Role of arachidonic acid in the regulation of the inflammatory response in (TNF-alpha-treated) rats
Ling, Pei-Ra
ABSTRACT. Background: This study examined whether adding arachidonic acid (AA) to a fish oil diet would alter certain of the anti-inflammatory effects of fish oil in response to tumor necrosis factor (TNF) infusion in rats. Methods: AA was given at 0.08 wt% of diet for 6 weeks. The total fat in each diet provided 20% of dietary energy. Four groups were pair-fed sunflower oil (S), S+AA, fish oil (F), or F+AA for 6 weeks. At the end of feeding, each animal received TNF-alpha (20 [mu]g/kg) infusion for 3 hours. After 1 hour of TNF infusion, a euglycemic and hyperinsulinemic clamp (10 mU/min per kilogram of insulin) was used to determine the actions of insulin. The insulin-stimulated glucose utilization in liver, muscle, and fat was determined by using ^sup 14^C-deoxyglucose. The plasma glucose, insulin, and corticosterone levels were determined at basal, 60 minutes, and the end of the experiment (180
minutes). The fatty acid composition of plasma phospholipids also was determined. Results: Fish oil significantly increased omega-3 fatty acids in phospholipids in both F and F+AA and decreased AA in F, compared with S. AA significantly restored the level of AA and reduced the increase of omega-3 fatty acids in phospholipids in F+AA compared with F, but had no impact on fatty acid composition when added to S. Corticosterone level was significantly lower with fish oil feeding but higher in both F and S containing AA compared with F and S, respectively. The highest glucose uptake in tissues was in F, followed by F+AA, and then S and S+AA. Conclusions: These results suggest that fish oil is anti-inflammatory principally through a reduction in the AA content of phospholipids. (Journal of Parenteral and Enteral Nutrition 22:268275, 1998)
Arachidonic acid (AA, 20:4omega-6) is one of the most abundant polyunsaturated fatty acids in membrane phospholipids. AA content in membrane is correlated with important physiologic functions, such as growth,1 the regulation of gene expression,2-5 insulin sensitivity,6 and growth hormone action.7
Linoleic acid (LA) is the precursor for the synthesis of AA; thus when LA is deficient, membrane AA level is reduced and an abnormal fatty acid, Mead acid (20:3omega– 9), is produced. Consequent to this change, there is a marked reduction in the inflammatory response to endotoxin.s Fish oil feeding also can produce a decrease in AA content, because eicosapentaenoic acid (EPA, 20:5omega-3) and docosahexaenoic acid (DHA, 22:6omega-3), which are enriched in fish oil, are preferentially and rapidly incorporated into membrane phospholipids to displace AA.
EPA is structurally similar to AA and competes with AA in the cyclo-oxygenase pathway, which leads to synthesis of different series of prostaglandins, thromboxanes, and prostacyclins9,10 that have substantially less immune suppression, and fewer proinflammatory and hypotensive effects.11 It is through this mechanism that fish oil assumes potential value in therapy for modulation of protein catabolism, immune depression, and certain consequences of insulin resistance that develop in response to infection or inflammation. Mechanisms proposed for the effects of an increase in dietary omega-3 fatty acids include changes in tissue signaling responsiveness, eicosanoid and other second messenger production, and responsiveness to eicosanoid actions.11-13 However, on the basis of the evidence that dietary AA has effects that are opposed to those observed with omega-3 fatty acids, the present study was designed to explore the hypothesis that the reduced availability of AA induced by fish oil feeding may play a predominant role in modulation of inflammatory responses to injury or infection. Because the production of AA from linoleic acid is closely regulated by two desaturase steps, little change in tissue membrane AA occurs despite major changes in dietary linoleic acid content. However, AA, which is a limited component of most diets, because it is principally found in membrane phospholipids of animal tissues, might be expected to bypass the regulatory steps when added in large amounts to the diet.
