Dietary fish oil modulation of macrophagetumoricidal activity

Supplement to Nutrition'Vol. 12, No. 1, 1996

Dietary Fish Oil Modulation of Macrophage Tumoricidal Activity K E N T L. E R I C K S O N , PHD, A N D NEIL E. H U B B A R D , PHD

From the Department of CeU Biology and Human Anatomy, University of California, School of Medicine, Davis, California, USA ABSTRACT Recent studies have shown that macrophages and their functions can be altered by dietary fat. Specifically,diets that are rich in n-3 fatty acids such as fish oils can have significant effects on macrophage cytolytic capacity and the production of select cytokines. The purpose of these studies was to characterize how dietary fish oils altered macrophage tumoricidal activity and the production of tumor necrosis factor-a (TNF-0t). Dietary menhaden fish oil (MFO) significantly decreased the ability of activated macrophages to kill tumor targets compared with macrophages from mice fed safflower oil (SAF), which is high in n-6 fatty acids. Those macrophages from mice fed MFO were hyporesponsive to intefferon-y. In addition, macrophages from mice fed MFO produced more TNF-ct after 24 h activation with lipopolysaccharide compared with macrophages from mice fed SAF. That difference in TNF-ct production was associated with a differentialproduction of and response to prostaglandin ~ . Although there are several possible mechanisms by which dietary fat may alter macrophage function and cytokine production, we have investigated signal transduction. Macrophages from MFO-fed mice had a greater increase in intracellular calcium mobilization after treatment with platelet-activatingfactor (PAF) than macrophages from mice fed SAF. Those differences may be related to an alteration in the PAF signalling pathway by increasing phospholipase C activity. Thus, dietary n-3 fatty acids may significantly alter macrophage tumoricidal activation and TNF-ct production through the modulation of PGE2 production and signal transduction. Key words: fish oil, n-3 fatty acids, macrophage, tumor necrosis factor-~ signal transduction,tumor cytolysis INTRODUCTION Macrophages are a diverse population of cells capable of carrying out a number of important biologic functions, including the production of select cytokines. Those functions can also be regulated by cytokines, bacterial products, and lipid-based mediators. It has also been shown that select macrophage functions may be differentially altered by dietary fats. However, the mechanisms by which dietary fatty acids alter those functions are not known. One possible mechanism is based on dietary fat alteration of eicosanoids, 20-carbon fatty acid metabolites, such as prostaglandins (PG), leukotrienes, and hydroxy fatty acids, l Various functional capacities, including tumoricidal activity, may be downregulated by either exogenous or endogenous PGE2. 2-4 Tumorigenesis at select sites, noteably breast, colon and prostate may be influenced by multiple factors including dietary fat and immune response. For example, dietary linoleic acid (LA) has been shown to enhance the growth and metastasis of rodent breast tumors.~ Although the mechanisms for enhanced tumor growth are not understood, the production of eicosanoids from the precursor, LA, undoubtedly plays an important role. The increased

PGEz production may suppress the tumoricidal capacity of various effector cells. In contrast to vegetable oils, which contain high levels of LA, fish oils contain high levels of the n-3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Those n-3 fatty acids are thought to be antiinflammatory either by the inhibition of cyclooxygenase activity or by the production of similar mediators, but with different biologic activities from arachidonic acid (AA)-dedved eicosanoids. TNF-~ is an important cytokine involved in the onset and regulation of inflammatory responses. A primary source of TNF-ct is the macrophage, which can also secrete eicosanoids. Although the regulation of TNF-~ synthesis can occur at multiple levels, production may be inhibited by PGE2.6 Therefore, it is important to determine how diets containing n-3 fatty acids affect the production of TNF-ct and how PGE2 regulates such production. Because macrophages can be pivotal in a number of immune functions as well as the production of cytokines, we have focused on those cells. The purpose of this article was to review how diets containing n-3 fatty acids affect routine macrophage function, with a particular emphasis on tumoricidal activities including the

Correspondence to: Kent Erickson, PhD, Dept. of Cell Biology and Human Anatomy, University of California, School of Medicine, Davis, CA 95616-8643, USA.

