Effects of dietary fish oil on NFκB gene expression and related signaling in spleen of chickens stimulated with lipopolysaccharide*

The study was conducted to investigate the effects of fish oil and maize oil on nuclear factor kappa B (NFκB) gene expression and the downstream pathways of intracellular signaling in spleen of chickens after lipopolysaccharide (LPS) stimulation. Two hundred eighty eight chickens were assigned in a 2×2 factorial design. Factors were dietary fat type (4.5% maize oil or 4.5% fish oil) and immunological stimulation (LPS or saline). LPS increased levels of eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and arachidonic acid (AA) of spleen in chickens after the second LPS stimulation on 28 d of age. Fish oil alleviated the increase of EPA and DHA in spleen of chickens after LPS stimulation at 27 d of age. Fish oil supplementation decreased prostaglandin 2 (PGE2) production and the activity of cyclooxygenase 2 (COX2) after LPS stimulation. LPS stimulation increased the activity of phospholipase C (PLC) in spleen of chickens. And fish oil inhibited activity of PLC in spleen of chickens stimulated by LPS. Meanwhile fish oil decreased the production of * Supported by the National Natural Science Foundation of China (No. 31001017), the Natural Science Foundation of Shaanxi Province (No. 2010JQ3002), the Support of Opening Program State Key Laboratory of Animal Nutrition (No. 2004DA125184F0809), the Natural Science Foundation Council of China (No. 30671520), the State Outstanding Young Scientist Scholarship of China (No. 30425037) and the Central University Special Fund Basic Research and Operating Expenses (No. QN2009020) 3 Corresponding author: e-mail: guoym9899@yahoo.com.cn Journal of Animal and Feed Sciences, 20, 2011, 206–223


INTRODUCTION
Polyunsaturated fatty acids (PUFA) are categorized according to the position of the last double bond near their methyl end into n-3 and n-6 series.Animals cannot introduce double bonds between carbon 9 and the methyl end of fatty acid.Therefore, animals cannot convert n-6 PUFA into n-3 PUFA or vice versa.Fish oil is the main dietary source of n-3 PUFA enriched with eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA); and maize oil is the main dietary source of n-6 PUFA enriched with arachidonic acid (AA).Fish oil modulates the immune responses including genes expression and lymphocytes proliferation, thereby exerting beneficial effects on inflammatory processes (Liu et al., 2003;Yang et al., 2008).Cyclooxygenase (COX), also known as prostaglandin endoperoxide synthase or as prostaglandin G/H synthase, is a key membrane bound enzyme responsible for the oxidation of AA to prostaglandin E 2 and the oxidation of EPA to prostaglandin E 3 .Two COX isoforms have been identified, COX 1 and COX 2 .COX 1 enzyme is expressed constitutively in most tissues and cells to control synthesis of those PGs primarily involved in the regulation of homeostatic functions throughout the body and remains constant under either physiological or pathological conditions.COX 2 is an intermediate response gene that encodes a protein normally absent from most cells but highly inducible in certain cells in response to inflammatory stimuli resulting in enhanced PG release.PGs formed by COX 2 primarily mediate pain and inflammation.COX 1 was present in both normal and neoplastic tissue equally.COX 2 is also induced by a variety of factors including tumour promoters, cytokines, growth factors and hypoxia and is overexpressed in rheumatoid arthritis (Simmons et al., 2004).Its utilization of AA also perturbs the level of intracellular free AA and subsequently affects cellular functions.The anti-inflammatory activity of corticosteroids correlates with preventing the release of AA from storage phospholipids, and glucocorticoids inhibit the expression of COX 2 (Tapiero et al., 2002).
Some reviews have summarized that constituents in fish oil are antiinflammatory, and the most likely candidates for this beneficial effect are the two n-3 fatty acids found in highest abundance in fish oil, EPA and DHA (McMurray et al., 2000;Calder and Field, 2002;Yang et al., 2008).Spleen, an important immune organ of poultry, can transport lymphocytes to peripheral blood and reclaim lymphocytes from peripheral blood.Lymphocytes of spleen can bind and recognize antigens, which activates immune responses.However, it is unclear how fish oil regulates signaling of lymphocytes by immune stress.NF-κB is found in almost all animal spleen cell types and is involved in cellular responses to stimuli such as stress, cytokines, free radicals, and bacterial or viral antigens (Gilmore, 2006;Perkins, 2007).NF-κB plays a key role in regulating the immune response to infection.Conversely, incorrect regulation of NF-κB has been linked to cancer, inflammatory and autoimmune diseases, septic shock, viral infection, and improper immune development.In order to assess immune capabilities regulated by fish oil after immune stimulation, it is essential to understand the immune signaling.The present study was designed to determine the effects of dietary fish oil and maize oil on NFκB gene expression and the downstream pathways of intracellular signaling in spleen of chickens subjected to an immune stress, LPS.

