Fish oils contain the long chain (n-3) polyunsaturated

Transcription

1 J Vet Intern Med 2004;18: Effect of Type of Dietary Polyunsaturated Fatty Acid Supplement (Corn Oil or Fish Oil) on Immune Responses in Healthy Horses Jean A. Hall, Robert J. Van Saun, Susan J. Tornquist, Joseph L. Gradin, Erwin G. Pearson, and Rosemary C. Wander The objective of this study was to compare effects of dietary polyunsaturated fatty acid supplementation (corn oil or fish oil) on selected immune responses in normal horses. Two groups of horses (n 5) were randomly assigned a dietary supplement with either 3.0% corn oil or fish oil for a period of 14 weeks. Plasma fatty acid profiles were monitored to ensure uptake of dietary fatty acids. Cell-mediated immunity was assessed by a delayed-type hypersensitivity (DTH) skin test to keyhole limpet hemocyanin (KLH), and humoral immunity was assessed by measuring antibody titers to KLH. Production of prostaglandin E 2 (PGE 2 ), expression of tumor necrosis factor- (TNF- ), and phagocytosis of latex beads by bronchoalveolar lavage fluid (BALF) cells were also assessed. Lipopolysaccharide (LPS)-stimulated BALF cells from horses fed corn oil showed a higher production of PGE 2 compared with those from horses fed fish oil at 6 and 12 weeks. Production of TNF- by LPS-stimulated BALF cells was higher in both groups of horses at 6, 8, and 12 weeks compared with pretrial values, and phagocytic activity of BALF cells was higher at 8 and 12 weeks, however, there were no differences between the 2 groups of horses. The DTH skin test and antibody titers to KLH revealed no differences between horses fed corn or fish oil. Based on these studies, dietary polyunsaturated fatty acids modulate the inflammatory response of horses. Both fatty acid supplements increased production of the proinflammatory cytokine TNF-, whereas only corn oil increased production of the proinflammatory eicosanoid PGE 2 by LPS-stimulated BALF cells. It is possible that fish oil, because it did not increase production of PGE 2, could have value in the treatment of equine recurrent airway obstruction or other equine inflammatory diseases. Key words: Bronchoalveolar lavage fluid; (n-3) Fatty acids; PGE 2 ; Phagocytosis; Tumor necrosis factor-. Fish oils contain the long chain (n-3) polyunsaturated fatty acids (PUFAs) eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Corn oil is a rich source of linoleic acid, an (n-6) PUFA. Increasing the amount of dietary (n-3) PUFA relative to (n-6) PUFA allows (n-3) fatty acids to be preferentially incorporated into cell membrane phospholipids. 1 The type of eicosanoids that cells produce, and consequently communication between cells of the immune system, can potentially be modulated through dietary supplementation of PUFA. 2 9 Consumption of fish oils also reduces cell-mediated immune responses. 2 9 For example, there is decreased lymphocyte proliferation, T cell-mediated cytotoxicity, natural killer cell activity, macrophage-mediated cytotoxicity, monocyte and neutrophil chemotaxis, major histocompatibility class II expression and antigen presentation, and adhesion molecule expression. 2 9 Dietary supplementation with fish oil can also affect production of certain cytokines. From the Department of Biomedical Sciences (Hall, Tornquist, Gradin) and the Department of Large Animal Clinical Sciences (Van Saun, Pearson), College of Veterinary Medicine, and the Department of Nutrition and Food Management, College of Home Economics and Education (Wander), Oregon State University, Corvallis, OR Dr Van Saun is presently affiliated with the Department of Veterinary Science, College of Agricultural Sciences, Pennsylvania State University, University Park, PA Dr Wander is presently affiliated with the Human Nutrition Research Laboratory, Department of Nutrition, School of Human Environmental Sciences, The University of North Carolina at Greensboro, Greensboro, NC Reprint requests: Jean A. Hall, DVM, PhD, Department of Biomedical Sciences, College of Veterinary Medicine, Oregon State University, Magruder Hall 105, Corvallis, OR ; Jean.Hall@oregonstate.edu. Submitted December 4, 2003; Revised February 27, 2004 and May 3, 2004; Accepted June 2, Copyright 2004 by the American College of Veterinary Internal Medicine /04/ /$3.00/0 Cytokines mediate the immune response by altering lymphocyte proliferation, differentiation, and activation. There is decreased production of the proinflammatory cytokines interleukin (IL)-1, IL-6, and tumor necrosis factor- (TNF- ) when a diet is supplemented with (n-3) PUFA. 4,5,10 14 Also, it has been shown that dietary consumption of PUFA affects other immune functions such as phagocytosis and antibody production. 