Nutrition 12:$34-$38, 1996 ©ElsevierScienceInc. 1996 Printed in the USA. All rights reserved.

0899-9007/96/$15.00 SSDI 0899-9007(95)00072-0

FISH OIL AND MACROPHAGE TUMORICIDAL ACTIVITY cytokine TNF-ct. Next, we focus on one of several possible mechanisms by which n-3 fatty acids alter those macrophage activities. An understanding of how n-3 fatty acids may regulate macrophage cytokine production might in turn lead to insights into how dietary EPA and DHA may be used to alter tumorigenesis.

$35

Statistical Analyses Statistical analysis was performed using a two-tailed Student's t test. Differences were judged to be statistically significant when p < 0.05. The number of mice used for each experiment depended on the number of cells required and the peritoneal macrophage yield.

METHODSANDMATERIAl,S

RESULTS

Mice

Dietary Fish Oil Modulation of Macrophage TumoricidalActivity

Specific pathogen-free, 5--6-wk-old C57BL/6NCr female mice were housed in autoclaved cages in a laminar flow hood and given autoclaved water and synthetic diet ad lib. The SAF and the MFO diets were isocaloric, containing 10% fat by weight. Diet formulation and fatty acid composition of the oils used have been previously described.7.8 The mice were fed these diets for 4 weeks before use, and displayed no differences in weight gain.

To assess how dietary n-3 fatty acids compared to n-6 fatty acids, mice were fed 10 wt% of fat from MFO or SAF, and macrophage activation and tumoricidal function was examined. Macrophages were exposed to interferon-'y (IFN-¥) with or without LPS for 4 h, and cytolysis of tumor targets was assessed. Macrophages from mice fed MFO were significantly less cytolytic (62% decrease) than macrophages from mice fed SAF. It is possible that the macrophages from MFO-fed mice were defective in the actual cytolytic mechanism, or that they did not respond to the priming agent, IFN-'/, or the activating agent, LPS, as well as macrophages from mice fed SAE To distinguish between those alternatives, macrophages were primed pharmacologically with phorbol myristate acetate and calcium ionophore; such treatment has been shown to mimic IFN-¥ priming of macrophages for cytolytic function,t5 No differences in tumoricidal activity were found in either of the dietary groups even when more LPS was added. The finding that the cytolytic capacity of each diet was enhanced equally by LPS suggests that dietary manipulation did not alter LPS responsiveness. Observations that priming with greater concentrations of IFN-)" restored the partial defect in activation indicate that macrophages from mice fed MFO were relatively hyporesponsive to IFN-3,. The anti-inflammatory effects of dietary MFO may be due to altered eicosanoid production. Thus, we next examined autoregulation by macrophage-produced PGE. When stimulated with high levels of LPS (100 ng/mL), macrophages from mice fed SAF had a lower cytolytic capacity than macrophages from mice fed MFO. Tumoricidal activity may decay over time as a result of LPS stimu!ation of PGE2 production, which can be inhibited with indomethacin. 2 Macrophages from the experimental groups were exposed to LPS with or without indomethacin for 24 h, after which time cytolysis was assessed. Macrophages from mice fed the SAF diet had little cytolytic activity when cultured with LPS alone. Conversely, macrophages from the MFO diet demonstrated high levels of cytolysis with or without indomethacin. Thus, it appears

Macrophages Inflammatory peritoneal macrophages were isolated from mice injected 3 days previously with sterile fluid thioglycollate broth as previously described.8 The cells, which were >90% macrophages, were harvested, washed, resuspended, and added to tissue culture wells. After 60 rain at 37°C, the monolayers were washed free of nonadherent cells. Macrophages made up >95% of the final adherent cell population, as judged by morphology and phagocytosis. All reagents contained <0.1 ng/mL endotoxin, as determined by the suppliers or the Limulus amebocyte lysate test. Glassware was heatsterilized at 180°C for 4 h to eliminate residual endotoxin.

