidation was obtained already with pure fatty acid (over 97%) preparations. Small amounts of mainly propane and butane were observed (l-4% based on the major hydrocarbon gas released). These hydrocarbons could have been formed from @cissionderived products of intermediate radicals [25]. Thus, /3-scission of the pentyl radical would give ethylene and propyl radical, which, by hydrogen abstraction, yields butane, propane and ethane. Indeed, small amounts of ethane were observed in the chromatograms on peroxidation of linoleic acid. This theory could account for the detection of minor amounts of hydrocarbons other than ethane and pentane shown by some investigators [26,27] and also by us in a few cases. In healthy animals, only ethane and pentane are released, as is shown in Fig. 3. It is clear that evolution of ethane, and even of pen-

0 0

.~.-*[email protected] 8’

I 10

Chemicals All chemicals used were analytical grade. The n-4 and n-7 polyunsaturated fatty acid methyl esters (purity over 95% by GLC) were synthesized as described earlier [21,22].

/

0

20

/

cl/”

20

I 30

40

I 50

60

time(mln)

2. Hydrocarbons evolved during peroxidation of fatty acids in a non-enzymatic model system. 0, Ethane from Z9, Z12,ZlSoctadecatrienoic acid; 0, pentane from Z9,212octadecadienoic acid; q, hexane from Z8,Zll,Z14-heneicosatrienoic acid. lOO-ml serum bottles provided with rubber caps were used as reaction vessels. The 10.0 ml reaction mixtures contained: 0.0002 M fatty acid (as ammonium salt); 0.125 M Tris/citrate buffer, pH 5.0; 0.02 M thioglycolit acid; 0.0001 M Fe3+ (either as FeC13 or as EDTA complex). Incubation was at room temperature (22-24’C) in purified air. For headspace analysis, air samples were drawn through the rubber caps with a gastight syringe and injected into the gaschromatograph. The experiment shown is representative of three similar experiments. Fig.

tane, is linear up to a period of 60 min, after which production of pentane slowly levels off, presumably by uptake of the gas into the animal’s fat depots. Evolution rates of hydrocarbons as given here, therefore, always represent initial rates. Even on stimulation of in vivo lipid peroxidation by injection of small amounts of carbon tetrachloride into the animal holding chamber, no hydrocarbons other than ethane and pentane have been observed, except small amounts of ethylene, which evolved at the higher concentration of carbon tetrachloride (less than 30% of the ethane evolution in control (non-stimulated) animals). It should be noted that the animals survive these levels of carbon tetrachoride, even when applied several times at l-week intervals. The relation between the concentration of carbon tetrachloride administered and the amounts of hydrocarbons released suggests that ethane is more readily released

563 CsHtz/lGOg body wt

nmd hydrocarbon/h

nmdes

0.e -

0.6 -

0.4 -

/O 0-O

0

20

1

40

60

1

80

,

100 [CCl4]l(nmolelml)

A

nmoles CzHs/lOOg body wt 10

Fig. 4. Relationship between the amount of injected carbon tetrachloride in the animal holding chamber and the subsequent release of hydrocarbons. Rats (four animals) on sunflower oil diets (see Table I) with adequate vitamin E and selenium were placed into the animal holding chamber and carbon tetrachloride was injected. The evolution of ethane (o) and pentane (a) by the animals was measured. Figures represented are means of independent determinations. Standard deviation was 15-25%.

