Short-Term Dietary Lipid Manipulation Does Not Affect Survival in Two Models of Murine Sepsis PHYLLIS CLOUVA-MOLYVDAS, MD*; MICHAEL D. PECK, MD†;

AND

J. WESLEY

ALEXANDER, MD, SCD‡

From the Shriners Burns Institute, Cincinnati, Ohio; the *Intensive Care Unit, Laiko General Hospital, Athens, Greece; the †University of Miami/Jackson Memorial Medical Center, Miami, Florida; and the ‡University of Cincinnati Medical Center, Cincinnati, Ohio

control diet that had 12% of total calories as corn oil. At the end of 2 weeks of feeding the experimental diets, mice were challenged with Ps aeruginosa intraperitoneally and mortality was recorded over 1 week. After 3 weeks of feeding the experimental diets, mice were challenged with S typhimurium and mortality was recorded over 2 weeks. No significant differences were seen on survival among groups fed different levels of fat, as well as different sources of fat. We conclude that, overall, 2and 3-week manipulation of dietary fat does not affect outcome from infection in these models. (Journal of Parenteral and Enteral Nutrition 16: 343-347, 1992)

ABSTRACT. Dietary lipid manipulation has been shown to have various effects on the immune system, depending on the amount of fat, degree of saturation, and type of fat used. In this study we investigated the role of different sources of fat at different levels on the survival of mice in two models of peritonitis, one with Pseudomonas aeruginosa and the other with Salmonella typhimurium. CF1 mice were pair-fed diets with 5% or 40% of total calories as fat. The source of fat used was coconut oil, oleic acid, safflower oil, or fish oil. Three other diets were tested as well, one with no fat, one with only 0.5% of total calories linoleic acid as the only source of fat, and a

Lipids have a wide range of immunomodulatory effects. These effects are thought to be mediated through modulation of eicosanoid synthesis, changes in cell membrane properties, changes at the receptor binding sites, and finally through synthesis, release, and/or binding of various substances. The role of dietary lipids in the modulation of immune function has been studied both in vitro and in vivo and has been reviewed exten-

affected by dietary lipids. The present study was designed dietary fat, as well as different sources of fat, on the survival of mice in two models of sepsis.

to compare different levels of

METHODS

sively.1-5

In

There are numerous studies on the effects of different levels of dietary fat, essential fatty acid deficiency, saturated vs polyunsaturated fat, different ratios of m-6 to w-3 fatty acids, as well as combinations of the above, on different aspects of the immune response with sometimes conflicting results. Humoral immunity, as assessed by primary and secondary responses to both T-cell-dependent and T-cell-independent antigens, seems to be affected by dietary lipid manipulation.~ Cellular immunity is also affected by dietary lipids. Their effect has been studied on delayed type hypersensitivity,7,S mitogenic responses to T-lymphocyte mitogens,9 and lymphocytemediated cytotoxicity.l0,11 Finally, dietary lipid manipulation has been found to alter macrophage and neutrophil

conducting this research,

the

investigators adhered

to the &dquo;Guide of Laboratory Animal Facilities Care&dquo; as set forth by the Committee on the Guide for Laboratory

Resources, National Academy of Sciences-National Research Council and to the regulations of the National Institutes of Health. These protocols were reviewed and approved by the Animal Care and Use Committee of the University of Cincinnati. Female CF1 mice weighing 20 to 22 g were obtained from Charles River Suppliers. They were housed five in a cage, in a well-ventilated room with 12-hour alternating light and dark cycles and with temperatures kept at 22 to 24°C. They were given water and natural chow (Wayne Rodent Blox, Libertyville, IL) ad libitum. The mice were allowed to acclimate in the laboratory facilities for 1 week. At the end of the acclimation period they were fed a standard purified diet based on the recommendations of the American Institute of Nutrition (AIN 76) (Table I) for 4 days ad libitum. They were then weighed and started on experimental diets. During this period the animals were pair-fed isocaloric amounts of food. All diets were prepared in our laboratory from food-