Assay for TNF-tx Macrophages were cultured in serum-free medium with or without bacterial lipopolysaccharide (LPS) for the times specified. An aliquot of the conditioned media was removed and tested for TNF-et activity by lysis of L929 fibroblasts in the presence of 2.5 ~tg/mL actinomycin D as previously described. 9.1° TNF-tx units were defined as the reciprocal of the dilution required to give 50% lysis of the monolayer. They were standardized against recombinant murine TNF-ct run in parallel for each experiment.

Cytolysis Assay Thioglycollate-elicited macrophages from mice fed the experimental diets were plated in 96-well dishes for these assays. Monolayers of activated macrophages were overlaid with 5~Cr-labeled P815 tumor cells to give an effector:target ratio of 10:1, and the cultures were incubated for 18 h. Percent cytolysis was determined as previously described. 8

Northern Analysis Macrophages isolated as described earlier were stimulated for various periods of time with 10 ng/mL LPS. Total RNA was isolated as previously described, jr,12run on agarose gels containing formaldehyde, and transferred to Nytran (Schleicher & Schuell, Keene, NH). Membranes were then probed with a 32p-labeled 1.37-Kb insert for murine TNF-tx cDNA or a 2.0-Kb insert for 13actin as previously described. 1°'12

300 25O 200 .~

150

Quantitation of lntracellular Calcium Concentrations Measurement of intracellular calcium was performed using the fluorescent indicator FURA-2/AM, as described elsewhere) 3 Briefly, macrophages at 5 x 106 cells/mL were incubated with 5 I.tM FURA-2/AM at 25°C in Hanks' buffered saline solution for 45 min, then washed and resuspended. After 20 s, various concentrations of PAF were added, and the FURA-2 fluorescence was monitored at excitation wavelengths of 340 and 380 nm and an emission of 510 nm. The intracellular calcium concentration was determined as previously described.t4

10o-

50 0

'~

0

.

.

.

.

I

5

'

'

'

•

I

10

.

.

.

.

r

15

.

.

.

.

i

20

.

.

.

.

i

25

FIG. I. Kinecticsof bioactive TNF-a production by macrophages from mice fed 5AF o r M F O . 43

S36 that sufficient PGE levels are produced by SAF macrophages, but not by MFO macrophages, to turn off cytolysis.

Dietary Fish Oil Alteration of TNF-ct Production " A number of different agents, including LPS, may induce the production of TNF-tx by macrophages. To assess how n-3 fatty acids may regulate TNF-~t production, mice were fed MFO- or SAF-containing diets. LPS stimulation resulted in more TNF-tx secretion after 24 h in macrophages from MFO-fed mice compared with macrophages from SAF-fed mice (Fig. 1). Both macrophage populations produced relatively large quantities of TNF-ct activity after 8 h of LPS stimulation. Although TNF-ct activity was reduced approximately 85% with time in macrophages from SAF-fed mice, TNF-ct activity in macrophages from MFO-fed mice was unchanged. The relatively high levels of TNF-ct found in MFO macrophage cultures at 24 h did not appear to be due to continuous production. When macrophages from SAF- or MFO-fed mice were treated with LPS for 8 h, then washed, and fresh LPS was added back to the cells for an additional 16 h, no additional TNF-ct was detectable. These results suggest that TNF-ct production was selflimiting and basically'complete by about 8 h after LPS exposure. The downregulation of TNF-ct production may partially be due to an alteration of transcription.6 However, there were no differences in mRNA levels for either macrophage population after LPS stimulation. The lack of TNF-ct production past 8 h and very little detectable mRNA for TNF-ct at 24 h suggest that suppression was more complex than termination of transcription.16-19 Moreover, macrophages from mice fed the SAF diet appeared to remove bioactive TNF-ct from conditioned media, whereas macrophages from mice fed MFO were unable to perform this task. Macrophages from animals fed MFO diets produced lower levels of PGE2. Thus, experiments were designed to assess whether the loss of TNF-tx activity was related to eicosanoid production. Indomethacin significantly increased TNF-tx levels after 24 h exposure to LPS in macrophages from mice fed the SAF diet. Conversely, TNF-ct production by macrophages from MFO-fed mice was not altered. These results suggest that the decrease in soluble TNF-ct levels following maximal production was mediated through a cyclooxygenase product, most likely PGE 2, and that macrophages from mice fed MFO might be deficient in this process. To examine the possibility, macrophages from both diet groups were treated with LPS, indomethacin to block endogenous PGE2 production, and various concentrations of exogenous PGE2. At 8 h, the amount of PGE2 required to inhibit TNF-~t production was similar for both macrophage groups. However, at 24 h, macrophages from mice fed SAF required one log less PGE2 than macrophages from MFO-fed mice for the suppression of TNF~. These results suggest that there may be at least two components of PGE2-mediated regulation of TNF-ct production. Thus, a concentration of PGE2 not altering the synthesis of TNF-tx at 8 h of culture appeared to induce the clearance of soluble TNF-tx. Dietary Fish Oil Upregulation of Signal Transduction Platelet-activating factor (PAF) is a lipid-based mediator which can be produced by macrophages. Not only can diet alter the concentration of PAF produced, but PAF may also stimulate a number of macrophage functions.2° An increase of intracellular Ca 2+ concentrations [(Ca2+)i] and protein kinase C (PKC) activation are early events in macrophages associated with PAF signal transduction.2~ To determine whether the signaling pathway was altered by dietary fat, we measured the PAF-stimulated increase of [Ca2+]i in macrophages from mice fed SAF- or MFOcontaining diets. PAF stimulated a biphasic increase of [Ca2+]i in both groups of macrophages in a dose-dependent manner. Macrophages from MFO-fed mice showed a 67% greater response after treatment with 25 nM PAF than did the macrophages from the SAF-fed mice. We hypothesize that the hyperresponsiveness of