A highly enriched AA source (ARASCO oil; Martek Biosciences Corporation, Columbia, MD) was used for the addition of AA in the diet. The amount of AA was given at 0.08 wt% of diet for 6 weeks. The study was conducted in an acute-stress condition induced by the infusion of tumor necrosis factor alpha (TNF-alpha) for 3 hours in rats. The reason for use of this model is that our previous studies demonstrated that infusion of TNF-alpha can produce a state of insulin resistance as one important component of the systemic inflammatory response (SIR).14 In the same model, prefeeding with fish oil (30% of total dietary energy from fat) significantly improved glucose uptake in abdominis muscle during a hyperinsulinemic and euglycemic condition, indicating that fish oil feeding could benefit insulin-stimulated glucose utilization.15 Therefore the effects of the addition of AA to fish oil could be compared with fish oil without AA supplementation in response to TNF-alpha infusion. In addition, sunflower oil was chosen as an additional control to determine the effects of dietary AA on plasma phospholipids when dietary omega-6 polyunsaturated fatty acids were sufficient to optimize tissue AA levels. In this study, we evaluated the effects of additional dietary AA on plasma phospholipid composition and changes in plasma corticosterone concentration and glucose metabolism as indicators of the SIR in response to TNF-alpha infusion.
Animal Preparation
Male Sprague-Dawley rats (weight, 45 to 50 g; Taconic Farms, Germantown, NY) were acclimatized in individual cages in a light-controlled room (12 hours on/12 hours off) at a temperature of 22 deg C to 24 deg C for 4 days. During this period, animals were fed a regular rat chow diet and had access to tap water ad libitum. Animals then were divided randomly into four groups to be pair-fed with (1) sunflower oil diet (S; n = 9), (2) sunflower oil plus arachidonic acid diet (AA) (S+AA; n = 8), (3) fish oil diet (F; n = 9), and (4) fish oil plus AA diet (F+AA; n = 7) for 6 weeks (Tables I and II) to be certain that fatty acid compositions of membrane were altered to an extent that pathophysiologic alterations could be identified, if present. The AIN-93G diet (Dyets, Inc, Bethlehem, PA) was the base diet. The total fat content of each diet was 10%, which provided 20% of dietary energy. AA supplementation was 20% of total dietary fat and about 0.08 wt% of the diet in each AA-supplemented diet. The sunflower oil diet (S) contained 100% of sunflower oil with 69.5% of linoleic acid. The S+AA contained 80% of sunflower oil and 20% of refined, bleached, deodorized (RBD)ARASCO oil with 40.6% of the fatty acids as AA in triglyceride form (Martek Bioscience Corp). The other major fatty acid components of RBD-ARASCO oil are 4.7% 16:0; 6.6% 18:0; 32.8% 18:1omega-9; 4.9% 18:2omega-6; 0.1% 18:3omega-6; 0.8% 20:0; 1.8% 22:0; and 1.3% 24:0. This oil is refined, bleached, and deodorized with antioxidants and contains 0.025% tocopherol and 0.025% ascorbyl palmitate. The fish oil diet contained 100% menhaden oil with 1.5% linoleic acid, 15.5% EPA, and 9.1% DHA. The F+AA contained 80% of menhaden oil and 20% of RBDARASCO oil. These four diets were isonitrogenous and isocaloric (Dyets, Inc). During the feeding period, the body weights of the animals were recorded three times weekly to ensure similar weight gain in all groups.
At the end of a 6-week feeding period, all animals underwent a surgical procedure for catheter placement under ether inhalation. A polyethylene catheter (PE 50, 0.011 x 0.024; Becton Dickinson, Parsippany, NJ) was placed into the right carotid artery for sampling of blood. Two other silicone elastomer catheters were placed in both jugular veins for administration of infusions of tracer, TNF-alpha, insulin, and glucose. The catheters were externalized in the midscapular region and attached to a flow-through swivel to permit free movement of the animals. After surgery, animals were allowed to recover in individual cages for 2 days and continuously consumed the same diet. Before the experiment, all animals were fasted overnight.
The experiment was approved by the Animal Care Committee of Beth Israel Deaconess Medical Center, which follows the guidelines established for the care and use of laboratory animals of the Institute of Laboratory Animal Resources of the National Research Council.