Animals and experimental procedures
These 288 chickens were assigned in a 2×2 factorial arrangement.The main factors were dietary fat type (4.5% maize oil or 4.5% fish oil) and immunological stimulation (LPS or saline).Birds were randomly assigned to four groups of seventy-two chicks each, with 6 replications of twelve chicks in a cage with constant lighting.Feed and water were available ad libitum in each dietary group.On 20 and 27 d of age, all chickens were injected intra-abdominally with either 1 mg/kg body weight (BW) of E. coli LPS (Escherichia coli serotype 055: B 5 , Sigma) or sterile saline.The animal care and experimental procedures conformed to the regulations of China Agricultural University Animal Care and Use Committee.Fish oil and maize oil were supplied by Dongmiao Co. Ltd. and China Corn Oil Co. Ltd., respectively.The fatty acid profiles of fish oil and maize oil were analysed by gas chromatography as shown in Table 1.

Spleen sampling and analytical procedures
Spleen sampling.After LPS injection at 20 d of age, the spleen from a chicken every replication on 21 d of age was aseptically removed.After LPS injection at 27 d of age, the spleen from a chicken every replication on 28 d of age was aseptically removed.The spleens were preserved in liquid nitrogen immediately and then transferred to appropriate tubes.All tubes were stored at -80°C until analysis.
Extraction and isolation of lipids and measurement of fatty acids composition.Lipids were extracted from spleen by modified method according to Kramer and Zhou (2001).The 0.5 g portions of tissues, stored frozen at -80°C, were quantitatively pulverized using a stainless steel mortar and pestle kept at dry ice temperature.The pulverized content was transferred into a beaker containing 5 ml chloroform/methanol (2:1), gently homogenized, and C:17 was added as an external standard.Then sample filtered through a sintered glass funnel to collect total lipids.
The methylation procedures were NaOH/methanol at 50°C for 30 min, and HCl/methanol at 80°C for 60 min, and then added 1 ml water and 6 ml heptane.Chromatographic analysis was performed on a HP6890 chromatograph equipped with a flame ionization detector system.Separations were conducted using a 100 m × 0.25 mm capillary column (CP-Sil 88) with a 0.25 μm stationary phase thickness.The carrier gas was purified helium at a flow rate of 1.7 ml/min.The gas chromatography (GC) conditions were as follows: injector temperature 250°C, detector temperature 250°C, initial oven temperature 180°C for 45 min, increased to 215°C at a rate of 10°C/min, then maintained at 215°C for 17 min.The sample and carrier gas mixture (helium) were carried at 1:40.
Measurement of phospholipase A 2 (PLA 2 ), COX and prostaglandin 2 (PGE 2 ).Activity of PLA 2 was measured according to the method of Reynolds et al. (1992).
The spleen was homogenized in 9 volume (W/V) of ice-cold PBS with a polytron tissue homogenizer for 2 min.The supernatant was obtained by centrifugation at 10000 rpm for 15 min, 4°C.Protein concentration of the supernatant was determined by the method of Lowry with bovine serum albumin as the standard (Stoscheck, 1990).
Activity of COX in spleen of chickens was determined according to the method of Meng et al. (2003).The 60 μl supernatant stemming from spleen of chicken was averagely divided into three tubes: A, B, C, respectively.The 60 μl reaction mixture containing 20 μl Tris-HCl buffer (pH 7.5, 100 mM), 20 μl hematin (H3281, Sigma; 20 μmol/l) and 20 μl 2',7'-dichlorodihydrofluorescein diacetate (D6883, Sigma; 50 μmol/l) were added to each tube.Then 20 μl nimsulide (N1016, Sigma) (2.5 μmol/l) were added to C tube.The tubes were incubated for 10 min at 30°C.Twenty μl substrate (9 μmol/l AA) was added to