3,15 18 The (n-3) PUFA may work by altering signal transduction, gene expression, or both within inflammatory and immune cells. 5 The goal of this investigation was to compare the effect of diets rich in (n-3) fatty acids on selected immune responses of healthy horses. Evaluated variables included immune functions of bronchoalveolar lavage fluid (BALF) cells (phagocytosis of latex beads, production of TNF-, and PGE 2 production), systemic antibody production to keyhole limpet hemocyanin (KLH), and a delayed-type hypersensitivity (DTH) skin test to KLH. Of particular interest in this study was the effect of dietary (n-3) fatty acids on TNF- and PGE 2 production. Our goal was to determine whether dietary supplementation with fish oil reduced expression of the proinflammatory eicosanoid PGE 2, proinflammatory cytokine TNF-, or both in BALF cells from normal horses. The horses used in his study were the same horses described in a companion paper. 19 In that paper, we measured leukotriene B (LTB) 4 and LTB 5 production in calcium ionophore stimulated peripheral blood neutrophils. In this paper, the focus is on immune responses of cells isolated from airways of these horses, as well as on in vivo tests to assess cell-mediated and humoral immunity. Materials and Methods Horses and Diets Horses and dietary treatments used for this study were the same as those previously described. 19 Briefly, 10 mature, nonpregnant, lightbreed mares ranging in age from 5 to 20 years and weighing between

2 (n-3) FA and Immune Responses in Horses and 553 kg (mean 502 kg) were fed hay-based diets with a supplement that differed only in its source of oil, either corn oil (low n- 3) or fish oil (high n-3). Horses were determined to be healthy based on repeated physical examinations, complete blood counts, and serum biochemical evaluations. Protocol Horses were randomly assigned to 1 of 2 feeding groups (n 5) and fed oil-supplemented diets for a period of 14 weeks following the diet acclimation period, as previously described. 19 Jugular venous blood was collected into tubes with ethylenediamine tetraacetic acid at 0, 6, 8, and 12 weeks. BALF was collected at 0, 6, 8, and 12 weeks. At 8 weeks horses were inoculated with KLH suspension. A second KLH booster inoculation was given at week 10, 2 weeks after the first inoculation. Cell-mediated immune response was evaluated using a DTH skin test the day after the second KLH inoculation, at 10 weeks and 1 day. At 12 weeks, the KLH antibody titer was determined. The experimental protocol was reviewed and approved by the Oregon State University Animal Care and Use Committee according to principles outlined by the National Institutes of Health. 20 Bronchoalveolar Lavage BALF cells were isolated to evaluate phagocytic activity and for stimulation to assess PGE 2 and TNF- production. Horses were sedated with 5 mg of detomidine a given IV for the procedure. The bronchus was infused 3 times with physiological saline (100-mL aliquots) and fluid was immediately aspirated after each infusion. Approximately ml of BALF was obtained from each horse. The BALF was kept on ice until centrifuged for 20 minutes at 400 g. Supernatants were discarded and cell pellets were combined and resuspended in 2 ml of Hank s balanced salt solution. Cell concentrations were determined using a Coulter ZB1 Counter. b Cell differential counts were determined by a certified medical technologist by way of microscopic examination of a Wright-Giemsa stained aliquot of these cells. Phagocytosis of Latex Beads Phagocytic activity of BALF cells was assessed using latex beads. 21 Briefly, for horses, cells were incubated on ice with either a monoclonal antibody c (mouse, anti-equine, pangranulocyte/monocyte marker, cell-line DH59B) 22 or an isotype negative control d (a purified mouse immunoglobulin G [IgG] 1 isotype standard) for 5 minutes in the dark. After the incubation, cells were washed, and a phycoerythrin-conjugated secondary antibody e (goat anti-mouse IgG, F(ab ) 2 ) was added to each well. After a second incubation on ice in the dark for 5 minutes, cells were washed and then resuspended. Fluorescent bead engulfment was assessed using a flow cytometer. Each assessment was based on counting 5,000 cells. Tumor Necrosis Factor- Activity Aliquots of BALF cells were resuspended in RPMI-1640 media containing 30 g/ml of lipopolysaccharide (LPS; Escherichia coli O55:B5). f Medium was also supplemented with 100,000 U/L penicillin, 100 mg/l streptomycin, 2 mmol/l-glutamine, and 5% fetal calf serum. Cells were incubated in tissue culture flasks for 40 hours at 37 C in a humidified atmosphere of 5% CO 2. After incubation, cells were centrifuged, and supernatants were filtered through a 0.45-micron filter g and stored at 70 C until assayed. TNF- activity was determined using a bioassay previously described. 