Experimental Design
On the morning of study, basal fasting arterial blood samples were drawn for determinations of plasma glucose, insulin, and corticosterone concentrations. Then, all animals in each dietary group received 20 [mu]g/kg recombinant murine TNF-alpha containing c200 pg endotoxin/ng protein (Genentech, San Francisco, CA). Half of the dose (10 [mu]g/kg) of TNF-alpha was administered by IV bolus, and the remainder (10 [mu]g/kg) was infused constantly over 3 hours. TNF-alpha was freshly prepared on the day of the experiment by mixing with 0.1% human albumin in saline. By the end of the first hour of infusion (60 minutes), blood was drawn again for determinations of plasma glucose, insulin, and corticosterone concentrations. Then a modification of the hyperinsulinemic-euglycemic clamp technique was used. Insulin was infused continuously at 10 mU/min per kilogram for 2 hours through one jugular catheter. After 1 minute of insulin infusion, 20% dextrose (Astra Pharmaceutical Products, Westborough, MA) was infused at a variable rate through another jugular catheter using a variable syringe infusion pump (Harvard Apparatus, South Natick, MA). Arterial blood was sampled every 10 minutes for measurements of glucose concentration. The rate of glucose infusion was adjusted empirically after each arterial plasma glucose determination to maintain a glucose level at ~100 mg/dL as a euglycemic condition. At 138 minutes of the initial infusion of TNF-alpha, a bolus of 5 [mu]Ci [^sup 14^C-2]deoxyglucose (^sup 14^C-DG) was injected IV; then 100 liL arterial blood samples were drawn at 140, 150, 160, 170, and 180 minutes for determinations of the decay of 14C-DG in plasma. After each sampling, an equal volume of saline was injected to replace the blood volume.
At the end of the infusions (180 minutes of total infusion time or 120 minutes during the hyperinsulinemiceuglycemic clamp), a blood sample was collected for determinations of plasma concentrations of glucose, insulin, and corticosterone and phospholipid fatty acid profiles, and then 5 mg pentobarbital sodium was IV injected to euthanize the animals. Pieces of rectus abdominis muscle, gastrocnemius muscle, liver, and abdominal mesenteric adipose tissues were removed and weighed for ^sup 14^C-DG measurement. All samples were stored at -20 deg C until analysis.
Analytical Procedures
Plasma total lipids were extracted with 2:1 chloroform/methanol.15 Internal standard (C13:0;
Sigma Chemical Co, St Louis, MO) was added to each sample before lipid extraction. Phospholipid was separated by thin-layer chromatography (TLC) plates using a mobile phase of petroleum ether, diethyl ether, and glacial acetic acid (80:20:1 vol/vol) and was identified relative to the migration of standards using dichlorofluorescein spray. The phospholipids were isolated from the plates and were hydrolyzed and methylated under nitrogen with 14% BF^sub 3^ in methanol for 45 minutes in a steam bath with a closed system. After methylation, the fatty acid methyl esters were resuspended in petroleum ether and analyzed by gas chromatography (5890 Series II; Hewlett-Packard, Palo Alto, CA) using a 50-m fused silica capillary column containing SP-2330 as the stationary phase with a 0.20-[mu]m film thickness (Supelco, Bellefonte, PA). The relative mole percent of individual fatty acids were identified and quantified using Chem Station software (Hewlett Packard), based on the relative responses of an internal standard of pure C13:0 methyl esters (NuChek Prep, Elusia, MN).
Plasma glucose was determined by the glucose oxidase method using a Beckman glucose analyzer II (Beckman, Brea, CA). Plasma ^sup 14^C-DG level and accumulation of ^sup 14^C-DG in tissues were determined using a method described elsewhere.-15-7 In brief, blood samples were deproteinized with Ba(OH)^sub 2^-ZnSO^sub 4^ and immediately centrifuged. An aliquot of the supernatant was used for counting ^sup 14^C-DG radioactivity. Tissue samples were placed in NaOH, digested at 60 deg C for 1 hour, and neutralized with HCl. The digested tissues were treated with HClO^sub 4^ or Ba(OH)-ZnSO^sub 4^ and centrifuged, and the supernatants were counted for ^sup 14^C-DG plus ^sup 14^C-DG-6-phosphate and ^sup 14^C-DG-6-phosphate, respectively.
Plasma insulin (Binax, South Portland, ME) and corticosterone (Nichols Institute, San Juan Capistrano, CA) were determined by radioimmunoassay using commercial kits.
Calculations
The apparent rate of glucose uptake in tissue was calculated on the basis of the accumulation of ^sup 14^CDG-phosphate in the tissue and the integrated ^sup 14^C-DG to glucose ratio in plasma during the 42 minutes after ^sup 14^C-DG injection, as described previously.14,15 The lumped constant in different tissues was assumed to be 1.0.