A tube and B tube without adding inhibitor (nimsulide).The reaction was performed for 20 min at 30°C.Then 2 ml of cold PBS were added to all tubes.A bilateral laser scanning fluorescence miscroscope (Cary Eclipse, Varian) with an argon ion laser beam for excitation was used for fluorescent measurement of COX activities and expression.Excitation and emission wavelengths were 450 and 550 nm, respectively.Data of COX activity indicated as fluorescence intensity (FI).FI C minus FI A was COX 1 activities.FI B minus FI C was COX 2 activities.FI B minus FI A was total COX activities, and total COX activities included COX 1 activities and COX 2 activities.The chicken spleen supernatant prepared as above was harvested and stored at -80°C until analysed.The PGE 2 levels of the supernatant were measured by PGE 2 kit (900-001, Assay Designs Inc.), according to the manufacturer instructions.CV of intra-assay precision determined by taking samples containing low, medium and high concentrations of PGE2 was 8.9, 5.8 and 17.5%, respectively.CV of interassay precision determined by measuring three samples with low, medium and high concentrations of PGE2 was 3.0, 5.1 and 3.9%, respectively.The percentage cross reactivity for PGE 2 , PGE 1 , PGE 3 , PGA 2 , PGF 1α was 100, 70, 16.3, 0.1, 1.4 and 0.1%, respectively.The percentage sample recovery was 101.3%.
Membrane fluidity of splenocytes.Splenocyte membrane was isolated by the methods of Giraud et al. (2000).The tissues (1 g) were homogenized at 4°C in a glass tube with a Teflon pestle attached to a motorized homogenizer (1000 rpm) with four 15-s bursts.The sample was filtered through cheese cloth and centrifuged at 1500 g for 10 min.The resulting microsomal pellet was resuspended in 1.0 ml NaHCO 3 (1 mM, pH 7.5), then 3 ml of 81% sucrose (w/v) were added.The microsome suspension from spleen was layered on a discontinuous sucrose gradient built up by carefully adding 1.5 ml 53.4% sucrose, 3.0 ml 48.0% sucrose and 1.5 ml 42.9%, then centrifuged at 70.000 g for 90 min at 4°C.The white, fluffy layer formed at the interface of the 42.9 and 48.0%sucrose layers.The pellet was harvested and washed with PBS (1 mM, pH 7.5, with 30% sucrose).The entire preparation procedure was conducted at 4°C.Protein concentration of the plasma membranes was determined by the method of Lowry with bovine serum albumin as the standard (Stoscheck, 1990).The same volume of sample prepared was mixed with 2 ml PBS (1 mM, pH 7.5) containing 1,6-diphenyl-1,3,5-hexatriene (DPH, 2×10 -6 mol/l prepared from a DPH stock solution of 2×10 -3 mmol/l in tetrahydrofuran).The suspension was incubated for 30 min at room temperature.The sample was excited with vertically polarized light (362 nm) and emission (438 nm) was measured through the polarizer both parallel and perpendicular to the excitation polarizer.The parallel (I//) and perpendicular (I') fluorescence intensities were recorded at 37°C after careful temperature equilibration.The term G, which was calculated by I 90°,0°/ I 90°,90°, was used to normalize the determinations of the parallel and perpendicular fluorescence intensities The fluorescence polarization (FP) was calculated: The fluorescence polarization refers to the rotational motion of the DPH distributed throughout the hydrophobic core of the lipid bilayers and is inversely proportional to the membrane fluidity; therefore, the lower the FP, the more fluid is the membrane.