function.12-14 From all the evidence accumulated so far, dietary lipids have been shown to influence different aspects of the immune response and to exert an overall effect on the immune response to various disease states. Neoplasms autoimmune diseases,16,17 infection,&dquo; trauma,’9 and transplanted organ survivaFo-22 have been shown to be

grade ingredients (ICN Biochemicals, Cleveland, OH). were based on the AIN 76 formula with either an increase of the fat content to 40% of total calories, or a decrease of the fat content to 5% of total calories. When the fat content was increased to 40% of total calories, ie,

They requests: J. Wesley Alexander, MD, ScD, University of Cincinnati College of Medicine, 231 Bethesda Avenue, Cincinnati, OH 45267-0558.

Reprint

343 Downloaded from pen.sagepub.com at UQ Library on March 13, 2015

344 TABLE I

Composition of experimental diets

EFA, essential fatty acids; AIN, American Institute of Nutrition. * Butylated hydroxytoluene was added in all diets as an antioxidant (0.2 mg/g of fat). t Corn oil was used as the fat source in the control diet. $ Four different oils were used in four different diets: safflower oil, coconut oil, oleic acid, and MaxEPA oil.

by weight, a weighted reduction of carbohydrates made. When the fat content was decreased a weighted increase in carbohydrates was made. Four different fat sources were used, coconut oil, safflower oil, oleic acid, and MaxEPA oil. In addition, a fat-free diet was used, as well as a minimum essential fatty acid diet that had 0.5% of total calories as linoleic acid as the only source of fat. The control diet used was the AIN 76 formula, with 12% of total calories as corn oil (Corn). Some of the fats used contained restricted or absent amounts of linoleic acid, and essential fatty acid deficiency could confound the results. Therefore, linoleic acid, 0.3% by weight, was added in all diets, except in the 0% fat diet. To prevent lipid oxidation, butylated hydroxytoluene, 0.02 mg% per gram of fat was added. In addition, all high-fat diets were kept frozen and packed in sealed bags gassed with N2. The composition of all the experimental diets used is shown in Table I. Safflower oil is predominately (>70%) linoleic acid (18:2w-6). Corn oil is also predominately (>60%) linoleic acid. Coconut oil is predominately saturated fats (>90%), with little or no linoleic acid. MaxEPA oil is composed of approximately 35% saturated fats, 25% monounsaturated fats, and 35% w-3 fatty acids (equal amounts of eicosapentaenoic and docosahexaenoic acids). Oleic acid is a monounsaturated fatty acid (l8:lw9). Six experiments were performed. Mice were age matched for each experiment. The first three experiments compared the effect of feeding the experimental diets for 2 weeks on the survival of mice challenged with Pseudomonas aeruginosa intraperitoneally (IP). At the end of the 2 weeks the animals were weighed and injected IP with 5 X 106 organisms of Ps aeruginosa. Mortality was recorded over 7 days. The next three experiments compared the effect of feeding the experimental diets for 3 weeks on the survival of mice challenged with Salmonella typhimurium IP. At the end of the 3 weeks the animals were weighed and injected IP with 104 organisms of S typhimurium. Mortality was recorded over a 2 weeks. Dilutions of Ps aeruginosa and S typhimurium were to 20% was

to inoculate two cultures. After incubation overnight at 37°C, the cultures were centrifuged at 3000 rpm for 10 minutes, and the resulting pellets were resuspended in

0.9% NaCl. After they were washed three times they were diluted to the desired final concentrations of organisms per milliliter. The number of viable organisms was confirmed by serial dilutions and plating. Statistics were calculated on a Maclntosh SE computer using Statview 512+ software. Analysis of variance (ANOVA) and Sheffe’s F test were used to compare weights. The X2 test of independence was used to analyze mortality rates. An a level of .05 was accepted as significant. Power calculations were done based on a .05, ~ .20, and on the observed differences among groups. Power was defined as the probability that if no significant difference was observed between groups, that indeed this lack of difference reflects the real relationship between =

=

groups.