23 Briefly, TNF- activity was measured using L929-8 cells, a subclone of a murine fibroblastoid cell line isolated by Branch et al. 24 The percentage of cell lysis as a measure of TNF- activity was calculated. In previous experiments by Tornquist et al, 23 specificity of the assay was demonstrated by preincubating equine TNF standard (LPSstimulated equine serum) and selected foal serum with an anti-equine TNF- neutralizing monoclonal antibody (HL801, a gift from R.J. MacKay and J. Cargile, University of Florida, Gainesville, FL) before the assay. Control wells received an equivalent amount of an irrelevant isotype-matched monoclonal antibody. Prostaglandin E 2 Production Concentration of PGE 2 in supernatants from LPS-stimulated BALF cells was determined using an enzyme immunoassay kit h according to the manufacturer s instructions. 25 Prostaglandin E 2 standard h was diluted to 100 pg/ml in enzyme immunoassay buffer and served as an internal standard. The interassay coefficient of variation (based on the internal standard) was 16%. Delayed-Type Hypersensitivity Skin Test Horses were sensitized with KLH i suspension administered IM (500 g of KLH emulsified in 1.0 mg of T1501 adjuvant for a total volume of 0.5 ml) at 8 and 10 weeks after initiation of the feeding trial. The KLH and adjuvant were combined in an oil-water emulsion as described 26 ; however, the preparation was sonicated instead of ground. The KLH was stored at 20 C until it was administered to the horses. For the DTH skin test, KLH was heat-aggregated according to the method described by Exon et al 27 and injected the day following the second KLH IM injection. Intradermal injections (0.05 ml /site; approximately 3 mg of KLH) were administered at 5 sites (2 sites on the external pinna of the left ear and 3 sites on the right lateral neck) that had been previously shaved of hair. Histamine base j (0.1 g/l) was administered as a positive control, and physiological saline was administered as a negative control. All injections were administered intradermally at least 2.5 cm apart. Examinations were made at 30 minutes and at 24, 48, 72, and 96 hours after intradermal injections. Ear thickness was measured before injecting the ear. Reactions were recorded according to diameter of induration (neck and ear) and thickness (ear) at each site. Keyhole Limpet Hemocyanin Antibody Titer Serum was collected for KLH antibody titer at 12 weeks and was measured by a modification of an indirect ELISA procedure previously described by Woodard. 26 Final results were expressed as the log of antibody titer. Interassay coefficient of variation (based on an internal standard) was less than 10%. Other Chemical Assays Complete blood cell counts were determined at 0, 6, 8, and 12 weeks using a hematology analyzer. k A certified medical technologist determined white cell differential counts by microscopic examination of blood smears after Wright-Giemsa staining. Statistical Analysis The trial was a completely randomized design. Data were analyzed using the mixed procedure of Statistical Analysis Systems 28 for repeated measures. Main effects were diet, week, and their interaction analyzed over time. For each dependent variable, a horse nested in treatment (diet) was subjected to differing covariance structures. The best model fit was determined by lowest parameter values for covariance structure. For main effects found to be significant, mean differences were determined by pairwise differences or probability values for differences of the least-squares means (PDIFF) for preplanned comparisons. 28 Data are reported as least squares means SEM unless otherwise indicated. A two-sample t-test was used to compare data not meeting repeated measures criteria. Data for the KLH antibody titer required a log transformation before comparing groups. Values were considered significant at P.05 unless otherwise indicated.