Measurement of phospholipase C (PLC), sphingomyelinases (SMase) and IP 3 .Membranous PLC and SMase activity were determined as described previously (Wu et al., 1997).After homogenizing, the spleen suspensions were centrifuged at 800 g, 4°C for 15 min, the supernatant was centrifuged again at 100.000 g at 4°C for 60 min.The pellets were resuspended in buffer (20 mmol/l Tris-HCl, pH 7.0, 10 mmol/l EGTA, 1 mmol/l DTT, 1 mmol/l PMSF, 0.5 mg/ml sodium deoxylcholate and 0.1 mol/l NaCl), treated by sonification several times for 2 min at 4°C, and used as membranous PLC and SMase.The supernatant was used for determination of membranous PLC and SMase.Protein concentration of the supernatant was determined by the method of Lowry with bovine serum albumin as the standard (Stoscheck, 1990).The IP 3 levels of the supernatant were measured by IP 3 Fluorescence Polarization kit (90-0037-01, Amersham Biosciences), according to the manufacturer instructions.
PLC assay.For Ca 2+ -dependent PLC, the reaction mixture (500 μl) contained 100 mmol/l Tris-HCl, pH 7.0, 1.5 mmol/l CaCl 2 (omitted in controls), and 5 mmol/l L-α-phosphatidylcholine (P3556, Sigma).The reaction was started by adding 10 μl of the sample to the mixture.After incubation at 37°C for 15 min, the reaction was stopped by adding 2 ml chloroform/methanol/HCl (2:1:0.035,vol/vol/vol).The reaction mixture was then vigorously agitated in a vortex mixer and centrifuged at 1000 g for 10 min.Each supernatant (160 μl) was transferred to another mixture (0.5 ml) containing 0.21 mol/l Tris, pH 9.8, 50 mmol/l MgCl 2 , and 4 units of alkaline phosphatase (TaKaRa).After incubation at 37°C for 30 min, 2 ml of chromogen solution (0.31% ammonium molybdate, 0.66 mol/l sulphuric acid, 0.011% malachite green and 0.018% Tween-20) was added, and the resulting mixture was incubated at 37°C for another 20 min.The optical density was measured at 660 nm (wavelength).Enzyme activity was expressed as nmol per min per mg of protein (nmol/min • mg).For Ca 2+ -independent PLC, the reaction mixture containing 1 mmol/l EDTA and the procedure were the same as in the assay for Ca 2+ -dependent PLC except that CaCl 2 was excluded.The control tubes for Ca 2+ -dependent PLC were used as test tubes for Ca 2+ -independent PLC, and the tubes in which samples were added after the reaction was stopped were set as control.
Real-time quantitative RT-PCR.Total splenic RNA was extracted using the the TRIzol Reagent (Invitrogen, Life Technologies) according to the manufacturer's instructions.Total RNA was reverse transcribed according to the method described by Sijben et al. (2001) and Lai et al. (2005).Cycle parameters for the RT procedure were 1 cycle of 37°C, 1 h; 1 cycle of 70°C, 15 min; and 1 cycle of 4°C, 5 min.The RT products (cDNA) were stored at -20°C for relative quantification by PCR.The reverse transcribed products (cDNA) were for relative quantification by PCR.NFκB amplification primers were designed using the Primer Premier 5.0 software programme.The following sequences of PCR primer pairs were used: forward 5'-GTG TGA AGA AAC GGG AAC TG-3', reverse 5'-GGC ACG GTT GTC ATA GAT GG -3'.
Real-time PCR for quantification of NFκB mRNA levels in chicken spleens were quantified using a method based on that of Sijben et al. (2001).Briefly, quantitative analysis of PCR was carried out according to optimized PCR protocols and SYBR Green PCR master Mix Kit (Applied Biosystems).For the PCR reaction, the following experimental run protocol was used: enzyme incubation (50°C for 2 min), denaturation programme (95°C for 10 min), amplification and quantification programme repeated 38 times (94°C for 20 s, different annealing temperature for different target genes for 20 s, 72°C for 20 s with a single fluorescence measurement), melting curve programme (65-95°C with a heating rate of 0.1°C/s and a continuous fluorescence measurement) and finally 72°C for 10 min.The annealing temperature for NFκB was 53°C.Quantification was based on the increased fluorescence detected by the DNA Engine Opticon 2 fluorescence detection system (MJ Research).