RESULTS

Experiments I, II,

and III

Table II summarizes the

weight changes

of the mice

fed the different diets, at the end of the 2-week period, for the first three experiments. To compare the weight changes of the different groups, a two-factor ANOVA was performed, one for the type of fat (p .0102) and =

the other for the level of fat Weight changes of the

follows: An aliquot of microorganisms, from stock cultures in 10% glycerol kept at -40°C, was used

prepared as

Downloaded from pen.sagepub.com at UQ Library on March 13, 2015

mice

(p

=

.0743). The weight gain

TABLE II weeks of feeding (experiments I, II, and III)

after 2

345 of the animals on safflower oil diets (0.8% ± 1.0) was significantly lower than the weight gain of the animals on MaxEPA oil diets (5.4% ± 1.2). There was no significant difference in the weight gain of the animals when the groups with different levels of fat (5% vs 40%) were compared. There was also no significant difference in weight gain between the group fed the 0% fat diet, the group fed the control diet, and the group fed the 0.5%

linoleic acid diet (ANOVA p .1215). The weight gain of the control mice was not significantly different from that of any of the other groups (one-way ANOVA/ Scheffe’s F test). The survival data for the first three experiments are shown in Table III. The overall survival for experiments I, II and III combined was 70% in the control group, 90% in the 0% fat group, 73% in the 0.5% linoleic acid group, 73% in the 5% coconut oil group, 73% in the 40% coconut oil group, 80% in the 5% oleic acid group, 73% in the 40% oleic acid group, 90% in the 5% safflower oil group, 80% in the 40% safflower oil group, 70% in the 5% MaxEPA oil group, and 80% in the 40% MaxEPA oil group. These differences were not statistically significant =

(overall X2, p

=

.6753,

power

=

.35).

Comparison of survival between groups fed 5% fat vs 40% fat, regardless of the type of fat, showed no significant difference (p .8772, power < .1). There was also no significant difference when the survival was compared between groups fed different types of fat, regardless of the level of fat in the diet (p .4283, power .31). =

=

=

Experiments I V, V, and

.0292). In these experiments there was a significant difference in weight gain between animals fed the control diet (11.8% ± 1.5) and animals fed the 0.5% linoleic acid diet (6.9% ± 1.4), whereas there was no significant difference between the above two groups and the group fed the 0% fat diet (ANOVA for the three groups, p .0369). The survival data of experiments IV, V, and VI are shown on Table V. The overall survival for experiments IV, V, and VI combined was 42% in the control group, 30% in the 0% fat group, 35% in the 0.5% linoleic acid group, 39% in the 5% coconut oil group, 29% in the 40% coconut oil group, 37.5% in the 5% oleic acid group, 33% in the 40% oleic acid group, 28% in the 5% safflower oil group, 39% in the 40% safflower oil group, 40% in the 5% MaxEPA oil group, and 45% in the 40% MaxEPA oil group. These differences were not statistically significant (overall X2, p .32). Length of .7097, power survival was not significantly different among the groups. Comparison of survival between groups fed 5% fat vs 40% fat, regardless of the type of fat, showed no significant difference (p .8772, power < .1). There was also no significant difference when the survival was compared between groups fed different types of fat, regardless of the level of fat in the diet (p .4282, power .31). Comparison of survival between groups fed 5% fat vs 40% fat, regardless of the type of fat, showed no significant difference (p .9606, power < .1). There was also no significant difference when the survival was compared between groups fed different types of fat, regardless of the level of fat in the diet (p .3784, power .33).