3 882 Hall et al Fig 1. Effect of feeding horses diets supplemented with oils that differed in the (n-6) and (n-3) fatty acid content on phagocytic activity of bronchoalveolar lavage fluid (BALF) cells. Horses fed a corn oil supplement (white bars) received a low level of (n-3) fatty acids, whereas horses fed a fish oil supplement (black bars) received a high level of (n-3) fatty acids. Each bar represents the percentage of phagocytosis (population of cells engulfing 1 or more fluorescent beads total population of cells 100%) after the horses had consumed their respective diets for 0, 6, 8, and 12 weeks. Values are mean SEM, n 4 or 5. Different letters above bars indicate when mean values differed significantly (P.05) across time of study. Results Horses and Diets As reported in a companion paper, 19 dietary treatment resulted in an altered plasma fatty acid profile, which is consistent with the supplement s fatty acid profile. Horses consuming fish oil experienced an elevation of (n-3) fatty acids, predominately EPA and DHA (but also arachidonic acid; AA), starting at 6 weeks and continuing through 12 weeks of the study. BALF Cytologies Differential cell counts on BALF were not influenced by dietary treatment, but the percentage of macrophages (P.001), lymphocytes (P.0001), and other cells (P.02, including eosinophils, mast cells, and epithelial cells) were altered by collection week. Neutrophil percent was not different across diet or time periods, ranging between 1.4 and 3.2%. The percentage of macrophages increased over the study collection periods from % to %. The mean percentage of macrophages at weeks 8 and 12 were greater than those in the pretrial period. The percentage of lymphocytes started at %, declined to % at week 8, and then increased to % at week 12. The percentage of other cells was highest during the pretrial period ( %) and declined over time ( % at week 12). Phagocytosis Fig 2. Effect of feeding horses diets supplemented with oils that differed in the (n-6) and (n-3) fatty acid content on TNF- production by lipopolysaccharide-stimulated bronchoalveolar lavage fluid (BALF) cells. Horses fed a corn oil supplement (white bars) received a low level of (n-3) fatty acids, whereas horses fed a fish oil supplement (black bars) received a high level of (n-3) fatty acids. Each bar represents the percentage of lysis of L929 cells by TNF- after the horses had consumed their respective diets for 0, 6, 8, and 12 weeks. Values are mean SEM, n 4 or 5. Bars with * above them denote a significant difference (P.05) from week 0 within a group of horses. Sample week as a main effect also influenced mean production of tumor necrosis factor- activity by BALF cells. Different letters above bars indicate when mean values for both groups of horses differed significantly (P.05) across time of study. Dietary means were different only at week 6. The pangranulocyte/monocyte marker did not distinguish between macrophages and granulocytes. Because the predominant BALF cell type was the macrophage, phagocytosis results were interpreted as macrophage activity. Percentage of BALF cells engulfing latex beads increased with week of study. Phagocytic activity was highest at week 12 compared with all other test times for horses fed both diets. Although the percentage of macrophages engulfing latex beads was numerically higher in corn oil fed than in fish oil fed horses at all time points, differences between the 2 groups of horses were not significant at any time point during the feeding period (Fig 1). Both dietary treatments showed similar changes over time. Tumor Necrosis Factor- Production There was an observed interaction between dietary treatment and sample period for mean production of TNF- activity by BALF cells (Fig 2). Sample week as a main effect influenced mean production of TNF- activity. For both groups of horses, all sample periods had higher TNF- activity compared with that of week 0. Weeks 6 and 12 had the highest mean values, whereas week 8 activity was greater than that of week 0, but lower compared with that of weeks 6 and 12. Diet did not influence TNF- activity; however, corn oil fed horses had higher (88.2 versus 83.9% lysis of L929 cells) activity than fish oil fed horses. Spontaneous production of TNF- activity by BALF cells incubated in media without LPS was 0, 8, 15, and 18% of LPS-stimulated production at 0, 6, 8, and 12 weeks, respectively.