Statistical analysis
Statistical analysis of the chicken data were conducted using the GLM procedure of SAS ( 2001) in a 2×2 factorial arrangement of treatments in a randomized complete block design, and the following model was used: where: Y ijk -the digestion index in the k th chicken on the a i diet on the b j stimulation, μ -the overall meal, a i -the fixed effect of the i th diet; b j -the fixed effect of the j th stimulation; (ab) ij -the interaction term and e ijk -the random error connected with ijk -observation.
Correlations analysis between COX and PGE 2 , PLC and IP 3 were conducted using the Pearson procedure.SAS codes for correlation coefficient test were provided in the following section.
proc corr data = a outp = corr; *outp with pearson correlation coefficient; var x y; run where: proc corr -analysis programme for correlations; outp -the product moment correlation coefficient; var -analytical parameters; a -the data file for COX, PGE 2 , PLC and IP 3 , x, y -the variables of COX and PGE 2 , PLC and IP 3 , respectively.

Growth performance
There were no significant effects of dietary oils on body weight gains (BW = 0.708 kg) on 21 d of age in the chickens.During the first LPS stimulation period (from 21 to 27 d of age), LPS stimulation did not reduced body weight gains of chickens.After the second LPS stimulation, chickens (BW = 1.65 kg) stimulated with LPS had lower body weight gains (P<0.05)than that in those (BW = 1.75 kg) treated with saline during week 4 to week 6.Feed conversion efficiency was significantly higher in chickens stimulated with LPS than in those treated with saline.

Major PUFA composition of splenocytes
In all 4 feeding groups, neither LPS stimulation nor dietary oils affected the levels of LA, EPA and DHA in spleen of chickens at 21 d of age, but the level of AA was decreased by the first LPS stimulation at 21 d of age (Table 2).Among the four groups, chickens fed fish oil diet had the highest levels of EPA and DHA in spleen after the second LPS stimulation.

Membrane fluidity of splenocytes
The effects of diets supplemented with different oils on membrane fluidity in splenocytes membrane of chicken stimulated with LPS were investigated by measuring FP (Table 3).FP is inversely correlated to membrane fluidity.After the first and second LPS stimulation, FP of splenocyes membrane of chickens on 21 and 28 d of age was significantly lowered in response to LPS stimulation.No significant differences in FP of splenocytes membranes were observed between chickens consuming fish-oil diets and chickens consuming maize oil.

PLA 2 , COX activities and PGE 2 production
There were no significant differences in activities of PLA, COX and COX 1 in spleen of chickens on 21 d and 28 day of age between groups consuming diets supplemented with fish oil and maize oil (Table 4).Both the first and second LPS stimulation increased the activities of PLA and COX.Meanwhile chickens given fish oil compared with those given maize oil had lower activities of COX.There were significant interactions on COX 2 activities and PGE 2 production between dietary oils and LPS stimulation in chickens.Fish oils ameliorated the increasing of COX 2 activities and PGE 2 production in spleen of chickens stimulated with LPS.

SMase activities in spleen
The effects of dietary different types of oils on SMase activities in spleen of chickens stimulated by LPS were shown in Table 5. Dietary oils did not significantly affect activities of SMase in spleen of chickens on 21 and 28 day of age.One day after the first and second LPS stimulation, the activities of acid SMase of membrane and plasma in spleen were significantly higher in chickens stimulated with LPS than that in chickens treated with saline.However, there was no significant interaction on the acid SMase activities in spleen of chickens between LPS stimulation and dietary oils.One d after the first LPS stimulation, the activities of membrane neutral SMase in spleen were significantly higher in chickens stimulated with LPS than that in chickens treated with saline.However, one day after the second LPS stimulation, the activities of plasma neutral SMase in spleen were significantly increased.There was no significant interaction on the neutral SMase activities in spleen of chickens between LPS stimulation and dietary oils.

Membrane PLC activities in spleen
Membrane Ca 2+ -independent PLC activities in spleen between chickens given maize oil and fish oil were not significantly different on 21 and 28 day of age (Table 6).LPS stimulation did not significantly affect the activities membrane Ca 2+ -independent PLC.There were significant interactions on membrane Ca 2+ -dependent PLC in the spleen between dietary oils and LPS stimulation in chickens.Fish oils decreased the increasing activity of membrane Ca 2+ -dependent PLC in the spleen of chickens after the first and second LPS stimulation.

IP 3 levels of splenocytes
No significant differences in splenocytes of chickens on 21 and 28 day of age were observed between groups given maize oil and fish oil (Table 7).After the first and second LPS stimulation, chickens on 21 and 28 day of age had higher levels of IP 3 in splenocytes than those of chickens injected by saline.There were significant interactions on IP 3 production between dietary oils and LPS stimulation in chickens.Fish oils attenuated the increase of IP 3 production in splenocytes of chickens stimulated with LPS.