Statistical Analysis
Data are presented as the means +/- SEM. Group means were compared by two-way analysis of variance (ANOVA) using the SYSTAT statistical software package for Windows (Version 5, 1992; SYSTAT, Inc, Evanston, IL). Comparisons among groups were made by Fisher’s least significant difference test for the changes in plasma phospholipid fatty acids, glucose, insulin, corticosterone, and insulin-stimulated glucose uptake in tissues when ANOVA was found to be >=95% confidence level. The effects of TNF-alpha infusion on plasma concentrations of glucose and corticosterone were compared by one-way ANOVA at different time points in each dietary groups. The significance also was defined by a p value equal to or greater than the 95% confidence level.
RESULTS
The initial body weight of the animals was not significantly different among groups. After 6 weeks of pairfeeding with four different diets, the final body weights also were not significantly different among groups.
Tables III and IV list the effects of different feeding diets on plasma phospholipid composition. The results showed that plasma fatty acid compositions closely reflected the differences in fatty acid contents of the various diets. Fish oil contains higher percentages of 16:0, 16:1omega-7, 20:5omega-3, 22:5omega-3 and 22:6omega-3 and lower percentages of 18:2omega-6 and 20:4omega-6 than sunflower oil. After 6 weeks of feeding, fish oil-based diets (F and F+AA) significantly increased the percentage of 16:0, 16:lomega-7, 18:1omega-9, 20:3omega-6, 20:5omega-3, 22:5omega-3, and 22:6omega-3 fatty acids and significantly decreased the percentage of 18:2omega-6, 20:2omega-6, and 20:4omega-6 fatty acids. As a result, the total content of omega-6 fatty acids was significantly lower, whereas the total content of omega-3 fatty acids was significantly higher in both F and F+AA groups than in S and S+AA groups. However, the total contents of saturated and polyunsaturated fatty acids in plasma phospholipids were not significantly different among the four different feeding groups.
AA in the diet significantly decreased the content of 18:2omega-6 and increased the content of 18:3omega-3 in plasma phospholipids in both S and F diets. On the other hand, AA had no impact on AA content in the S+AA group compared with the S group. In contrast, AA significantly increased and restored the level of 20:4omega-6 in the F+AA group to that observed in S group, but reduced the increases of 18:1omega-9, 20:3omega-6, 20:5omega-3 (EPA), 22:5omega-3 (DPA), and 22:6omega-3 (DHA) compared with the F group. Therefore, the ratio of EPA to AA, DPA to AA, DHA to AA, and total group of co-3 fatty acids to AA was significantly lower in the F+AA group compared with the F group (Table V). However, the omega-3 content of the F+AA group remained significantly higher than that in the S or S+AA group.
Table VI lists the changes in plasma levels of glucose. Basal plasma glucose and insulin concentrations were not significantly different among groups. After 1 hour of TNF-alpha infusion (60 minutes), plasma glucose concentrations were significantly higher than the basal levels in all groups, except in the F group. Further, the plasma glucose concentrations were significantly higher in the S+AA and F+AA group than those in the S and F groups, respectively. However, the insulin level was maintained. During the insulin clamp (from 60 to 180 minutes), plasma insulin levels were increased by insulin infusion, and plasma glucose in each group was monitored and controlled at the level of 100 mg/dL.
Table VII lists changes in plasma corticosterone levels in different groups. Similar to the changes in the plasma glucose concentrations, TNF-alpha infusion for 1 hour (60 minutes) also significantly increased the levels of plasma corticosterone in all groups, The increase was slightly but not significantly less in the F group than in the other three groups. However, larger increases in corticosterone concentrations were observed in both S and S+AA groups than in both F and FA groups. AA in the diet also significantly increased the plasma corticosterone levels in both S and F groups. At the end of the clamp (180 minutes), the corticosterone concentrations were reduced but still were significantly higher than basal levels, and there were no significant differences among groups at this time point.
Table VIII lists the data that reflect the effects of TNF-alpha on insulin action in different feeding groups under euglycemic and hyperinsulinemic conditions. The glucose infusion rate necessary to maintain plasma glucose levels at similar hyperinsulinemic condition was not significantly different among groups. However, higher rates of ^sup 14^C-DG uptake reflecting increased insulin sensitivity were found in fish oilcontaining groups in the liver and fat tissues than in the sunflower oil-containing groups. When the comparison was made between the F and F+AA groups, animals fed the F+AA diet had a lower rate of ^sup 14^C-DG uptake in both liver and fat. However, there was no difference in the rates between the S and S+AA groups. There also were trends toward an increase in the rates of ^sup 14^C-DG uptake in abdominis muscle and gastrocnemius muscle in fish oil-containing groups, but these changes did not reach statistical significance.