4 (n-3) FA and Immune Responses in Horses 883 Fig 3. Effect of feeding horses diets supplemented with oils that differed in the (n-6) and (n-3) fatty acid content on production of prostaglandin E 2 (PGE 2 ) by lipopolysaccharide-stimulated bronchoalveolar lavage fluid (BALF) cells. Horses fed a corn oil supplement (white bars) received a low level of (n-3) fatty acids, whereas horses fed a fish oil supplement (black bars) received a high level of (n-3) fatty acids. Each bar represents the concentration of PGE 2 in pg/ml from stimulated BALF cells after the horses had consumed their respective diets for 0, 6, and 12 weeks (Overall diet effect, P.03). Values are means SEM, n 4 or 5. Bars with P values above them indicate times when values between groups differed significantly. Different letters over bars within corn oil treatment indicate when mean values are different (P.01). Prostaglandin E 2 Quantification Production of PGE 2 by BALF cells was influenced by diet and sample week, but not diet by week interaction (Fig 3). Overall mean PGE 2 production was greater for horses fed corn oil (9,600 pg/ml) compared with those fed fish oil (4,020 pg/ml) over the 12 weeks of study. Production of PGE 2 for corn oil fed horses was greatest at week 12, whereas there was no difference across time periods for fish oil fed horses. Within sample weeks, PGE 2 production for corn oil fed horses was greater than those fed fish oil at 6 weeks (8,340 versus 2,770 pg/ml) and 12 weeks (12,540 versus 3,730 pg/ml). Spontaneous production of PGE 2 by BALF cells incubated in media without LPS was 0% of LPS-stimulated production at 0, 6, 8, and 12 weeks, respectively. The ratio of EPA to AA in the plasma was correlated with PGE 2 production by stimulated BALF cells. Across all weeks and dietary treatments, there was a negative correlation (correlation coefficient 0.46; P.01) between EPA : AA ratio and PGE 2 concentration. Across both dietary treatments, EPA : AA ratio was correlated (correlation coefficient 0.69; P.04) with PGE 2 production at week 12. Delayed-Type Hypersensitivity Skin Tests Both groups of horses had large reactions on the neck in response to the KLH intradermal injection: there was a central ring of induration surrounded by a larger area of swelling. Wheal diameter was influenced by time, but not diet, although there was a tendency (P.08) for an interaction between diet and time. Overall, wheal diameter increased from 0.5 hour ( mm, corn oil; mm, fish oil) to 24 hours ( mm, corn oil; mm, fish oil), then declined over time. In horses fed corn oil, the reaction area remained greater at 48 hours ( mm) and 72 hours ( mm) than at 0.5 hour. The reaction area was not different between 0.5 and 96 hours ( mm). In horses fed fish oil, the reaction area at 48 hours ( mm) was greater than it was at 0.5 hour, but less than it was at 24 hours. Reaction areas at 72 hours ( mm) and 96 hours ( mm) were not different from those of 0.5 or 48 hours. The area of DTH reaction was also measured at 2 independent points on the ear (base and tip). Reaction area at the ear tip was influenced by time, and diet by time interaction, but not by diet. Across dietary treatments, the reaction area was greater at all time points compared with 0.5 hour. Comparisons between time points were difficult to interpret because of large variation in mean values at 72 and 96 hours. Numerically, mean areas increased at each successive time up to 72 hours, and they remained high at 96 hours. In corn oil fed horses, reaction area at 24 hours ( mm 2 ) was the only time point different from 0.5 hour ( mm 2 ). In contrast, fish oil fed horses had a constantly increasing reaction area whereby all time points were greater than they were at 0.5 hour. No differences were found between time points for 48, 72, and 96 hours. Reaction area at the ear base was influenced by time, but not by diet or diet by time interaction. Across both treatments, reaction area was increased at all time points compared with 0.5 hour, but none of the time points were different from each other. Keyhole Limpet Hemocyanin Antibody Titer The humoral immune response to KLH was measured at 12 weeks; 2 weeks after horses received the second inoculation with KLH protein. There were no significant differences in antibody titers between the 2 groups of horses. The log titer was in corn oil fed horses and in fish oil fed horses. This was a significant response following sensitization to a foreign protein in both groups of horses. Total White Blood Cell and Differential Counts There was an interaction between diet and week on mean white blood cell count. Mean white blood cell count was influenced by week of study, but not by diet (9, cells/ L, corn oil; 8, cells/ L, fish oil). Most of the weekly variation in white blood cell count resulted from weekly effects and from diet by week effects on neutrophil counts. No other blood cell type was influenced by diet, week, or their interaction. All horses showed an alternating pattern for white blood cell and neutrophil counts, in which lower counts were observed at 0 and 8 weeks and higher counts at 6 and 12 weeks. Even though there were significant weekly changes in white blood cell and neutrophil counts, all values remained within normal reference intervals for horses.