FISH OIL AND MACROPHAGE TUMORICIDAL ACTIVITY macrophages from MFO-fed mice to PAF was due to the high n-3 fatty acids found in the MFO but not in the SAF diet. Therefore, the effect of n-3 and n-6 fatty acids in vitro on the PAF signaling pathway was assessed. EPA- or DHA-treated macrophages showed a greater increase in [Ca2+]i than did LA-treated macrophages. With 25 nM PAF, EPA-treated macrophages had a 55% and DHAtreated macrophages had a 51% greater transient calcium response compared to LA-treated macrophages. Collectively, these results showed that the in vitro treatment of macrophages with n-3 fatty acids most commonly found in MFO enhanced PAF signaling similarly to macrophages from mice fed a MFO-rich diet. There are several possible mechanisms for the hyperresponsiveness of macrophages from MFO-fed mice to PAF. First, this effect of MFO may be related to its ability to suppress PGE2 production. This possibility was excluded, because maximum [Ca2+]i rise occurred within 10-15 s after PAF stimulation, and at least several minutes were required after PAF stimulation for a sufficient amount of any eicosanoid metabolite to accumulate to suppress PAF action. Moreover, blocking eicosanoid metabolism by indomethacin did not alter the differences in [Ca2+]i response between macrophages of SAF- and MFO-fed mice. A second possible explanation is that the dietary n-3 fatty acids altered the PAF receptor number, affinity, or both, resulting in the hyperresponsiveness of MFO macrophages to PAF. However, the PAF receptor number and affinities were not altered by dietary n-3 fatty acids. A third possibility is that dietary n-3 fatty acids increased the signal transducing activity through the PAF receptor, without altering the receptor number and affinity. The binding of PAF to its receptor transduces the signal across the membrane through the G protein to the effector component PLC. 2~ To test this possibility, we used mastoparan, a wasp venom toxin known to activate PIP2 hydrolysis and an increase in [Ca2+]i in various cell types. 22-26 Thus, mastoparan can activate G protein-linked PLC without the involvement of G protein-linked receptor. If PLC in MFO macrophages has a higher potential, then mastoparan should cause a higher Ca2÷ mobilization in MFO macrophages than SAF macrophages. Stimulation of macrophages with a mastoparan analog, Mas7, resulted in a higher Ca2÷ response in macrophages from mice fed MFO compared to macrophages from mice fed SAF. An inactive analog of mastoparan, Masl7, did not stimulate a Ca2÷ response in either macrophage population. These results indicate that PLC activation via G protein was higher in macrophages from MFO- than SAF-fed mice. DISCUSSION In this article, we have reviewed work demonstrating that dietary fish oils containing high levels of EPA and DHA can differentially alter macrophage tumoricidal activity as well as the production of cytokines, notably TNFqx. Studies to determine the effects of dietary n-3 fatty acids on macrophage function and cytokine production have been undertaken because, first, fatty acids are known to modulate a variety of macrophage functions.27-3° Second, dietary n-3 fatty acids are known to lessen the severity of several inflammatory and immune disorders 31-36 as well as the growth of LA-sensitive tumors. 37-4° Our findings suggest that macrophages from animals fed MFO were hyporesponsive to IFN-T. However, 10-fold more IFN-T rendered those hyporesponsive macrophages equally competent for tumoricidal activity. The basis for this defect is not known, although the addition of cyclooxygenase inhibitors during activation had no effect on tumoricidal activity. In contrast, macrophages activated by LPS alone appear to be autoregulated by PGE2.2 When macrophages from mice fed SAF were activated by LPS, tumoricidal activity was low unless macrophages were treated with indomethacin. Conversely, LPS-activated macrophages from mice fed MFO had a high level of cytolysis whether or not indomethacin was added. Those macrophages do not have the