NFκB gene expression of spleen
There was an interaction between LPS stimulation and dietary oil (P=0.0079for the first stimulation; P=0.0017 for the second stimulation) for the mRNA abundance of NF κ B (Table 8).Among chickens stimulated with LPS, those fed fish oil diets had lower mRNA abundance of NF κ B than those fed maize oil diets.

Correlation analysis
Table 9 showed the correlation between COX 2 and PGE 2 , PLC and IP 3 in spleen of chickens on 21d and 28 d of age.A good correlation existed between COX 2 and PGE 2 (R 2 =0.791;P<0.05, 21d of age; R 2 =0.756;P<0.05, 28 d of age).And there was a good correlation between IP 3 productivity and membrane Ca 2+ -dependent PLC activity.

DISCUSSION
In our study, there was no significant interaction in splenocytes fatty acids profile between the first LPS injection and fish oil or maize oil, but LPS stimulation had dramatic effects on splenocytes AA in chickens (Table 2).This result goes against previous studies (Selvaraj and Cherian, 2004;Wiesenfeld et al., 2005).The effects of dietary oils on changes of fatty acids composition of splenocytes was not greatly significant compared with the culturing lymphocytes in the medium supplemented with fatty acids (Yaqoob et al., 1995).It was a pity that we can not explain it commendably.Our future studies will investigate what could lead to the phenomena.The second LPS stimulation decreased the proportions of LA and AA, whereas there was significant interaction in the proportions of EPA and DHA of splenocytes in chickens on 28 d of age.And maize oil inhibited the increasing proportions of DHA and EPA in spleen of chickens stimulated with LPS.The result suggested that dietary fish oils increase the proportions of n-3 PUFA (DHA and EPA) of splenocytes in chicken stimulated by LPS.Some studies showed that the increasing availability of n-3 PUFA in the diets (fish oil, as a good source of n-3 PUFA) played an important role in ameliorating inflammatory immune responses (Korver and Klasing, 1997;Korver et al., 1998;Gaines et al., 2003;Liu et al., 2003;Yang et al., 2006).
In the present study, we found that LPS stimulation increased the membrane fluidity of splenocytes in chickens (Table 3).These effects have been observed previously in the other study (Calder et al., 1994).Altered fatty acids composition might be expected to influence immune cell function.Changes of fatty acid composition in membrane of splenocytes led to changes of membrane fluidity.Supplementation of the culture medium with particular fatty acids (AA, DHA and EPA) led to the increasing of membrane fluidity of lymphocytes (Calder et al., 1994).However, the effects were not observed in our study (Table 3).The present study indicated that fish oil would not change the membrane fluidity in splenocytes of chickens at inflammation status.The fatty acid composition changes of splenocytes membrane resulted from diet in our study were less extreme than those changes seen in cell culture (Yaqoob et al., 1995).We cannot explain this result, and we will investigate the possible reason.No significant interactions between fish oil and LPS on membrane fluidity of splenocyte indicated that fish oil would not modulate immune inflammatory of chickens stimulated with LPS through changing the membrane fluidity of splenocytes.
Our present study showed that dietary fish oil did not significantly affect the PLA 2 activity, but effectively inhibited the COX 2 activity and PGE 2 production of spleen in chickens stimulated with LPS (Table 4).From our study, we find a good correlation between COX 2 and PGE 2 (R 2 =0.791,P<0.05, 21 d of age; R 2 =0.756,P<0.05, 28 d of age).In addition, the correlation between COX 2 activity and PGE 2 production of spleen on 21 and 28 d of age indicated that COX 2 played a key role in PGE 2 production (Table 9).Similar to the result, Arntzen et al. (1998) found that after supplementing with n-3 PUFA, PGE 2 were significantly reduced in decidual cell stimulated by LPS.Whereas, supplementing with n-6 PUFA, there was a significant increase in PGE 2 of decidual cell stimulated by LPS.The result is consistent with our previous findings in laying hens too (Guo et al., 2004).Dietary PUFA affects immune cellular functions not only by themselves but also by metabolites, such as PGE 2 (He et al., 2007).Usually, the highly abundant AA is liberated from membrane phospholipids by PLA 2 to be used for eicosanoid synthesis.AA metabolized through COX gives rise to PGE 2 .Li and Watkins (1998) reported that dietary fish oil reduced ex vivo PGE 2 biosynthesis in rats fed diets rich in n-3 PUFA.n-3 PUFA showed suppression of COX 2 expression and PGE 2 production in the cultured murine monocytic cells when stimulated in vitro with LPS (Lee et al., 2003).Based on these studies, it is crucial for the lower PGE 2 production of spleen in chickens stimulated with LPS that fish oil inhibits COX 2 activity.It may be that one way fish oil exerts its beneficial effects on inflammatory responses of chickens stimulated with LPS is by lessening of inflammatory mediators, such as PGE 2 .