5 884 Hall et al Discussion Increased phagocytosis of latex beads by pulmonary alveolar macrophages from horses fed corn or fish oil for 12 weeks was similar to what has been observed in rats. 15 Although the percentage of macrophages engulfing latex beads was higher in horses fed corn oil than in horses fed fish oil, and differences increased as the duration of the study progressed, there were no significant differences between groups of horses at any time point during the feeding period. Membrane fluidity is important in determining macrophage adhesion and phagocytic activity (eg, degree of unsaturation of fatty acids in culture media of macrophages affects subsequent phagocytic activity). 15 Phagocytic activity of macrophages increases after they are cultured with highly polyunsaturated fatty acids, yet macrophages cultured with EPA and DHA have lower phagocytic activity than AA-cultured cells. In our study, horses fed fish oil had higher plasma levels of EPA and DHA, but numerically lower phagocytic activity by pulmonary alveolar macrophages than horses fed corn oil, the latter having higher plasma levels of linoleic acid. The horses response to a DTH skin test following KLH administration after consuming corn or fish oil for 12 weeks was unexpected. In dogs, a high dietary intake of (n-3) PUFA suppresses the DTH response to intradermal injection of KLH antigen. 29 The diameter of induration was significantly smaller at 24 and 48 hours in dogs supplemented with a high level of (n-3) PUFA compared with dogs that consumed medium or low levels of an (n-3) PUFA supplement. In this study, the DTH reactions on the neck following KLH administration were most pronounced at 24 hours in horses fed both corn oil and fish oil. At 48 and 72 hours they were reduced in size, with those from horses fed fish oil being smaller. This was not considered a typical DTH response. These results were more consistent with an inflammatory response to KLH. Reactions at the ear base were not different between horses fed corn or fish oil, although wheal diameter and reaction area in both groups of horses was greatest at 72 hours, which is a more typical DTH reaction. Reactions at the ear tip in general persisted longer in fish oil fed horses for elevation, wheal diameter, and reaction area. At this location, fish oil fed horses had a more pronounced increase in reaction area with time. The KLH antigen has been used previously to evaluate DTH responses in horses, with the horses reactions to KLH starting early (4 hours) and subsiding after 24 hours. 30 The DTH skin response of horses to KLH might be a combination of both humoral and cellular mechanisms. 30 In our study, horses humoral immune response was evaluated by measuring the production of antibodies to KLH. IgG antibody titers were not significantly different between groups of horses in our study. A more definitive test of humoral immunity would involve measuring additional classes of immunoglobulin. In particular, measurement of IgE antibody titers would help define whether horses reaction to KLH was the result of an immediate hypersensitivity reaction. Yamada et al 16 have shown that unsaturated fatty acids can inhibit production of IgG, IgM, and IgA by cultured lymphocytes from rat mesenteric lymph nodes, while stimulating production of IgE. Those researchers concluded that this was, in part, related to production of oxidation products from fatty acids containing higher numbers of carbon atoms, double bonds, or both. In the study reported here, supplementation with either corn oil or fish oil enhanced production of TNF- by LPSstimulated BALF cells. Compared with pretrial levels, BALF cells from horses fed fish oil produced significantly higher levels of TNF- activity at 6, 8, and 12 weeks, whereas BALF cells from horses fed corn oil produced significantly higher TNF- activity at 6 and 12 weeks. In contrast, a decrease (approximately 10-fold) in TNF- production by LPS-stimulated peritoneal macrophages was noted in another study when horses were fed a diet enriched in -linolenic acid for 8 weeks. 31 It is interesting that in the study by Morris et al, 31 when cells were cultured in media without LPS, cells from horses consuming the -linolenic acid diet produced significantly more TNF- than cells obtained pretrial, before dietary intervention. Thus, it is difficult to interpret the effects of dietary -linolenic acid supplementation on TNF- production in that study. 31 In another study, horses were infused IV with 20% lipid emulsions enriched with (n-3) or (n-6) fatty acids. 