p

=

=

=

=

=

=

=

=

=

=

VI

Table IV summarizes the weight gain of the animals fed different diets at the end of the 3-week period for experiments IV, V, and VI. To compare the weight changes of the different groups, a two-factor ANOVA was performed, one for the type of fat (p = .0311) and one for the level of fat (p .0292). The weight gain of the animals on the MaxEPA oil diet (8.2% ± 1.3) was significantly lower than the weight gain of the animals on the safflower oil diet (12.3% ± 1.2) and of the animals on the oleic oil diet (12.0% ± 1.1) (ANOVA p .0311). The weight gain of the animals on the 40% fat diets was significantly higher than the weight gain of the animals on the 5% fat diets (11.8% ± 0.9 vs 9.3% ± 0.7) (ANOVA =

=

DISCUSSION

In this with

study two different models of sepsis were used, intraperitoneal injection of Ps aeruginosa and the other with intraperitoneal injection of S typhimurium. The sepsis induced by the intraperitoneal injection of Ps aeruginosa is acute, clearance of the organism is very rapid, and death results within 48 hours. The host defenses in this model are predominantly phagocytes, especially neutrophils. The sepsis induced by S typhione

murium has a much slower course and death occurs within 2 weeks. This organism is an intracellular parasite and requires cell-mediated defenses for eradication. The types of fat used in this study were chosen because

TABLE III

Survival of mice fed different

*

Data

are

presented

as

the number of mice that

lipid diets after intraperitoneal challenge

survived/total

with Pseudomonas

number of mice in the group.

Downloaded from pen.sagepub.com at UQ Library on March 13, 2015

aeruginosa*

346 Weight changes of the

mice

TABLE IV weeks V, and VI)

after 3

of feeding (experiments IV,

of their fatty acid composition. Most of those fatty acids have been shown, from previous studies, to have an effect on the immune system. Linoleic acid (18:2 m-6) is the precursor of arachidonic acid, which gives rise to the dienoic prostaglandins (PGs), and the tetraenoic leukotrienes (LTs). These compounds have a wide range of biologic effects and play an important role in inflammatory and immune reactions. Prostaglandin E2 inhibits various aspects of the immune response, including lymphocyte proliferation, lymphokine secretion, and macrophage, and natural killer cell activity. 9,23-26 I,Ts also modify lymphocyte function in vitro .2’ Administration of PGs of the E series has anti-inflammatory or immunosuppressive effects in vivo.2$ Modification of the availability of substrate by dietary manipulation can clearly alter PG and LT synthesis. Furthermore, dietary linoleic acid and polyunsaturated fats (PUFAs) of the m-6 family can encourage the development of infection by enhancing Gram-negative bacteria survival .21 It has been shown to be immunosuppressive22 and to worsen survival after burn. 30 The MaxEPA oil is rich in co-3 PUFAs, which are precursors of trienoic PG and pentaenoic LT, whose effects are generally less potent than those of the dienoic PG and tetraenoic LT, and are less inflammatory.31 The w-3 PUFA can decrease production of dienoic PG and tetraenoic LT32 by decreasing membrane phospholipid arachidonic acid. Oleic acid has been found to be immunostimulatory. Animals fed oleic acid had markedly hyperstimulated macrophages with dramatic increases in stimulation in-

dices when added back to a normal mixed lymphocyte response cultures.14 Coconut oil might also lower arachidonic acid in membrane phospholipids by changing eicosanoid pathways. Chronic coconut oil feeding has been found to lower spleen phospholipids significantly.33 We did not find any differences in survival in animals fed different levels of fat (0%, 5%, and 40% of total calories) or different sources of fat (corn oil, coconut oil, oleic acid, safflower oil, and MaxEPA oil) compared with animals fed the control diet (12% of total calories as corn oil) or the minimum essential fatty acid diet (0.3 g% linoleic acid). Our animals although pairfed showed differences in weight gain at the end of the feeding period. However, weight gain did not appear to affect survival in these experiments, so we presume that small changes in nutritional status did not influence the overall outcome. Although the concentration of protein by weight was kept constant, the percentage of protein expressed as percent of calories varied between 17% and 22% among the various diets. This fluctuation is due to the difference in caloric density between the various diets. We do not believe that this has any physiologic importance because much higher differences in the protein content of diets (5%, 20%, and 40% of protein) have been shown in previous experiments not to influence survival in the intraperitoneal Pseudomonas infection