DISCUSSION
The results of this study demonstrated that AA levels in plasma phospholipids can be regulated by both the dietary fish oil and dietary fish oil + AA intakes, but are unchanged by AA supplementation in a vegetable oil diet. In this study, sunflower oil contained 69.6% of LA but no AA. After 6 weeks of feeding, however, the AA content in plasma phospholipids was 25.5% of the total fat, indicating that the dietary LA, as the precursor for AA, was sufficiently desaturated (via Delta6 desaturase and Delta5 desaturase) and elongated to AA, and the AA was incorporated into phospholipids. The process of de novo synthesis of AA mainly occurs in the liver. Providing AA supplementation to vegetable oil did not further influence AA levels, although one might propose that dietary AA was incorporated into phospholipids with a compensating reduction in the endogenous contribution. Fish oil contained only 1.5% LA but 2.4% AA in total fat. As a result of the fish oil diet, the level of AA in plasma phospholipids in the F group was significantly lower, at only half of the level in the S group. However, when AA was added to F (F+AA), dramatic effects on plasma phospholipid fatty acid composition were found. Even though the additional dietary AA was a small amount, it was substantial, -8% of fatty acids, in relation to normal intakes. Consequently, the plasma AA level was enhanced significantly compared with F group, suggesting that the dietary AA can be incorporated readily and effectively into plasma phospholipids to restore the AA level in phospholipids to normal levels but not beyond. This effect suggests that dietary AA can overcome the block in AA production or incorporation into phospholipids induced by fish oil, or more specifically by EPA.
It is important to comment further on the observation that the amount of AA in plasma phospholipid fatty acids and presumably tissue membranes is tightly regulated. Whelan et al18 reported that a doubling of LA intake resulted in only a slight and insignificant increase in tissue AA levels in both liver and peritoneal exudate cells in mice. They concluded that the conversion of dietary LA to tissue AA was saturable. Consistent with their findings, the AA content in plasma phospholipids in this study was maintained at a similar level in the S and S+AA groups, although the S diet had 20% more LA than the S+AA diet. Because AA is an inhibitor of both the Delta5 desaturase and Delta6 desaturase,19,20 it is reasonable to speculate that dietary AA has a feedback mechanism to control the AA content in phospholipid pool when a certain level of AA has been achieved. Further conversion of dietary LA to AA is presumably inhibited. However, the AA can be from either de novo synthesis or direct dietary supply. In the F+AA group, the dietary AA only restored the AA content to the level seen in the S group. Additional dietary AA did not further increase the AA level in plasma phospholipids in the S+AA group compared with the S group. Therefore, our results further suggest that the process of the incorporation of dietary AA into phospholipids is saturable.
Basal plasma glucose and insulin concentrations were not different among the groups. In this study, all the diets were isocaloric and isonitrogenous with 20% of dietary calories from fat. In the dietary fat source, the differences were the types of fatty acids (omega-3 vs omega-6) and the different amounts of dietary AA. Because all animals were pair-fed for 6 weeks, the nutritional conditions were well controlled. Therefore, these data demonstrated that dietary intake of fish oil and dietary AA had no effect on circulating glucose and insulin levels in a normal condition.
However, after TNF-alpha infusion, the metabolic responses were altered significantly with different dietary feedings. At 60 minutes, the plasma glucose concentration was increased significantly in all groups except the F group. Because plasma insulin levels were not different among groups at 60 minutes, the increased glucose levels in the S, S+AA, and F+AA groups indicate that a state of insulin resistance had developed during TNF infusion. Because insulin affects glycogenolysis and gluconeogenesis and glucose utilization, the changes in glucose and insulin concentrations may reflect any one of the sum of these alterations in insulin action.