6 A/

06 /

/

06

0.4

02

, 0

20

40

60

,

, 80

(

, 100

time(min) Fig. 3. Time course for total accumulated amount of pentane (A) and ethane (B) by rats after injection of different amounts of carbon tetrachloride in the animal holding chamber. Carbon tetrachloride: A, 50 ~1; o, 30 ~1; o, control. Rats (four animals) on sunflower oil diets (see Table I) with adequate vitamin E and selenium were placed into the animal holding chamber and carbon tetrachloride was injected in amounts as indicated. Evolution of alkanes was measured. Figures are means of independent determinations. Standard deviation was 15 -25%.

than pentane (Fig. 4). It was, therefore, of interest to study the possible relationship between the composition of the alkanes produced as a result of in vivo lipid peroxidation and the fatty acid composition of tissue lipids, e.g. of liver. In Table IV results of an experiment with essential fatty acid-deficient rats receiving different amounts of various polyunsaturated fatty acids are given. In the same way as pentane is formed from n-6 acids, propane is formed from n4 acid (18 : 3) and hexane from n-7 acid (21 : 3). This indicates that

the alkanes measured originate from the terminal end, the methyl end of the fatty acids administered. Also, a direct relationship is shown between the amounts of a specific alkane produced and its corresponding fatty acid as present in the liver phospholipids. The fatty acid compositions of liver and fat pads of these essential fatty acid-deficient rats after various dietary treatments are given in Table V. These figures show that in essential fatty acid-deficiency, only small amounts of the n-6 fatty acids remain in the phospholipids; in the triacylglycerols these acids are hardly present. Yet a measurable amount of pentane is released when the in vivo lipid peroxidation is stimulated. Release of pentane, however, is much higher when the animal on a linoleic acid-free diet receives a daily dose of 700 mg linoleic acid for 1 week. It is also shown that the 21 : 3 (n-7) acid is incorporated mainly into liver phospholipids, contrary to the 18 : 2 (n-6) acid (linoleic acid), which is also incorporated into liver triacylglycerols. The fact that roughly similar quantities of pentane and hexane are formed, may indicate that, at least under these experimental conditions, mainly membrane phospholipids are peroxidized. The relationship between the composition of alkanes produced as a result of in vivo lipid peroxidation, and the fatty acid composition of liver phos-

564 TABLE I COMPOSITION OF DIETS AND CONTRIBUTION OF DIETARY COMPONENTS TO ASSIMILABLE ENERGY P: protein;C; --

carbohydrate;

F: fat. Total diet (energy 6): P, 23; C, 42; F, 35. -Ingredients

Diet Sunflower oil or lard

g/kg

Soycomil Cornstarch Sunflower oil or lard a Cellulose Salt mix b

348.3 432.9 162.1 39.6 NaCl CaHP04 .2 H20 CaCOa CeHsNaa07 .2 Hz0 Trace minerals

J/kJ

20.0 24.8 9.3 2.3

P: 23O;C: 67 c: 353 F: 350

2.46 0.10 0.30 0.14

Vitamins

4.2

Total Essential fatty aciddeficient

g/MJ

0.24

1000

Casein (23 energy c/c) Saccharose (72 energy %) Cellulose Salt mix Vitamins

1000

57.43

234.5 664.3 56.7 20.3 3.8

a Sunflower oil, tocophecoi stripped to less than 5 mg/kg; lard, tocopherol diet where appropriate. b Where appropriate, Se was added as NazSeOs.

stripped

to 6 mg DL-wtocophero1

acetate

per MJ of

TABLE II MINERALS (mg/MJ) AVAILABLE IN DIET

From Soycomil Added Total

Ii

Na

Ca

Mg

381

2 41

50 146

71

43

196

71

381 --_

TABLE III SELENIUM AND VITAMIN E CONTENT OF EXPERIMENTAL DIETS Data on selenium content from atomic absorption analysis. Diet

Se (mglkg)

or-Tocopherol acetate (mglkg)

Deficient Supplemented

0.03 0.15

0 75

Soycomil

0.10

0

Cl

P

Fc

cu -

25

40 106

2 2

0.3 0.4

25

146

4

0.7 -

pholipids, is illustrated again in Table VI. In these experiments the ratio of pentane : ethane in the expired breath of rats was determined as well as the ratio of n-6 : n-3 fatty acids present in their liver phospholipids. The healthy animal produces more ethane than pentane, although there are considerably more n-6 than n-3 fatty acids present in the liver phospholipids. The same relative trend is followed in animals receiving various amounts of sunflower oil as a source of linoleic acid. Since the way in which lipid peroxidation is