mode1.34 The lack of effect of dietary lipid manipulation on survival between the groups cannot exclude the fact that changes in individual aspects of immune function exist, but it is possible that they outweigh each other, so that a net effect is not visible. This lack of effect could also be attributed to the length of the feeding period. It is possible that 2 to 3 weeks might not be enough to show an effect on survival. In our study we used the 2- and 3week period of feeding that was shown to have had an effect in a murine burn model in our laboratory.35 Although incorporation of dietary fatty acids into cell membranes is very rapid,36 effects on specific aspects of the immune system require a variable amount of time. Finally, it is possible that levels of fat higher than those tested might have had an effect on survival. It is also possible that the route of administration of fat (parenteral us enteral) as well as the quality of fat might influence survival in these and other animal models of infection.

TABLE V Survival of mice fed different lipid diets after intraperitoneal challenge with Salmonella typhimurium*

*

Data

are

presented

as

the number of mice that

survived/total number of mice. Downloaded from pen.sagepub.com at UQ Library on March 13, 2015

347 ACKNOWLEDGMENTS

This work was supported by the Shriners of North America and US Public Health Service Grant AI12936.

18.

19. 20.

REFERENCES 21. 1. Meade

2. 3. 4.

5.

6.

7.

CJ, Mertin J: Fatty acids and immunity. Adv Lipid Res 16:127-165, 1978 Gurr MI: The role of lipids in the regulation of the immune system. Prog Lipid Res 22:257-258, 1983 Johnston PV, Marshall LA: Dietary fat, prostaglandins and the immune response. Prog Food Nutr Sci 8:3-35, 1984 Johnston PV: Dietary fat, eicosanoids and immunity. Adv Lipid Res 21:103-141, 1985 Johnston PV: Lipid modulation of immune responses. IN Contemporary Issues in Clinical Nutrition II. Nutrition and Immunology, Chandra RK (ed). Alan R. Liss Inc, New York, 1988, pp 37-86 DeWille JW, Fraker PJ, Romsos DR: Effects of essential fatty acid deficiency and various levels of dietary polyunsaturated fatty acids on humoral immunity in mice. J Nutr 109:1018-1027, 1979 DeWille JW, Fraker PJ, Romsos DR: Effects of dietary fatty acids on delayed type hypersensitivity in mice. J Nutr 111:2039-2043,

22.

619, 1988 23. Webb DR, Weider KJ, Nowowiejski I: Prostaglandins in lymphocyte suppressor mechanisms. Adv Immunopharmacol 2:383-395, 1980 24. Bankhurst AD: The modulation of human killer cells activity by prostaglandins. J Clin Lab Immunol 7:85-91, 1982 25. Goodwin JS, Ceuppens J: Regulation of the immune response by prostaglandins. J Clin Immunol 3:295-310, 1983 26. Herman J, Rabsion AR: Prostaglandin E 2 depresses natural cytotoxicity by inhibiting interleukin 1 production by large granular

27. 28.

1981 8. Thomas 9.

10. 11.

12.

13.

14. 15.