FISH OIL AND MACROPHAGE TUMORICIDAL ACTIVITY ability to autoregulate cytolytic capacity, possibly because there was no substrate adequate to produce sufficient levels of PGE 2. Any change in the regulation of macrophage activation by PGE 2 may be critical for a function such as cytokine production. Our work also suggests two possible means by which PGE 2 may regulate TNF-et expression. The termination of TNF-ct production, which appears to occur in macrophages from mice fed both SAF- and M F O - c o n t a i n i n g diets, may be due to the generation of nanomolar concentrations of PGE 2. However, the regulation of TNF-ct expression is complex, with potential control points independent of PGE2.16 Although the clearance of TNF-~ was probably mediated by PGE 2, it cannot be ruled out that some EPA or DHA may be involved. Several other papers have reported alterations in the production of cytokines by leukocytes following dietary modification with n-3 fatty acids, with differentiating results. For example, unfractionated human peripheral blood mononuclear cells taken from volunteers who supplemented their normal diets with marine lipid concentrate produced less TNF-ct in response to various signals than their presupplementation levels.41 That study assessed total immunoreactive TNF-et contained in both cell-free and cell-associated fractions combined; the significance of cell-associated TNF-ct is not clear. Moreover, it is difficult to compare studies directly using defined synthetic diets 42.43 with supplementation of undefined diets.4~ The apparent discrepancy in TNF-ct production must be assessed with caution, because macrophages from different species and anatomic sources may have different mechanisms for regulating TNF-tx production. Care must also be taken in interpreting TNF-ct concentrations determined solely by enzyme-linked immunosorbent assay, as several reports have demonstrated a discrepancy between immunoreactivity and bioactivity.44.45 Finally, other studies with dietary n-3 fatty acids have reported an increase in TNF-ct p r o d u c t i o n by L P S - s t i m u l a t e d r e s i d e n t p e r i t o n e a l macrophages. 46,47 Despite these controversies, because TNF-ct is an important cytokine involved in many aspects of inflammation and immune response, our finding that diets high in EPA and DHA may alter the longevity but not maximal production of