32 Fatty acid analysis of monocyte membrane phospholipids demonstrated changes in the fatty acid composition persisting for up to 7 days after infusion. In vitro production of TNF- by peripheral blood monocytes was diminished by (n-3) lipid infusion and was unchanged or increased by (n-6) lipid infusion. Unstimulated monocytes had significantly lower TNF- production after (n-3) infusion compared with that before (n-3) infusion. 32 The inconsistencies between these two studies 31,32 and ours may indicate a differential response to constitutively produced TNF- and inducible TNF- in the presence of (n-3) fatty acids. Feeding dietary (n-3) PUFA has been shown to decrease PGE 2 production by peripheral blood mononuclear cells in dogs. 29 Dogs consuming diets high in (n-3) fatty acids showed a significant decrease in PGE 2 production compared with dogs that consumed diets low in (n-3) PUFA. In the study reported here, LPS-stimulated BALF cells from horses fed fish oil produced lower levels of PGE 2 than those from horses fed corn oil. It has been shown that high levels of PGE 2 can diminish TNF- production by increasing cellular levels of cyclic adenosine monophosphate, whereas low levels of PGE 2 stimulate cyclic guanosine monophosphate production, which leads to increased TNF- production. 11 However, the results of our study are not directly in support of the relationship that high levels of PGE 2 decrease TNF- production, because clearly, BALF cells from horses fed corn oil should have produced less TNF- than those from horses that were fed fish oil. In the current study, there was a significant, negative correlation between the ratio of EPA : AA in plasma fatty acids and the production of PGE 2 by LPS-stimulated BALF cells. Thus, as the EPA : AA ratio increased (eg, in the fish oil fed horses), PGE 2 production decreased. We did not measure the fatty acid profiles of membrane phospholipids of BALF cells in our study. However, others have shown that equine monocytes have increased (n-3) fatty acid incorporation into their membrane phospholipids when fed (n-3)

6 (n-3) FA and Immune Responses in Horses 885 enriched dietary supplements 33 or given (n-3) enriched IV infusions. 32 The PGE 2 monoclonal antibody used in this study had a 43% cross-reactivity with PGE 3. Because PGE 2 and PGE 3 were not separated by high-performance liquid chromatography before quantification, it could be that actual PGE 2 concentrations were even lower in the horses fed fish oil, because part of the recorded PGE 2 may have represented increased PGE 3, similar to the LTB 4 and LTB 5 findings in the accompanying paper by Hall et al. 19 Even though there were significant weekly changes in white blood cell and neutrophil counts, all values remained within normal reference intervals. Differential cell counts on BALF showed that neutrophil percentages were not different across diet or time periods. The changes in the other cell types (macrophages, lymphocytes, and others) were statistically but not clinically significant. In summary, dietary supplementation with either corn oil or fish oil modulated the inflammatory response of normal horses in several ways. Phagocytic activity of BALF cells was increased after dietary supplementation with either corn or fish oil compared with pretrial levels. Both fatty acid supplements increased production of the proinflammatory cytokine TNF- after 6 weeks of supplementation. In addition, corn oil increased production of the proinflammatory eicosanoid PGE 2 after 12 weeks of supplementation. Supplementation with fish oil appears to inhibit the LPS-induced increase in PGE 2 production by BALF cells. Footnotes a Dormosedan, SmithKline Beecham Animal Health, Exton, PA b Coulter Electronics, Inc., Hialeah, FL c VMRD Inc., Pullman, WA d Pharmingen International, San Diego, CA e Jackson ImmunoResearch Inc., West Grove, PA f Sigma Chemical Company, St. Louis, MO g UNIFLO-Plus, Scheicher and Schuell, Keene, NH h Cayman Chemical Company, Ann Harbor, MI i Calbiochem-Behring Diagnostics, La Jolla, CA j Histatrol Center Laboratories, Port Washington, NY k Baker 9010, Serono-Baker Instrument, Inc., Allentown, PA Acknowledgments This project was funded in part by the Agricultural Research Foundation, Strand Agriculture Hall Suite 100, Oregon State University, Corvallis, OR The authors thank Kathleen O Hara, Bernadette Stang, and Lisa Boeder for their valuable technical assistance. References 1. Cook HW. Fatty acid desaturation and chain elongation in eukaryotes. In: Vance DE, Vance JE, eds. Biochemistry of Lipids and Membranes. Menlo Park, CA: The Benjamin/Cummings Publishing Company; 1985: Calder PC. N-3 polyunsaturated fatty acids and immune cell function. Adv Enzyme Regul 1997;37: Calder PC. Dietary fatty acids and lymphocyte functions. Proc Nutr Soc 1998;57: Calder PC. Dietary fatty acids and the immune system. Nutr Rev 1998;56:S70-S Calder PC. Immunoregulatory and