565 TABLE IV CARBON TETRACHLORIDE (INITIAL CONCENTRATION 0.23 pmol/ml)-INDUCED RELEASE OF HYDROCARBONS IN EXPIRED BREATH OF ESSENTIAL FATTY ACID-DEFICIENT RATS FED POLYUNSATURATED FATTY ACIDS Essential fatty aciddeficient rats were fed diets containing either oleic acid methyl ester, sunflower oil, or 28, Zll, Z14-heneicosatrienoic acid methyl ester. After the rats had been on the ,diets for the periods indicated, they were fasted overnight and used for alkane measurements as described in Methods and Materials. Figures are means of independent determinations. Standard deviation was between 15 -25% of the mean values shown. After alkane measurements, animals were killed by decapitation and the livers were quickly removed for fatty acid analysis of phospholipids. Number of animals

Fatty acid administered

Amount

Number of carbon atoms of alkanes produced (pmol/h per 100 g) c2

8 4 4 1

18 : 1 (n-9) Sunflower oil Sunflower oil 18 : 3 (n4)

4

21 : 3 (n-7)

6 --

21 : 3 (n-7)

a Incorporation

of 18 : 3 (nA);n.d.,

700 mg single dose 700 mg single dose 3 Cal%, 1 week 700 mg single dose 700 mg single dose 3 Cal%, 1 week

189 170 146 468 148 306

C5

c3

c6

153 807 971 99 160 348

140 -

Total amount of polyunsaturated fatty acids in liver phospholipids (%) a

_ 580 2420

n-3

n4

n-l

1.6 1.4 1.1 n.d 1.3 1.1

1.2 20.5 21.4 n.d 4.3 3.9

-

n.d 13.0 14.3

not determined, see Ref. 28. Cal%, calorie 70.

stimulated may also influence the ratio of C5 : C2 alkanes produced by lipid peroxidation in viva, lipid peroxidation in rats was stimulated by a dietary deficiency of vitamin E and selenium. All animals on the deficient diet were clearly deficient after 13 weeks, when the data on alkane production were collected. Deficiency symptoms as measured after 13 weeks are listed in Table VII. Since selenium deficiency is shown most clearly in the relative activities of glutathione peroxidase in erythrocytes and liver [29,30], data are included for relative activities of glutathione peroxidase in these tissues. Liver glutathione peroxidase was also measured with H202 as substrate since liver also contains the nonselenium-dependent glutathione peroxidase [3 I]. Glutathione peroxidase activity in both tissues fell below 10% levels in animals on the deficient diet. The greatly enlarged spleen and the increase in neutrophilic granulocytes are indicative of a vitamin E deficiency. In agreement with literature data [lo] the ratio of pentane : ethane produced was slightly increased, suggesting that vitamin E (and perhaps also selenium) protects n-6 acids slightly better than n-3 acids (Table VIII). Thus, the composition of alkanes in expired breath of rats, as a result of lipid peroxidation in vivo,

is affected by the fatty acid composition of tissue lipids, on the one hand, and by the way lipid peroxidation is stimulated (dietary deficiencies, drugs), on the other. Discussion

The major hydrocarbon gases that are evolved from peroxidation of polyunsaturated fatty acids with variously positioned terminal double bonds are derived from the terminal or methyl end of the carbon chain, according to Scheme I [23,24]. In our

R1hR2-R,CH; .&R2 . H’ Rr CHs --, R1CH3 fatty acids

Rl CH3

ethane

(n-3)

CH3CHa

propane

(n-4)

CH3 KHz)3

CH3KHz 14

-

pentane

(n-6)

hexane

(n-7)

Scheme I. Reaction from carbon chain.