S37 soluble TNF4x could be significant. There are a number of possible mechanisms by which dietary n-3 fatty acids may influence cytokine production. One possible explanation is that dietary n-3 fatty acids alter signal transduction induced by IFN-T, LPS, or PAF. For example, LPS may activate phosphatidyl-inositol (PI) turnover with a subsequent rapid rise in [Ca2+]i. The action of PLC on PI leads to the generation of diacylglycerol, an important activator of PKC. It has also been shown that PAF stimulated an increase in [Ca2÷]i, IP production, and PKC activation. 2~ Because it is possible to alter the acyl composition of phospholipids, it is possible that dietary fat may alter levels of one or more enzymes involved in the signal transduction pathways. Our results showed for the first time that the dietary n-3 fatty acids enhanced the PAF signaling pathway by increasing the signal transduction through PAF receptor in macrophages. This is in contrast with previous findings in other cell types, in which n-3 fatty acids suppressed the PAF signaling pathway.4a-5° However, differences in cell types may account for this difference. The mechanism by which n-3 fatty acids alter the signal transduction capacity of the PAF receptor may involve an increase in the activation potential of PLC, resulting from changes in the phospholipid composition of the cell membrane. Because PAF induces the production of T N F - ~ 2° further work will be necessary to define the mechanism and pathophysiologic consequences of the hyperresponsiveness of n-3 fatty acid-altered macrophages to PAF and its relationship to cytokine production. In summary, diets that were rich in n-3 fatty acids such as fish oil significantly altered the ability of macrophages to kill tumor cells compared to macrophages from mice fed vegetable oil. Dietary fish oils altered the longevity but not the maximal production of soluble TNF-ct. These effects of dietary n-3 fatty acids may be mediated through an alteration in signal transduction. ACKNOWLEDGEMENT This work was supported by Grant CA47050 from the National Institutes of Health.

REFERENCES 1. Erickson KL, Hubbard NE. A possible mechanism by which dietary fat can alter tumorigenesis: lipid modulation of macrophage function. In: AICR, ed. Diet and breast cancer. New York: Plenum Press, 1994:67 2. Taffet SM, Russell SW. Macrophage-mediated tumor cell killing: regulation of expression of cytolytic activity by prostaglandin E. J Immunol 1981;126:424 3. Bonney RJ, Davies P. Possible autoregulatory functions of the secretory products of mononuclear phagocytes. Contemp Topics Immunobiol 1984;13:199 4. Heidenreich S, Gong JH, Schmidt A, Nain M, Gemsa D. Macrophage activation by granulocyte/macrophage colony-stimulating factor: priming for enhanced release of tumor necrosis factor-alpha and prostaglandin E2. J Immunol 1989;143:1198 5. Erickson KL, Hubbard NE. The effects of dietary fat on tumor growth and metastasis. In: Roche AE ed, The role o f nutrients in cancer treatment. Columbus, OH: Ross Laboratories, 1991:66 6. Kunkel SL, Spengler M, May MA, Spengler R, Larrick J, Remick D. Prostaglandin E2 regulates macrophage-derived tumor necrosis factor gene expression. J Biol Chem 1988;263:5380 7. Chapkin RS, Somers SD, Schumacher L, Erickson KL. Fatty acid composition of macrophage phospholipids in mice fed fish or borage oil. Lipids 1988;23:380 8. Somers SD, Chapkin RS, Erickson KL. Alteration of in vitro routine peritoneal macrophage function by dietary enrichment with eicosapentaenoic and docosahexaenoic acids in menhaden fish oil. Cell Immunol 1989;123:201 9. Gifford GE, Lohmann-Matthes ML. Requirement for the continual

presence of lipopolysaccharide for production of tumor necrosis factor by thioglycollate-inducedperitoneal routine macmphages. Int J Cancer 1986;38:135 10. Somers SD, Erickson KL. Regulation of murine macrophage function by IL-4. I. Activation of macrophages by a T-T cell hybridoma is due to IL-4. Cell Immunol 1989;122:178 11. Chomczynski E Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-cldoroform extraction. Anal Biochem 1987;162:156 12. Caput D, Beutler B, Hartog K, Thayer R, Brown-Shimer S, Cerami A. Identification of a common nucleotide sequence in the 3'untranslated region of mRNA molecules specifying inflammatory mediators. Proc Nail Acad Sci USA 1986;83:1670 13. Chao W, Liu H, Hanahan DJ, OIson MS. Regulation of plateletactivating factor receptors in rat Kupffer cells. J Biol Chem 1989;264:20448 14. Liu H, Chao W, Olson MS. Regulation of the surface expression of the platelet-activating factor receptor in IC-21 peritoneal macrophages: effects of lipopolysaccharide. J Biol Chem 1992;267:20811 15. Somers SD, Weiel JE, Hamilton TA, Adams DO. Phorbol esters and calcium ionophore can prime murine peritoneal macrophages for tumor cell destruction. J Immunol 1986;136:4199 16. Beutler B. Cachexia: a fundamental mechanism. Nutr Rev 1988;46:369 17. Burchett SK, Weaver WM, Westall JA, Larsen A, Kronheim S, Wilson CB. Regulation of tumor necrosis fector/cachectin and IL-1 secretion in human mononuclear phagocytes. I Immunol