Prefeeding with fish oil increased insulin sensitivity in response to TNF-alpha infusion. After 3 hours of TNF-alpha infusion, the highest rate of ^sup 14^C-DG uptake also was found in the F group under hyperinsulinemiceuglycemic conditions. These results are consistent with the results from other studies21-23 and our previous report15,24 that fish oil feeding improves the insulin-stimulated glucose utilization in peripheral tissues after cytokine administration. Because feeding of fish oil resulted in a significant increase in omega-3 fatty acids with a concomitant decrease in omega-6 fatty acids in plasma phospholipids, it is possible that the omega-3 fatty acids act as competitive inhibitors of the cyclo-oxygenase and lipoxygenase enzymes, which are then converted to biologically less active prostaglandins of the 3 series, and leukotrienes of the 5 series.25-30 As a consequence, the insulin resistance engendered by TNF-alpha infusion is reduced. Studies have demonstrated that incorporation of omega-3 fatty acids into the membrane after 6 weeks of fish oil feeding can reduce PGE^sub 2^ production.31 This explanation also is supported by the evidence that the changes in plasma glucose concentration and insulin-stimulated glucose uptake were correlated well with the results of plasma corticosterone levels in the circulation. The fish oil-fed group had the lowest level of corticosterone in plasma. These results suggest that the alterations in glucose metabolism and insulin action are explained by the antiinflammatory effects of fish oil, because the levels of stress or counterregulatory hormones of the SIR, such as cortisone, glucagon, and catecholamines, principally impair insulin-mediated glucose uptake.32-34
The present results also showed that adding AA to the S and F diets elevated the plasma levels of glucose and corticosterone and increased insulin resistance, indicating an enhancement of inflammatory response. Nevertheless, it is not possible to state that AA increased the inflammatory response within diet groups. The rates of ^sup 14^DG-deoxyglucose in tissues tended to be lower in AA groups. However, significant differences were only observed between the F and F+AA groups, especially in the liver and fat tissues where, despite normalization of phospholipid AA levels, complete reversal of insulin resistance was not seen. All these findings demonstrate that dietary AA countered the metabolic alterations in response to TNF-ca infusion but were most effective when added to the F diet. The benefits of fish oil feeding were diminished by adding AA to the diet. Although omega-3 fatty acids were lower in the F+AA diet, they were still elevated compared with the S and S+AA diets, making the changes in AA the most likely explanation for the changes seen. Taken together, these observations suggest that there is a critical link between dietary AA intake and the inflammatory response when animals are in a stress condition, as induced by TNF-alpha infusion in this study. In addition, the present results also suggest that AA is proinflammatory, particularly when given with fish oil, and that the reduced AA availability plays a major role in mediating the anti-inflammatory effects of fish oil.
It is worth noting that the AA content in plasma phospholipids was significantly lower in the F group but similar among the S, S+AA, and F+AA groups. Further, the content of AA in plasma phospholipids appears to track with the alterations of plasma corticosterone concentrations in response to TNF-alpha infusion. The lowest value at 60 minutes was found in the F group, with similar values in the F+AA and S groups and the highest values in the S+AA group, although the latter difference was not statistically significant. The mechanisms for these changes are not clear. It has been established that free AA, which is liberated from membrane phospholipids, can be oxidized enzymatically to a variety of eicosanoids, many of which are proinflammatory and contribute to the pathologies of many diseases. Dietary AA also is a poor substrate for beta-oxidation, which may favor triglyceride synthesis.35,36 Furthermore, AA is an important second messenger in its own right.37 Thus it is possible that the AA content in the serum free fatty acid pool, which is mostly contributed by dietary fatty acids, may be important in regulating the whole body response to TNF-alpha. However, further studies are required to determine more fully the relationship between dietary AA, AA in circulating triglyceride levels, AA in membrane phospholipids, and the metabolic responses during TNF-alpha infusion. A third possibility for altering cytokine and homonal responsiveness has been related to the alteration in membrane fluidity by changing their polyunsaturated fatty acid content.6 At least grossly, this was not found because polyunsaturated and saturated fatty acids were similar among groups. Further, the total number of double bonds in the F and F+AA groups were similar despite the greatest phospholipid differences between these two groups.
In summary, the results of this study add to the existing evidence that AA content in phospholipids, which is the most abundant component in the membrane, is tightly regulated but can be altered by dietary fish oil and AA plus fish oil. There is limited evidence that the dietary intake of fish oil or modest intakes of AA have substantial metabolic effects, in terms of plasma levels of glucose, insulin, and corticosterone, under normal conditions. During an SIR induced by TNF-alpha infusion, fish oil and AA supplementation may be useful tools to probe the role of second messengers in the SIR. Further studies will be needed to determine if dietary AA supplementation of vegetable oil diets can intensify the SIR.