IK, Erickson KL: Dietary fatty acid modulation of murine T-cell responses in vivo. J Nutr 115:1528-1534, 1985 Marshall LA, Johnston PV: The influence of dietary essential fatty acids on the rat immunocompetent cell, prostaglandin synthesis and mitogen-induced blastogenesis. J Nutr 115:1572-1580, 1985 Erickson KL: Dietary fat influences on murine melanoma growth and lymphocyte mediated cytotoxicity. JNCI 72:115-120, 1985 Erickson KL, Thomas IK: Susceptibility of mammary tumor cells to complement mediated cytotoxicity after in vitro or in vivo fatty acid manipulation. JNCI 75:333-340, 1985 Leitch AG, Lee TH, Ringel EW, et al: Immunologically induced generation of tetraene and pentaene leukotrienes in the peritoneal cavities of menhaden-fed rats. J Immunol 132:2559-2565, 1984 Lefkowitch JB: Essential fatty acid deficiency inhibits the in vivo generation of leukotriene B4 and suppresses levels of resident and elicited leucocytes in acute inflammation. J Immunol 140:228-233, 1988 Perez RV, Babcock GF, Alexander JW: Altered macrophage function in dietary immunoregulation. JPEN 13:1-4, 1988 Karmali RA: Eicosanoids and cancer. Prog Clin Biol Res 222:687697, 1986

Ferretti A, Izui S, Strom TB: A fish oil diet rich in eicosapentaenoic acid reduced cycloxygenase metabolites and suppresses lupus in MRL-1pr mice. J Immunol 134:1914-1919, 1985 17. Morrow WJW, Ohashi Y, Hall J, et al: Dietary fat and immune 16.

Kelley VE,

function. 1. Antibody responses, lymphocyte function in (NZB X NZW)F1 mice. J Immunol 135:3857-3863, 1985 Cook A, Wise WC. Callihan CS: Resistance of essential fatty aciddeficient rats to endotoxic shock. Circ Shock 6:333-342, 1979 Alexander JW. Saito H, Trocki O, Ogle C: The importance of lipid type in the diet after burn injury. Ann Surg 204:1-8, 1986 Mertin J: Effect of polyunsaturated fatty acids on skin allograft survival and primary and secondary cytotoxic response in mice. Transplantation 21:1-4, 1976 Foegh ML, Alijani MR, Helfrich GB, et al: Fatty acids and eicosanoids in organ transplantation. Prog Lipid Res 25:567-572, 1986 Perez RV, Munda R, Alexander JW: Dietary immunoregulation of transfusion induced immunosuppression. Transplantation 45:614-

29.

30.

31.

lymphocytes. Clin Exp Immunol 57:380-384, 1984 Rola-Pleszczynski M: Immunoregulation by leukotrienes and other lipoxygenase metabolites. Immunol Today 6:302-307, 1985 Zurier RB: Prostaglandins, immune responses, and murine lupus. Arthritis Rheum 25:804-809, 1982 Wan JM-F, Teo T, Babayan V, Blackburn GL: Lipids and the development of immune dysfunction and infection. JPEN 12:43S47S, 1988 Saito H, Trocki O, Heyd T, et al: Effect of dietary unsaturated fatty acids and indomethacin on metabolism and survival after burn. Abstr American Burn Assoc 17:27(Abstr), 1985 Sanders TAB, Younger K: The effect of dietary supplements of ω3 polyunsaturated fatty acids on the fatty acid composition of platelets and plasma choline phosphoglycerides. Br J Nutr 45:613616, 1981

32.

Simopowlos AP, Kifer RR, Martin RE (eds): Health Effects of Polyunsaturated Fatty Acids in Sea Foods. Academic Press, New York,1986

33. Wan J, Grimble RF: Effect of dietary linoleate content on the metabolic response of Escherichia coli endotoxin. Clin Sci 72:383-

385, 1987 34.

Clouva-Molyvdas P, Peck MD, Alexander JW: Short-term dietary protein manipulation does not affect survival from intraperitoneal

Pseudomonas infection in mice. JPEN 14:366-370, 1990 35. Peck MD, Alexander JW, Ogle CK, et al: The effect of dietary fatty acids on response to Pseudomonas infection in burned mice. J Trauma 1990, 30:445-452, 1990 36. Morita I, Takahashi R, Saito R, et al: Effects of eicosapentaenoic acid on arachidonic acid metabolism in cultured vascular cells and platelet-species difference. Thromb Res 31:211-217, 1983

Downloaded from pen.sagepub.com at UQ Library on March 13, 2015