S38 1988; 140:3473 18. Sariban E, Imamura K, Luebbers R, Kufe D. Transcriptional and posttranscriptional regulation of tumor necrosis factor gene expression in human monocytes. J Clin Invest 1988;81:1506 19. Beutler B, Krochin N, Milsark IW, Luedke C, Cerami A. Control of cachectin (tumor necrosis factor) synthesis: mechanisms of endotoxin resistance. Science 1986;232:977 20. Howard AD, Erickson KL. The induction and augmentation of macrophage tumoricidal responses by platelet-activating factor. Cell Immunol 1995;164:105 21. Chao W, Olson MS. Platelet-activating factor: receptors and signal transduction. Biochem J 1993;292:617 22. Garcia-Sainz JA, Romero-Avila T, Olivares-Reyes JA, Macias-Silva M. Guinea pig hepatocyte alpha lA-adrenoceptors: characterization, signal transduction and regulation. Eur J Pharmacol 1992;227:239 23. Higashijima T, Uzu S, Nakajima T, Ross EM. Mastoparan, a peptide toxin from wasp venom, mimics receptors by activating GTP-binding regulatory proteins (G proteins). J Bioi Chem 1988;263:6491 24. Mousli M, Bronner C, Bueb JL, Tschirhart E, Gies JP, Landry Y. Activation of rat peritoneal mast cells by substance P and mastoparan. J Pharmacol Exp Ther 1989;250:329 25. Higashijima T, Burnier J, Ross EM. Regulation of Gi and Go by mastoparan, related amphiphilic peptides, and hydrophobic amines: mechanism and structural determinants of activity. J Biol Chem 1990;265:14176 26. Weingarten R, Ransnas L, Mueller H, Sklar LA, Bokoch GM. Mastoparan interacts with the carboxyl terminus of the alpha subunit ofGi. J Biol Chem 1990;265:11044 27. Mahoney EM, Hamill AL, Scott WA, Cohn ZA. Response of endocytosis to altered fatty acyl composition of macrophage phospholipids. Proc Natl Acad Sci USA 1977;74:4895 28. Schroit AJ, Gallily R. Macrophage fatty acid composition and phagocytosis: effect of unsaturation on cellular phagocytic activity. Immunology 1979;36:199 29, Mahoney EM, Scott WA, Landsberger FR, Hamill AL, Cohn ZA. Influence of fatty acyl substitution on the composition and function of macrophage membranes. J Biol Chem 1980;255:4910 30. Schlager SI, Meltzer MS. Role of macrophage lipids in regulating tumoricidal activity. II. Internal genetic and external physiologic regulatory factors controlling macrophage tumor cytotoxicity also control characteristic lipid changes associated with tumoricidal cells. Cell Immunol 1983;80:10 31. Whitaker MO, Wyche A, Fitzpatrick F, Sprecher H, Needleman P. Triene prostaglandins: prostaglandin D3 and eicosapentaenoic acid as potential antithrombotic substances. Proc Natl Acad Sci USA 1979;76:5919 32. Dyerberg J, Bang HO, Stoffersen E, Moncada S, Vane JR. Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis? Lancet 1978;ii: 117 33. Kremer JM, Bigauoette J, Michalek AV, Timchalk MA, Lininger L, Rynes RI, Huyck C, Zieminski J, Bartholomew LE. Effects of manipulation of dietary fatty acids on clinical manifestations of rheumatoid arthritis. Lancet 1985;i: 184 34. Pricker JD, Robinson DR, Steinberg AD. Dietary enrichment with the polyunsaturated fatty acid eicosapentaenoic acid prevents proteinuria

FISH OIL AND MACROPHAGE TUMORICIDAL ACTIVITY and prolongs survival in NZB × NZW FI mice, J Clin Invest 1981;68:556 35. Leslie CA, Gonnerman WA, Ullman MD, Hayes KC, Franzblau C, Cathcart ES. Dietary fish oil modulates macrophage fatty acids and decreases arthritis susceptibility in mice. J Exp Med 1985;162:1336 36. Prickett JD, Trentham DE, Robinson DR, Dietary fish oil augments the induction of arthritis in rats immunized with type II collagen. J Immunol 1984;132:725 37. Carroll KK, Braden LM. Dietary fat and mammary carcinogenesis. Nutr Cancer 1985;6:254 38. Erickson KL, Schlangur DS, Adams DA, Fregeau DR, Stern JS. Influence of dietary fatty acid concentration and geometric configuration on murine mammary tumorigenesis and experimental metastasis. J Nutr 1984;114:1834 39. Abraham S, Hillyard LA. Effect of dietary 18-carbon fatty acids on growth of transplantable mammary adenocarcinomas in mice. J Natl Cancer Inst 1983;71:601 40. Ip C, Carter CA, Ip MM. Requirement of essential fatty acid for mammary tumorigenesis in the rat. Cancer Res 1985;45:1997 41. Endres S, Ghorhani R, Kelley VE, Georgilis K, Lonnemann G, van der Meer JW, Cannon JG, Rogers TS, Klempner MS, Weber PC, et al. The effect of dietary supplementation with n-3 polyunsaturated fatty acids on the synthesis of interleukin-1 and tumor necrosis factor by mononuclear cells. N Engl J Med 1989;320:265 42. Billiar TR, Bankey PE, Svingen BA, Curran RD, West MA, Holman RT, Simmons RL, Cerra FB. Fatty acid intake and Kupffer cell function: fish oil alters eicosanoid and monokine production to endotoxin stimulation. Surgery 1988;104:343 43. Somers SD, Erickson KL. Alteration of tumor necrosis factor-alpha production by macrophages from mice fed diets high in eicosapentaenoic and docosahexaenoic fatty acids. Cell Immunol 1994;153:287 44. Petersen CM, Moiler BK. Immunological reactivity and bioactivity of tumour necrosis factor [letter]. Lancet 19885:934 45. Moiler B, Mogensen SC, Wendelboe P, Bendizen K, Petersen CM. Bioactive and inactive forms of tumor necrosis factor-alpha in spinal fluid from patients with meningitis. J Infect Dis 1991;163:886 46. Hardard'ottir I, Whelan J, Kinsella JE. Kinetics of tumour necrosis factor and prostaglandin production by murine resident peritoneal macrophages as affected by dietary n-3 polyunsaturated fatty acids. Immunology 1992;76:572 47. Hardard'ottir I, Kinsella JE. Tumor necrosis factor production by murine resident peritoneal macrophages is enhanced by dietary n-3 polyunsaturated fatty acids. Biochim Biophys Acta 1991;1095:187 48. Bankey PE, Billiar TR, Wang WY, Carlson A, Holman RT, Cerra FB. Modulation of Kupffer cell membrane pbospholipid function by n-3 polyunsaturated fatty acids. J Surg Res 1989;46:439 49. Sperling RI, Benincaso AI, Knoell CT, Larkin JK, Austen KF, Robinson DR. Dietary omega-3 polyunsaturated fatty acids inhibit phosphoinositide formation and chemotaxis in neutropbils. J Clin Invest 1993;91:651 50. Weber C, Aepfelbacher M, Lux I, Zimmer B, Weber PC. Docosahexaenoic acid inhibits PAF and LTD4-stimulated [Ca2+]iincrease in differentiated monocytic U937 cells. Biochim Biophys Acta 1991;1133:38