scheme of derivation

of hydrocarbons

566 TABLE V FATTY ACID COMPOSITION (%) OF LIVER AND FAT PAD OF EFFICIENT FATTY ACID-DEFICIENT RATS FED DIETS Essential fatty acid-deficient rats were fed diets containing either oleic acid methyl ester, sunflower oil, or 28, Zll, Z14,-heneicosatrienoic acid methyl ester. After the rats had been on the diets for the periods indicated, they were killed by decapitation and liver and fat pads were quickly removed for fatty acid analysis of phospholipids and triacylglycerols. See also Methods and Diet

Liver phospholipids

Liver triacylglycerols

Fatty tissue

Fatty acid or oil

Amount

Oleic

700 mg

Sunflower

16 : 0

15.0

8.0

16.7

700 mg 3 energy 70

_

18.5 16.4

5.4 6.7

17.0 17.4

21 : 3 (n-7)

700 mg 3 energy %,

_ _

20.5 18.6

6.5 6.2

16.3 17.6

Oleic

700 mg

1.8

29.0

13.9

Sunflower

700 mg 3 energy %

1.8 1.6

33.9 34.2

8.8 11.8

2.3 2.2

21 : 3 (n-7)

700 mg 3 energy %

1.9 1.3

24.6 31.7

16.5 11.1

1.6 2.0

Oleic

700 mg

1.9

25.9

16.8

1.7

Sunflower

700 mg 3 energy %

1.9 1.9

24.6 25.5

16.5 17.0

1.6 1.6

700 mg 3 energy %

1.7 2.0

24.9 25.3

15.8 18.2

1.6 1.6

21 : 3 (n-7)

TABLE VI RELATION BETWEEN RATIO OF PENTANE/ETHANE PRODUCED BY IN VIVO LIPID PEROXIDATION (STIMULATED BY INJECTION OF 50 jd CARBON TETRACHLORIDE IN ANIMAL HOLDING CHAMBER) AND RATIO OF n-6/n-3 FATTY ACIDS OF LIVER PHOSPHOLIPIDS OF ESSENTIAL FATTY ACID-DEFICIENT RATS RECEIVING VARIOUS DIETARY TREATMENTS (MEAN FIGURES) For experimental V. ~~ Ratio of alkanes produced CsHrz : C2H6

conditions and diets see legend of Table

Ratio of n-6 : n-3 fatty acids of liver phospholipids

Treatment

0.7

3.7

none (controls)

0.8 * 0.12

4.0

essential fatty aciddeficient

3.5 f 0.70

17.0

+sunflower oil (single dose)

9.0 + 1.35

25.0

+ sunflower oil (1 week)

-~

16 : 1 (n-7)

14 :o

18 :0

1.9

in vitro experiments, small amounts of mainly ethylene, propane and butane were observed (l-4% based on major hydrocarbon evolved). /3-Scission of the radicals formed has been proposed [25] as the predominant route for formation of these minor products. We have not studied the conditions for formation of these minor products in detail, but others [2324] have shown that replacement of iron with Cu2+ greatly stimulates formation of ethylene from an n-3 acid in in vitro systems. In view of the complexities of even this relatively simple in vitro system, we decided that lipid peroxidation in the living animal should be studied first with respect to major products formed and to the effects which the composition of cellular lipids can exert on them. The presence of hydrocarbons in faecal gases forms a possible source of errors in the determinations of alkanes in exhaled air from animals. This problem was briefly mentioned in the literature [lo], but no data on the possible extent of this contamination were

567

OF DIFFERENT FATTY ACIDS Materials. Values are the averages of at least four determinations. tion was less than 10% of the mean values shown.

Fatty acids present in traces (
Fatty acid (%) 18 : 1 (n-9)

18 : 2 (n-6)

29.4

1.7

24.7 24.1

6.1 5.0

26.0 25.1

1.0 1.2

19 : 3 (n-7)

20 : 3 (nd)

20 : 3 (n-9)

20 :4 (n-6)

15.0

5.5

6,6 3.2

12.2 16.7

7.0 8.9

3.3 2.7

2.2 1.8

44.5

21 :3 (n-7)

21 :4 (n-7)

22 : 5 (n-61

22 : 6 (n-3 1.6

3.9 4.1 2.9

1.4 1.1 1.3 1.1

8.9 11.4

1.6

42.9 41.2

3.9 2.6

51.1 47.9

1.0 1.2

50.2

3.0

51.1 48.7

1.0 1.9

52.3 49.2

0.9 0.8

1.8

1.2 1.2

1.5

1.1

TABLE VII EFFECTS OF VITAMIN E- AND Se-DEFICIENT DIET ON CLINICAL DATA AND GLUTATHIONE PEROXIDASE ACTIVITY Animals on vitamin E-and Sedeficient diets were observed regularly. Data presented were obtained after the animals had been on diets for 13 weeks. Values are mean t S.E. Clinical data

Body weight/g Spleen weight/mg Protein in urine (score O-3) Lymphocytes, blood (X103/n1) Neutrophilic granulocytes, blood (XlOs/~l) Glutathione peroxidase (CH3)3COOH, erythrocytes Glutathione peroxidasc (HzOa), liver (70) Glutathione peroxidase (CHs)sCOOH, liver (%)

Vi~m~E~dSe

(%)

Number of animals

+

-

396 I 9 508 f24 1.2 It 0.10 14.ort 1.5 2.4 + 0.5 100 100 100

392 1:18 756 -c 33 1.0 i 0.13 13.1 zf 1.2 5.9-t 0.6 7 -t3 2 + 1 8 tl

8 8 16 8 8 8 8 8

568 TABLE VIII RATIO OF PENTANE: ETHANE IN EXPIRED BREATH OF RATS AFTER STIMULATION OF IN VIVO LIPID PEROXIDATION BY VARIOUS TREATMENTS

Remaining hydrocarbon In the animal hddirg chamber(%)

Animals that had been on vitamin E- and Sedeficient diets (for composition see Methods and Materials) for 13 weeks were fasted overnight and used for alkane measurements, as described in Methods and Materials. Other animals received sunflower oil diets (see Table I) with adequate vitamin E and selenium. Values are mean + S.D. (%). Treatment

Number of animals

0.6 ? 0.12 0.9 f 0.13

none (controls) -vitamin E; -Se

8 8

0.7 f 0.10

+cc4

4

Ratio of alkanes produced CsHiz

: CzH6

reported. As is evident in Table IX, the composition of alkanes in these faecal gases is different from that produced in expired air as a result of either stimulated or non-stimulated in vivo lipid peroxidation. If faecal or intestinal gases form an appreciable contribution to the alkanes present in the animals’breath, especially ethylene, propane and butane would have been easily detected. However, under our conditions, using rats fasted overnight, considerable amounts of faeces were never present, and only ethylene was detected some-

TABLE IX HYDROCARBONS EVOLVED (pmol) FROM FAECES OF RATS ON A SUNFLOWER OIL DIET

0.6 g

Faeces (0.6 g) of rats on sunflower oil diets (see Table I) with adequate vitamin E and selenium were placed in lOO-ml serum bottles provided with rubber caps and incubated at room temperature, 22-24’C. Headspace samples were drawn through the rubber caps with a gastight syringe and injected into the gas chromatograph. A representative experiment is shown. Time (min) 20 40 60 90 120 5 days

C2H4

C2H6

C3W3

14 12 22 28 36 1050

21 20 33 41 56 500

14 16 _ 43 44 5 990

C4H10

5 4 19 25 3500

CsH12

8 _ 13 22 19 450

20 -

I

0

I

15

30

,

45

60

90

120 time(min)

Fig. 5. Uptake of hydrocarbons by a rat (350 g) placed in (2.5 1) animal holding chamber and provided with known amounts of various hydrocarbons. Known amounts of hydrocarbons were injected into the animal holding chamber. No hydrocarbon loss could be measured under identical conditions in the absence of the animal. o, Ethane; a, propane; 0, pentane; 0, hexane.

times, in low concentrations. Methane was not quantitated because it was too close to the air pressure peak following sample injection. The presence of alkanes in air due to (industrial) air pollution may be another problem. All gases except ethane and possibly ethylene are taken up by the animal, presumably in its fat depots (Fig. S), and are slowly released thereafter. This phenomenon increases with increasing chain length of the alkanes. It is, therefore, necessary to have the animal in alkane-free air for several hours. Release of hydrocarbons by animals previously exposed to 10e6 g/kg levels of these gases has then reduced to undetectable levels.

569

In this investigation the effects of varying the composition of fatty acids in liver lipids, by changes in dietary lipids, on the products of in vivo lipid peroxidation have been studied. Lipid peroxidation occurs in suspensions of microsomal fractions and the endogenous fatty acids in the membrane phospholipids are the substrates of lipid peroxidation [32]. It may, therefore, be expected that the nature and quantities of various polyunsaturated fatty acids in the membrane phospholipids are important determinants of the products of lipid preroxidation. As has been shown earlier [28,32], it was found that the nature of the dietary lipids has a marked effect on the fatty acid composition of the phospholipids of the liver (Table V). Linoleic acid was readily incorporated into liver phospholipids and, to a certain extent, also into liver triacylglycerols, but the 21 : 3 (n-7) acid was mainly, if not exclusively, incorporated into phospholipids. It has been demonstrated that also 18 : 3 (n-4) acid is incorporated into membrane phospholipids, although to a lesser extent than the n-7 acid [28]. The variations in adipose tissue triacylglycerols as a result of these dietary manipulations are probably not significant. The fact that the composition of the alkanes in exhaled air, the products of in vivo lipid peroxidation, corresponds closely to the fatty acid composition of liver phospholipids demonstrates that the alkanes are indeed derived from the terminal end of the fatty acid molecule. Since the 21 : 3 (n-7) fatty acid was not present to any extent in triacylglycerols, it follows that under our experimental conditions liver or adipose tissue triacylglycerols do not make significant contributions to alkane production. The possibility exists that CCL, induces lipid peroxidation, and consequently alkane formation, only in the endoplasmic reticulum of the liver cell, whereas fatty acids were determined in the whole liver. However, the fatty acid composition of the liver microsomes (not shown) was nearly similar to that of whole liver. Other dietary or experimental conditions may of course involve other cellular lipids, such as triacylglycerols, as indicated by the occurrence of the Yellow Fat disease, a vitamin E deficiency symptom that has been observed in depot fat of various animal species [33]. It should be noted that essential fatty acid-deficient animals have appreciable quantities of palmitoleic

acid (n-7) in their phospholipids as well as in their triacylglycerols (Table V). However, no hexane was detected in expired air of these animals unless dietary quantities of 21 : 3 (n-7) were given. This indicates that palmitoleic acid is much more resistant to peroxidation than the highly unsaturated 2 1 : 3 (n-7). The 20: 3 (n-9) abundantly present in essential fatty acid-deficient animals (‘Mead acid’) should, on peroxidation, yield octane. Retention times for this alkane were too long under our experimental conditions to permit quantitation. An interesting but not well understood finding is that under most conditions more ethane than pentane is produced as a result of in vivo lipid peroxidation, although membrane phospholipids contain much more n-6 than n-3 acids. This may indicate that vitamin E has a more specific role in the protection of unsaturated fatty acids, distinct from that of other lipid antioxidants [34]. However, since ethane is probably retained by the organism to a less extent than pentane, it is possible that this fact is responsible for the unexpected ratio of the alkanes in expired air, as derived from the n-3/n-6fatty acids in tissue lipids. Therefore, more pentane could be produced than is detectable in the animal chamber air. It is clear, therefore, that the fatty acid composition of cellular lipids is an important, but not the only dietary determinant of the products and extent of lipid peroxidation in vivo. In healthy animals lipid peroxidation in vivo is low, but is set free by dietary deficiencies and certain chemicals and drugs. Acknowledgements The authors want to thank Mr. J.S.W. Kleinekoort for animal care and Mr. H.C.M. Van der Knaap for statistical advice. References Recknagel, R.O. (1967) Pharmacol. Rev. 19,145-208 Slater, T.F. (1966) Nature 209,36-40 Tappel, A.L. (1973) Fed. Proc. 32,1870-l 874 Desai, I.D. and Tappel, A.L. (1963) J. Lipid Res. 4, 204-207 Packer, L., Deamer, D.W. and Heath, R.L. (1967) Adv. Geront. Res. 2, 77-120 Green, J. (1972) Ann. N.Y. Acad. Sci. 203,29-44 Tappel, A.L., Fletcher, B. and Deamer, D.W. (1973) J. Geront. 28.415-424