ACKNOWLEDGMENT
The authors thank Martek Biosciences Corporation, Columbia, Maryland, for the gift of oils, which made the study possible. This work was supported in part by Grants DK 45750 and DK 50411 awarded by the National Institutes of Health and by Ross Products Division, Abbott Laboratories, Columbus, Ohio.
Received for publication, December 23, 1997.
Accepted for publication, April 3,1998.
REFERENCES
1. Carlson SE, Werkman SH, Peeples JM, et al: Arachidonic acid status correlates with first year growth in preterm infants. Proc Natl Acad Sci USA 90:1073-1077, 1993
2. Jurivich DA, Sistonen L, Sarge KD, et al: Arachidonate is a potent modulator of human heat shock gene transcription. Proc Natl Acad Sci USA 91:2280-2284, 1994
3. Clarke SD, Jump DB: Dietary polyunsaturated fatty acid regulation of gene transcription. Annu Rev Nutr 14:83-98, 1994
4. Thams P, Hedeskov CJ, Capito K: Exogenous arachidonic acid inactivates protein kinase C in mouse pancreatic islets. Acta Physiol Scand 149:227-235, 1993
5. Stuhlmeier KM, Kao JJ, Bach FH: Arachidonic acid influences proinflammatory gene induction by stabilizing the inhibitorKBa/nuclear factor-KB (NF-KB) complex, thus suppressing the nuclear translocation of NF-KB. J Biol. Chem. 272:24679-24683, 1997
6. Borkman M, Storlien LH, Pan DA, et al: The relation between insulin sensitivity and the fatty acid composition of skeletal muscle phospholipids. N Engl J Med 328:238-244, 1993
7. Nakamura MT, Phinney SD, Tang AB, et al: Increased hepatic Delta6 desaturase activity and differential alteration of liver and adipose fatty acid composition with growth hormone expression in the MG 101 transgenic mouse. Lipids 31:139-143, 1996
8. Cook JA, Wise WC, Knapp DR, et al: Sensitization of essential fatty acid-deficient rats to endotoxin by arachidonate pretreatment: Role of thromboxane A2. Circ Shock 8:69-76, 1981
10. Curtis-Prior M: Prostaglandins: Biology and Chemistry of Prostaglandins and Related Eicosanoids. New York: Churchill-Livingston, 1988.
10. Parker CW: Lipid mediators produced through the lipoxygenase pathway. Annu Rev Immunol 5:65-84, 1987
11. Kinsella JE, Lokesh B, Broughton S, et al: Dietary polyunsaturated fatty acids and eicosanoids: Potential effects on the modulation of inflammatory and immune cells: An overview. Nutrition 6:24-44, 1990
12. Lands WEM: Biochemistry and physiology of to-3 fatty acids.
FASEB J 6:2530-2536, 1992
13. Terano T, Salmon JA, Higgs GA, et al: Eicosapentaenoic acid as a modulator of inflammation. Effect on prostaglandin and leukotriene synthesis. Biochem Pharmacol 35:779-785, 1986 14. Ling PR, Bistrian BR, Mendez B, et al: Effects of systemic infusions of endotoxin, tumor necrosis factor and interleukin-1 on glucose metabolism in the rat: Relationship to endogenous glucose production and peripheral tissue glucose. Metabolism 43:279-284, 1994 15. Sierra P, Ling PR, Istfan NW, et al: Fish oil feeding improves muscle glucose uptake in tumor necrosis factor-treated rats. Metabolism 44:1365-1370, 1995
16. Hom FG, Goodner CJ, Berrie MA: A (3H)-2-deoxyglucose method for comparing rates of glucose metabolism and insulin responses among rat tissues in vivo. Validation of the model and the absence of an insulin effect on brain. Diabetes 33:141-152, 1984
17. Ferre P, Leturgue A, Burnol AF, et al: A method to quantify glucose utilization in vivo in skeletal muscle and white adipose tissue of the anaesthetized rat. Biochem J 228:103-110, 1985
18. Whelan J, Broughton KS, Surette ME, et al: Dietray arachidonic and linoleic acids: Comparative effects on tissue lipids. Lipids 27:85-88, 1992
19. Brenner RR, Peluffo RO, Nervi AM, et al: Competitive effect of alpha- and gamma-linolenyl-CoA in linoleyl-CoA desaturation to gamma-linolenyl-CoA. Biochim Biophys Acta 176:420-422, 1969 20. Leikin AI, Brenner RR: Microsomal delta 5 desaturation of eicosa-8,11,14-trienoic acid is activated by a cytosolic fraction. Lipids 24:101-104, 1989
21. Waldhausl W, Ratheiser K, Komjati M, et al: Increase of insulin sensitivity and improvement of intravenous glucose tolerance by fish oil in healthy man. IN Health Effects of Fish and Fish Oils, Chandra RK (ed). ARTS Biomedical Publishers & Distributors, St. John’s, Newfoundland, 1989, pp 171-187 22. Storlien LH, Kraegen EW, Chisholm DJ, et al: Fish oil prevents insulin resistance induced by high-fat feeding in rats. Science 237:885-887, 1987
23. Ginsberg BH, Brown TJ, Simon I, et al: Effect of the membrane lipid environment on the properties of insulin receptors. Diabetes 30:773-780, 1981
24. Ling PR, Istfan N, Colon E, et al: Effect of fish oil on glucose metabolism in the interleukin-la-treated rat. Metabolism 42:81-85, 1993
25. Prescott SM: The effect of eicosapentaenoic acid on leukotriene B production by human neutrophils. J Biol Chem 259:7615-7621, 1984
26. Billiar TR, Bankey PE, Svingen BA, et al: Fatty acid intake and Kupffer cell function: Fish oil alters eicosanoid and monokine production to endotoxin stimulation. Surgery 104:343-349,1988 27. Chanmugam PS, Boudreau MD, Hwang DH: Dietary (n-3) fatty acids alter fatty acid composition and prostaglandin synthesis in rat testes. J Nutr 121:1173-1178,1991
28. Kinsella JE, Lokesh B: Dietary lipids, eicosanoids and the immune system. Crit Care Med 18:s94-113, 1990 29. Whelan J, Surette ME, Hardardottir, et al: Dietary arachidonate enhances tissue arachidonate levels and eicosanoid production in Syrian hamsters. J Nutr 123:2174-2185,1993 30. Mancuso P, Whelan J, DeMichele SJ, et al: Dietary fish oil and fish and borage oil suppress intrapulmonary proinflammatory eicosanoid biosynthesis and attenuate pulmonary neutrophil accumulation in endotoxic rats. Crit Care Med 25:1198-1206, 1997
Endres S, Ghorbani R, Kelley VE, et al: The effect of dietary supplementation with 3-w polyunsaturated fatty acids on the synthesis of interleukin-1 and tumor necrosis factor by mononuclear cells. N Engl J Med 320:265-271, 1989 Lang CH, Dobrescu C, Meszaros K: Insulin-mediated glucose uptake by individual tissue during sepsis. Metabolism 39:10961107, 1990
Evans DA, Jacobs DO, Wilmore DW: Tumor necrosis factor enhances glucose uptake by peripheral tissues. Am J Physiol 257:R1182-1189, 1989
Van de Poll T, Romijin JA, Endert E, et al: Tumor necrosis factor mimics the metabolic response to acute infection in healthy human. Am J Physiol 261:E457-465, 1991 Leyton J, Drury PJ, Crawford MA: In vivo incorporation of labeled fatty acids in rat liver lipids after oral administration. Lipids 22:553-558, 1987
Niot I, Gresti J, Biochot J, et al: Effect of dietary n-3 and n-6 polyunsaturated fatty acids on lipid-metabolizing enzymes in obese rat liver. Lipids 29:481-489, 1994
Kent JD, Sergeant S, Burns DJ, et al: Identification and regulation of protein kinase C-delta in human neutrophils. J Immunol 157:4641-4647, 1996
From the Nutrition/Infection Laboratory, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston
Correspondence: Bruce R. Bistrian, MD, PhD, Nutrition/Infection Laboratory, Beth Israel Deaconess Medical Center, West Campus, One Deaconess Road, Boston, MA 02215.
Copyright American Society for Parenteral and Enteral Nutrition Sep/Oct 1998
Provided by ProQuest Information and Learning Company. All rights Reserved