anti-inflammatory effects of n-3 polyunsaturated fatty acids. Braz J Med Biol Res 1998;31: Calder PC. Omega 3 polyunsaturated fatty acids, inflammation and immunity. World Rev Nutr Diet 2001;88: Calder PC. Polyunsaturated fatty acids, inflammation, and immunity. Lipids 2001;36: Miles EA, Calder PC. Modulation of immune function by dietary fatty acids. Proc Nutr Soc 1998;57: Kinsella JE, Lokesh B, Broughton S, Whelan J. Dietary polyunsaturated fatty acids and eicosanoids: potential effects on the modulation of inflammatory and immune cells: an overview. Nutrition 1990;6: James MJ, Gibson RA, Cleland LG. Dietary polyunsaturated fatty acids and inflammatory mediator production. Am J Clin Nutr 2000;71:343S 348S. 11. Calder PC. N-3 polyunsaturated fatty acids and cytokine production in health and disease. Ann Nutr Metab 1997;41: Endres S, Meydani SN, Dinarello CA. Effects of omega 3 fatty acid supplements on ex vivo synthesis of cytokines in human volunteers. Comparison with oral aspirin and ibuprofen. World Rev Nutr Diet 1991:66: Meydani SN, Endres S, Woods MM, et al. Oral (n-3) fatty acid supplementation suppresses cytokine production and lymphocyte proliferation: comparison between young and older women. J Nutr 1991; 121: Sasaki T, Kanke Y, Kudoh K, et al. Dietary n-3 polyunsaturated fatty acid and status of immunocompetent cells involved in innate immunity in female rats. Ann Nutr Metab 2000;44: Calder PC, Bond JA, Harvey DJ, et al. Uptake and incorporation of saturated and unsaturated fatty acids into macrophage lipids and their effect upon macrophage adhesion and phagocytosis. Biochem J 1990;269: Yamada K, Hung P, Yoshimura K, et al. Effect of unsaturated fatty acids and antioxidants on immunoglobulin production by mesenteric lymph node lymphocytes of Sprague-Dawley rats. J Biochem (Tokyo) 1996;120: Morrow WJ, Ohashi Y, Hall J, et al. Dietary fat and immune function. I. Antibody responses, lymphocyte and accessory cell function in (NZB NZW)F1 mice. J Immunol 1985;135: Fritsche KL, Johnston PV. Effect of dietary omega-3 fatty acids on cell-mediated cytotoxic activity in BALB/C mice. Nutr Res 1990; 10: Hall JA, Van Saun RJ, Wander RC. Dietary (n-3) fatty acids from Menhaden fish oil alter plasma fatty acids and leukotriene B synthesis in healthy horses. J Vet Intern Med 2004;18: National Research Council. Guide for the Care and Use of Laboratory Animals. Publication (rev.). Bethesda, MD: National Institutes of Health; Hall JA, Tooley KA, Gradin JL, et al. Effects of dietary n-6 and n-3 fatty acids and vitamin E on the immune response of healthy geriatric dogs. Am J Vet Res 2003;64: Davis WC, Marusic S, Lewin HA, et al. The development and analysis of species specific and cross reactive monoclonal antibodies to leukocyte differentiation antigens and antigens of the major histocompatibility complex for use in the study of the immune system in cattle and other species. Vet Immunol Immunopathol 1987;15: Tornquist SJ, Oaks JL, Crawford TB. Elevation of cytokines associated with the thrombocytopenia of equine infectious anaemia. J Gen Virol 1997;78: Branch DR, Shah A, Guilbert LJ. A specific and reliable bioassay for the detection of femtomolar levels of human and murine tumor necrosis factors. J Immunol Methods 1991;143: Hawkins DL, MacKay RJ, MacKay SL, Moldawer LL. Human

7 886 Hall et al interleukin 10 suppresses production of inflammatory mediators by LPS-stimulated equine peritoneal macrophages. Vet Immunol Immunopathol 1998;66: Woodard LF. Adjuvant activity of water-insoluble surfactants. Lab Anim Sci 1989;39: Exon JH, Bussiere JL, Mather GG. Immunotoxicity testing in the rat: an improved multiple assay model. Int J Immunopharmacol 1990;12: Littell RC, Milliken GA, Stroup WW, Wolfinger RD. SAS System for Mixed Models. Cary, NC: SAS Institute; Wander RC, Hall JA, Gradin JL, et al. The ratio of dietary (n- 6) to (n-3) fatty acids influences immune system function, eicosanoid metabolism, lipid peroxidation and vitamin E status in aged dogs. J Nutr 1997;127: Hodgin EC, McGuire TC, Perryman LE, Grant BD. Evaluation of delayed hypersensitivity responses in normal horses and immunodeficient foals. Am J Vet Res 1978;39: Morris DD, Henry MM, Moore JN, Fischer JK. Effect of dietary alpha-linolenic acid on endotoxin-induced production of tumor necrosis factor by peritoneal macrophages in horses. Am J Vet Res 1991;52: McCann ME, Moore JN, Carrick JB, Barton MH. Effect of intravenous infusion of omega-3 and omega-6 lipid emulsions on equine monocyte fatty acid composition and inflammatory mediator production in vitro. Shock 2000;14: Henry MM, Moore JN, Feldman EB, et al. Effect of dietary alpha-linolenic acid on equine monocyte procoagulant activity and eicosanoid synthesis. Circ Shock 1990;32: