Effect of age and extent of dietary restriction on hepatic microsomal lipid peroxidation potential in mice

I I ~;EVI~ R ~;( IFNTIFI( P[ B[ ISHERS; I R I I & N D

Mechanisms of Ageing and Development 72 (1993) 155-163

Effect of age and extent of dietary restriction on hepatic microsomal lipid peroxidation potential in mice L. Jane Davis *a, Bruna Tadolini a, Pier Luigi Biagi a, Roy Walford b, Federico Licastro c aDepartment of Biochemistry. University of Bologna, Via lrnerio 48, 40128 Bologna, Italy hDepartment of Pathology, University of California, Los Angeles. CA, USA ¢Department of Experimental Pathology, University of Bologna, Via San Giacomo 14, 40127 Bologna, Italy (Received 26 August 1993; accepted 26 August 1993)

Abstract

Lipid peroxidation potential in hepatic microsomes from young and old mice following two different caloric restriction regimens was measured by a colorimetric thiobarbituric acid method under conditions where Fe 2÷ autoxidation and free oxygen radical production were undetectable. Peroxidation was highest in the young (3.5-month-old) slightly restricted group (caloric intake 75% of ad libitum mice) but very low in young severely restricted (caloric intake 50% ofad libitum mice) and in both old (27-month-old) slightly and severely restricted groups. Very old (45-month-old) severely restricted animals had intermediate lipid peroxidation potentials. Fatty acid composition of liver homogenates was also determined. Significant differences between groups were found for only three fatty acids. Linoleic acid (18:2(n-6)) decreased in aged slightly restricted animals while it remained stable in severely restricted animals during aging. Dihomo-gamma-linolenic acid (20:3(n-6)~ was higher in very old restricted animals than in old slightly restricted animals. Docosahexaenoic acid (22:6(n-3)) decreased in old slightly restricted animals. These results indicated that the effect of diets on hepatic fatty acid composition and the potential for microsomal lipid peroxidation in mice was dependent on the degree of caloric restriction and age. Key words: Aging; Food restriction; Lipid peroxidation; Lipid composition; Thiobarbituric acid * Corresponding author. 0047-6374/93/$06.00 © 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. SSDI 0047-6374(93)01400-3

156

L.J, Davis et a l . / Mech. Ageing Dev. 72 (1993) 155-163

1. Introduction

Restriction of caloric intake, along with supplementation of essential nutrients is a reproducible method for increasing longevity in rodents. For instance, restriction of calories in the diet has been shown to extend average and maximum life span of mice and rats [1]. Interestingly, life span correlates with the severity of caloric restriction (CR). In fact, severe caloric restriction equal to 50% caloric intake of ad iibitum mice induces longer survival than that observed in mice whose caloric restriction was approximately 75% of the calories given to ad libitum fed animals [1]. The reasons for the differences in life extension observed in slightly (75%) and severely (50%) restricted mice still remain unclear. One reason may be that the incidence of chronic degenerative disease associated with aging is delayed and their intensity decreased according to the amount of the CR [2]. The mechanism(s) responsible for the antiaging effects of CR are not well understood. The free radical theory of aging predicts that senescence might be attributed at least in part to damages caused by radical generation associated with cellular metabolic activity [3]. According to this notion, one of the mechanisms responsible for life extension induced by CR might be a decreased generation of radicals followed by diminished membrane damages. However, the influence of CR upon these damaging reactions is yet to be completely defined. Lipid peroxidation can be initiated by free radical extraction of hydrogen atoms from unsaturated fatty acids and experimental variables may influence the detection of lipid peroxidation. For instance, iron salts are known to trigger lipid peroxidation by reacting with molecular oxygen to form the hydroxyl radical. The addition of Fe z+ to membranes in phosphate buffer or other moderately basic buffers, causes the autoxidation of the metal, the generation of the hydroxyl radical and lipid hydroperoxide (LOOH) independent lipid peroxidation [4,5]. Trace amounts of LOOH react with Fe 2+, which activates LOOH dependent lipid peroxidation [6]. In the absence of Fe z+ autoxidation, Fe 2+ dependent lipid peroxidation gives an estimate of the peroxidation potential of the membranes due to the presence of LOOH. Thus, it is crucial to use defned experimental systems and buffers with or without iron for determination of lipid peroxidation. Here we present results regarding the effect of dietary restriction on lipid peroxidation potential using experimental conditions where the Fe 2+ dependent lipid peroxidation was evaluated in hepatic microsomes from mice fed two different diets where the CR was 25% and 50°/,, respectively of unrestricted animals. The fatty acid composition of liver homogenates was also monitored. 2. Materials and methods 2.1. Animals and diets

Females of a F1 hybrid mouse strain (C3H-SW/Sn × C57BLI0.RII1/Sn) were maintained as previously described [7] and fed a semipurified diet. In brief, minimally restricted mice received approximately 75% of the caloric intake normally consumed by ad iibitum mice. This group of animals represents the control group for the severely restricted animals. The severely restricted mice followed a similar

L.J. Davis et al./ Mech. Ageing Dev. 72 (1993) 155-163

157

semipurified diet containing 50% of the calories normally consumed by ad libitum animals.

2.2. Preparation of hepatic microsomes Livers were aseptically removed and kept at -80°C until analysis. Frozen livers were weighed and the total liver homogenized in three volumes of MOPS 50 mM pH 7.0 with 1.15% KC1 with a Potter-Elvehjem homogenizer. Homogenates were centrifuged at 600 x g and the supernatants then centrifuged at 9000 × g. The pellets were washed and resuspended in MOPS 5 mM buffer, pH 6.5. The resuspended pellet was centrifuged as before and supernatant was pooled with the supernatant from the first centrifugation. Cytosolic proteins were separated from microsomal fractions by centrifugation of the combined supernatants at 105 000 x g, (60 rain, 4 °C). Microsomal protein concentrations were determined using the biuret reaction [8]. To minimize sample size used for protein determination, 10/~1 of cytosolic protein fractions were brought to a final volume of 0.5 ml with 0.75 N NaOH. After vortex mixing, 0.5 ml of the biuret reagent was added.

2.3. Lipid peroxidation measurement Potential lipid peroxidation of microsomes was measured in vitro by following the formation of thiobarbituric acid reactive materials according to Buege and Aust [9]. Latent lipid peroxidation was triggered by Fe z+ addition. Samples were incubated at room temperature in 5 mM MOPS buffer, pH 6.5. Lipid peroxidation was determined by the measurement of thiobarbituric acid reactive materials. The reactions were stopped by the addition of 1.5 ml of 1% thiobarbituric acid containing 10 #1 of 2% butylated hydroxytoluene (BHT) and of 1.5 ml of 20% glacial acetic acid, pH 3.5. The tubes were heated for 10 rain at 100°C to develop the color. The resulting chromogens were measured at 532 nm against appropriate blanks.

2.4. Fatty acid analysis The fatty acid composition of total liver homogenates was determined by gas chromatography (Carlo Erba model 4160) using a capillary column (SE 52 0.10-0.15 txm) at a programmed temperature (60-320°C) as previously described [10].

2.5. Statistical analysis Statistical evaluation was made with the unpaired Student's t-test.

3. Experimental results

3.1. Potential lipid peroxidation Iron induced microsomal lipid peroxidation potential measured as a function of protein concentration is reported in Fig. 1 and statistically analysed in Table 1. Young (3.5-month-old) slightly restricted animals (SRY) showed higher peroxidation potential than all other groups, statistical differences are reported in Table 2. In general, potential peroxidation was significantly higher in SRY animals and increased with increasing protein concentration. A very low potential peroxidation was present in old (27-month-old) slightly restricted (SRO) animals. Peroxidation

158

L.J. Davis et al. J Mech. Ageing Dev. 72 (1993) 155-163

0.05

0.04

I~

o sR't(3) '7 RY(2)

sRo(v)

• •

RO(7)

[3 RVO(4)

0.03

t-o tt~ 0.02 .,m

e~

.,~

)

0.01

0.00

0.0

I

l

1

1

I

I

I

0.2

0.4

0.6

0.8

1.0

1.2

1.4

[P~o'mn~] nag

Fig. 1. Lipid peroxidation potential of murine hepatic microsomes as a function of protein concentration. Absorbanee values are expressed as mean ± S.E. The iron salt concentration was 10 p.M. The five groups of animals are: SRY = 75% caloric intake of ad libitum animals, 3.5 months old; SRO = 75% caloric intake, 27 months old; RY = 5(/'/0 caloric intake, 3.5 months old; RO = 50% caloric intake, 27 months old; RVO = 50% caloric intake, 45 months old mice. The number of animals evaluated are in parenthesis.

Table 1 Summary of significant differences, P-values, according to the unpaired Student's t-test between groups for lipid peroxidation as a function of protein (Fig. 1) or Fe 2+ (Fig. 2) concentration Analysis

SR 35 vs. SR 27 SR 35 vs. R 27 SR 3'5 vs. R 45

Protein concentration (mg) 0.4

0.8

1.2

1.5

0.003 0.037 --

0.029 0.049 --

0.0004 0.02 0.047

0.007 0.027

Fe 2+ cocnentration

SR 3"5 vs. SR 27 SR 35 vs. R 27 SR 27 vs. R 45

(#M)

5

10

15

20

0.02 0.04

0.0004 0.02 0.008

0.001 0.003

0.007 0.007

SR, slightly restricted; R, restricted. Superior numbers are age in months of the animal group.

0.33 0.16 21.7 2.1 0.21 12.1 21.3 16.3 0.31 0.42 0.68 15.3 0.76 3.2 4.0

44444444± ~ 4:~ 444-

0.05 0.3 0.6 0.7 0.06 2.6 3.1 0.7 0.10 0.03 0.32 1.2 0.17 0.3 0.5

0.48 0.17 27.3 2.9 0.26 8.5 30.6 7.8 0.21 0.78 0.75 12.1 0.61 3.1 2.4

4- 0.17 4- 0.02 4- 2.4 4- 0.4 ± 0.06 4- 1.0 4- 6.2 4- 1.6 4- 0.02 + 0.19 4- 0.13 ± 2.0 + 0.04 ± 0.8 -4- 0.6

0.33 0.19 25.8 2,0 0.38 10 22.0 14.3 0.23 0.44 0.79 13.3 0.79 3.5 0. 7

4- 0.02 ± 0.04 4- 1.2 ± 0.1 4- 0.04 4- 0 4- 0 4- 0.9 .a- 0.08 4- 0.05 (1) (I) (1) + 0.4 4- 0.1

3.5 (2)

3.5 (3)

27 (7)

Restricted

Slightly restricted

T h e n u m b e r s in p a r a t h e s i s r e p r e s e n t the n u m b e r o f a n i m a l s e v a l u a t e d .

14:0 15:0 16:0 16:1 17:0 18:0 18:1 18:2 18:3n6 18:3n3 20:3 20:4 22:4 22:5n6 22:6n3

F a t t y acid

Table 2 Percentage o f fatty acids in total h o m o g e n a t e s o f m u r i n e livers for e a c h a g e g r o u p o f b o t h diets

0.36 4- 0.09 0.20 4- 0.04 27.1 4- 0.8 3.4 ± 0.3 0.45 4- 0.07 9.2 4- 0.5 22.8 4- 1.5 12.9 ± 0.7 0.28 4- 0.05 0.48 4- 0.05 1.3 4- 0.3 13.5 4- 0.7 0.68 4- 0.12 3.3 4- 0.4 2.64-0.1

27 (71

0.36 + 0.05 0.34 + 0.13 28.4 4- 1.4 4.0 4- 0.22 0.46 + 0.11 8,5 4- 0.5 21.6 4- 2.0 13.3 4- 1.0 0.29 + 0.13 0.61 4- 0.15 1.4 4- 0.16 12.8 4- 0.7 0.71 4- 0.12 3.3 4- 0.3 2.24-0

45 (41

i

%

L.,I l)avis et a l

160

0.05

0.04

Meclr Ageing Dev. 72 (1993) 155-163

t o

sRv(3)

.

V

RY(2)

,

sRo(7) RO(7)

[] RVO(4)

0.03 ~X2 if3

o

0.02

e~ <

0.01

o.oo i I

10

I

20

[Fe +z] ~M Fig. 2. Potential lipid peroxidation of murine hepatic microsomes as a function of iron concentration. Absorbance values are expressed as mean ± S.E. Protein concentration was 1 mg/ml. The five groups of animals are: SRY = 75% caloric intake of ad libitum mice, 3.5 months old; SRO = 75% caloric intake, 27 months old; RY = 50% caloric intake, 3.5 months old: RO = 50OA,caloric intake, 27 months old; RVO = 50% caloric intake, 45 months old mice. The number in parenthesis represents the number of animals evaluated.

potential was very low in livers from y o u n g (3.5-month-old, RY), old (27-month-old, RO) and very old (45-month-old, R V O ) severely restricted mice. The effect o f increasing iron concentrations on lipid peroxidation potential is presented in Fig. 2. Protein concentration was held constant at 1 mg/ml and once again, the S R Y animals had the highest potential for lipid peroxidation. In these experimental conditions lipid peroxidative potential was low in SRO, R Y and R O mice. Lipid peroxidation increased in R V O animals. However, this later cohort of mice still showed lower peroxidative potential than S R Y mice. The R V O animals had a statistically significant higher potential lipid peroxidation than S R O animals only when 10 ~tM Fe 2÷ was used.

3.2. Fatty acid composition of hepatic homogenate The percentage o f fatty acids in total homogenates o f murine livers are reported in Table 2. Significant differences between groups were found for only three fatty acids. Linoleic acid (18:2(n-6)) decreased in S R O animals (P < 0.011) while it remained stable in restricted animals during aging. Dihomo-gamma-linolenic acid (20:3(n-6)) was higher in R V O animals than in S R O animals (P < 0.018). Docosahexaenoic acid (22:6(n-3)) decreased in S R O animals (P < 0.015).

L.J. Davis et al./Mech. Ageing Dev. 72 (1993) 155-163

161

4. Discussion

One causal explanation of biological aging is suggested by the free radical theory, originally proposed by Harman [3]. Free radicals are able to induce the peroxidation of polyunsaturated fatty acids. Thus, the free radical theory of aging is currently used as synonymous with the lipid peroxide theory. Age-dependent NADPH-dependent lipid peroxidation has been reported to occur in liver microsomes and mitochondria [11]. The free radical theory of aging predicts that senescence might be attributed at least in part to damages caused by radical generation associated with cellular metabolic activity. One of the mechanisms responsible for life extension induced by dietary restriction might be a decreased generation of radicals followed by diminished membrane damages. It is interesting to note that restriction of calories in the diet has been reported to decrease the age-associated lipid peroxidation in livers from mice [12] or rats [131 up to 24 months of age. The present report is the first to show data on lipid peroxidation for 45-month-old animals. Lipid peroxidation dependence on FeC12 concentration is complex. It has been shown that the transition metal catalysis is an essential factor in the generation of activated oxygen species and that pH and metal chelation affects the reactivity of metals with 'oxy-radicals' [14]. Moreover, buffers usually used for in vitro measurements of biochemical variables and, in particular, the oxidative parameters affects metal autoxidation [5]. In the absence of Fe 2÷ autoxidation, the lipid peroxidation initiated by Fe 2÷ estimates the peroxidation potential of membranes due to the presence of LOOH. In the present contribution we have studied the effects of age and diets on lipid peroxidation of liver microsomes using an experimental protocol where Fe 2÷ autoxidation and free radical oxygen production were absent. Comparison of the lipid peroxidation of different microsomes evaluated at a fixed protein and iron concentrations, could be misleading. Therefore, we determined the dependence of lipid peroxidation on both protein and iron concentration for each sample. Our results show that potential lipid peroxidation could be masked at some combinations of fixed concentrations of protein or iron. Our data also show that peroxidative potential was affected by dietary restriction. The main difference between the two dietary regimens was observed in young slightly restricted and very old severely restricted mice. The present findings extend previous data reported by Koizumi et al. [12] who found a decrement of lipid peroxidation in livers from female severely caloric restricted mice. Our observation suggests that severe CR acts early in the life on this parameter. Protection against peroxidative damage during development might play a role in decelerating senescence and in extending life span to its maximum. Hepatic microsomal yields in 45-month-old CR animals was very low because they had very small livers (data not shown). Therefore, percentages of fatty acids were measured in total liver homogenates. Linoleic acid content decreased in slightly CR cohorts. Linoleic acid is a delta-6desaturase (EC 1.14.99.5) substrate which gives rise to metabolites dihomo-gammalinolenic acid (20:3(n-6)) and docosahexaenoic acid (22:6(n-3)). The decrease in linoleic acid was reflected by a decrease in docosahexaenoic acid in slightly CR

162

L.J. Davis et al./ Mech. Ageing Dev. 72 1/993 ~ 155-16~

cohorts. In severely CR cohorts linoleic acid remained stable and was reflected by a stable docosahexaenoic acid concentration. However, in the slightly CR animals, the dihomo-gamma-linolenic acid remained stable with age while increasing in severely CR ones. The activity of delta-6-desaturase decreases with age in liver or other tissues [15,16]. This decrement might be induced by an age-associated change of membrane lipid composition. Severe CR appears also to influence this ageassociated metabolic pathway. Our data are partially in accordance with those of Laganiere and Yu [13] who reported that linoleic acid increased in caloric restricted male Fischer 344 rats; the slight discrepancies with the above study might be attributed to species and sex differences. In conclusion, our results indicated that CR affected the age-related peroxidative damage of hepatic microsomes depending upon the extent of restriction. Furthermore, severe caloric restriction significantly decreased the oxidative potential of hepatic biomembranes earlier than slight dietary restriction.

5. Acknowledgements This research was funded by MURST and CNR grants.

6. References 1 R. Weindruch, R.L. Walford, S. Fligiel, and D. Guthrie, The retardation of aging by dietary restriction: longevity, immunity and lifetime energy intake. J. Nutr., 116 (1986) 641-54. 2 R.T. Bronson and R.D. Lipman, Reduction in rate of occurrence of age related lesions in dietary restricted laboratory mice. Growth. Dev. Aging, 55 ( 1991) 169-84. 3 D. Harman, Aging: a theory based on free radical and radiation chemistry. Z Gerontol., 11 (1956) 298-300. 4 J.M. Guneridge, Reactivity of hydroxyl-like radicals discriminated by release of thiobarbituric acidreactive material from deoxy sugars, nucleosides and benzoate. Biochem. J.. 224 (1984) 761-7. 5 B. Tadolini, Iron autoxidation in Mops and Hepes buffers. Free Rad. Res. Commun., 4 (1987) 149-60.

6

J.M. Gutteridge and P.J. Kerry, Detection by fluorescence of peroxides and carbonyls in samples of arachidonic acid. Br. ,L PharmacoL. 76 (1982) 459-61. 7 F. Licastro, R. Weindruch, L.J. Davis and R.L. Walford, Effect of CR upon the age-associated decline of lymphocyte DNA repair activity in mice. Age 11, (1988) 48-52. 8 A.G. Gornall, C.J. Bardawill and M.M. David, Determination of serum proteins by means of the biuret reaction. J. Biol. Chem., 177 (1949) 751-766. 9 J.A. Buege and S.D. Aust, Microsomal lipid peroxidation. In S. Fleischer and L. Packer (eds.), Methods in Enzymology, no, 52, Academic Press, New York, 1978, pp. 302-310. 10 L.J. Davis, P.L. Biagi, C Spanb, R.L. Walford and F. Licastro, Caloric restriction maintains linoleic and arachidonic acid levels in murine liver during aging. Aging, 3 (1991) 405-7. 11 T.J. Player, D.J. Mills and A.A. Horton, Age-dependent changes in rat liver microsomal and mitochondrial NADPH-dependent lipid peroxidation. Biochem. Biophys. Res. Commun.. 78 (1977) 1397- 1402. t2 A. Koizumi, R. Weindruch and R.L. Walford, Influences of dietary restriction and age on liver enzyme activities and lipid peroxidation in mice. J. Nutr., 117 (1987) 361-7. 13 S. Laganiere and B.P. Yu, Anti-lipoperoxidation action of food restriction. Biochem. Biophys. Res. Commun., 145 (1987) 1185-91.

L.J. Davis et al. /Mech. Ageing Dev. 72 (1993) 155-163 14

163

S.D. Aust, L.A. Morehouse and C.E. Thomas, Role of metals in oxygen radical reactions. J. Free. Rad. Biol. Med., 1 (1985) 3-25. 15 P.L. Biagi, A. Bordoni, S. Hrelia, M. Celadon and D.F. Horrobin, Gamma linolenic acid dietary supplementation can reverse the aging influence on rat liver microsome delta-6-desaturaase activity. Biochim. Biophys. Acta, 1083 (1991) 187-192. 16 J.A. Lopez Jimenez, A. Bordoni, S. Hrelia, C.A. Rossi, E. Turchetto, S.Z. Navarro and P.L. Biagi, Evidence for a detectable delta-6-desaturase activity in rat heart microsomes:aging influence on enzyme activity. Biochem. Biophys. Res. Commun., 192 (1993) 1037-1041.