570

8 Barber, A.A. and Bernheim, F. (1967) Adv. Geront. Res. 2,355-403 9 Riely, CA., Cohen, G. and Lieberman, M. (1974) Science 183,208-210 10 Dillard, C.J., Dumelin, E.E. and Tappel, A.L. (1977) Lipids 12,109-114 11 Hafeman, D.G. and Hoekstra, W.G. (1977) J. Nutr. 107, 656-665 12 Hafeman, D.G. and Hoekstra, W.G. (1977) J. Nutr. 107, 666-672 13 Konze, J.R. and Elstner, E.F. (1978) Biochim. Biophys. Acta 528,213-221 14 Lieberman, M. and Hochstein, P. (1966) Science 152, 213-214 15 Fu, PC., Zic,V. and Ozimy, K. (1979) Biochim. Biophys. Acta 585,427-434 16 Koster, U., Albrecht, D. and Kappus, H. (1977) Toxicol. Appl. Pharmacol. 41,639-648 17 Young, R.G. and Tappel, A.L. (1978) Exp. Geront. 13, 457-459 18 Kates, M. (1972) in Laboratory Techniques in Biochemistry and Molecular Biology, Vol. 3, 347-3.53 (Work, T.S. and Work. E., eds.), Elsevier/North-Holland, Amsterdam 19 Giinzler, W.A., Kremers, If. and FlohB, L. (1974) Z. Klin. Chem. Klin. Biochem. 12,444-448 20 Jacobs, E.E., Jacob, M., Sanadi, D.R. and Bradley, LB. (1956) J. Biol. Chem. 223,147-156 21 Beerthuis, R.K., Nugteren, D.H., Pabon, H.J.J. and Van

Dorp, D.A. (1968) Recueil87,461-475 22 Beerthuis, R.K., Nugteren, D.H., Pabon, H.J.J., Steenhoek, A. and Van Dorp, D.A. (1971) Recueil 90,943953 23 Lieberman, M. and Mapson, L.W. (1964) Nature 204, 343-345 24 Dumelin, E.E. and Tappel, A.L. (1977) Lipids 12, 894900 25 Pryor, W.A. (1976) Free Radicals in Biology, Academic Press, New York, Vol. 1, pp. 35-36 26 Horvat, R.J. and Lane, W.G. (1964) Nature 203, .523524 27 Dumelin, E. and Tappel, A.L. (1977) Lipids 12 (ll), 894-900 28 Houtsmuller, U.M.T. (1975) in The Role of Fats in Human Nutrition, (Vergroesen, A.J., cd.) pp. 331-351, Academic Press, New York 29 Hafeman. D.G., Sunde, R.A. and Hoekstra, W.G. (1974) J. Nun. 104.580-587 30 Pedersen, N.D., Whanger, P.D. and Weswig, P.H. (1975) Nutrition Reports international 11,429-435 31 Lawrence,R.A.andBurk, R.F. (1976) Biochem. Biophys. Res. Commun. 71,952-958 32 Hammer, CT. and Wills, E.D. (1978) Biochem. J. 174, 585-593 33 Wensvoort, P. (1974) Neth. J. Vet. Sci. 99,1060-l 065 34 Green, J., Edwin, IX., Diplock, A.T. and Bunyan, J. (1961) Nature 189,747-749