In our present study, LPS stimulation increased SMase activity of spleen in chickens.However, fish oil did not ameliorate the increasing of SMase activity (Table 5).No significant interactions between fish oil and LPS on SMase of splenocyte indicated that fish oil did not modulate inflammatory response of chickens through changing the SMase activity of splenocytes.Signaling through the sphingomyelin pathway is ubiquitous and evolutionarily conserved, ceramide may paly a role as a second messenger in intracellular signal transduction.The ceramaide generation is via SMase (Pena et al., 1997).The outcome of signaling through this pathway is apoptosis of lymphocytes.The signaling pathway for ceramide involves the action of SMase, sphingomyelin-specific forms of PLC, which hydrolyses the phosphodiester bond of sphingomyelin, a phospholipid preferentially found in the plasma membrane of mammalian cells yielding ceramide and phosphorylcholine (Kolesnick and Martin, 1998).Receptors as distinct as those for TNFα, IL-1 signal via the sphingomyelin pathway following ligand binding, which gives rise to inflammatory responses (Pushkareva et al., 1995;Spiegel et al., 1996).
In our work, we evaluated the participation of the PLC activity and IP 3 production in the activation of splenocytes in chickens stimulated by LPS (Tables 6 and 7).Phosphatidylcholine specific PLC is able to hydrolyse phosphatidylinositol-4,5-bisphosphate (PIP 2 ), then produce the DAG necessary for PKC activation in several signaling receptors and concomitant production of IP 3 for Ca 2+ mobilization.Recent data have suggested that several cytokine receptors as well as other cell surface receptors can mediate DAG-dependent PKC activation without activation of phosphoinositide-PLC isoforms (PLC-γ 2 ), IP 3 production in female BALB/c mice (Moreno-García et al., 2005).Whereas, our data demonstrated that LPS stimulation mediated its crucial activity of spleen in chickens that leads directly to the activation of PC-PLC and IP 3 production.While it is clear that the PLC signal is absolutely involved in inflammatory responses in cultured murine lymphocytes (Kolesnick and Martin, 1998).From our study, we found that fish oil attenuated the PLC activity and IP 3 production in spleen of LPS-injected chickens.Meanwhile, there was a good correlation between IP 3 productivity and membrane Ca 2+ -dependent PLC activity (Table 9).These events indicated that fish oil would ameliorate immune inflammatory through changing the PLC activity and IP 3 production in splenocytes of chickens stimulated with LPS.Kumar and Chakrabarti (2000) reported that Ca 2+ mobilization initialized by IP 3 was necessary to activation of PKC, which participated the spleen T lymphocyte proliferation stimulated by Con A. Fish oil decreased IP 3 production in splenocytes of chickens, which would inhibited intracellular Ca 2+ mobilization.And inhibition of intracellular Ca 2+ mobilization down-regulated Ca 2+ -dependent PLC activity that it was involved with production of IP 3 again.
Meanwhile, our data demonstrated that there were significant interactions between fish oil and LPS stimulation on NFκB gene expression in spleen of chickens at 21 and 28 d of age.And fish oil ameliorated the increase of NF κ B mRNA in chickens stimulated with LPS (Table 8).It is possible that fish oil ameliorates inflammatory responses in chickens stimulated with LPS through attenuating NF κ B gene transcription, the similar results was seen in piglets (Liu et al., 2003).In a previous study, EPA prevented LPS-induced TNFα expression by preventing NFκB activation in cultured monocytic THP-1 cells (Zhao et al., 2004).Similarly, RAW cells grown in EPA-rich media had less TNF mRNA expression and a decreased activation of the NF κ B (Lo et al., 1999).Several proinflammatory cytokines, such as TNFα, IL-1, IL-6, are encoded by target genes of NFκBactivation pathway.PKC-dependent degradation of IκB result in translocation of NFκB into the nucleus, which give rise to the production of the pro-inflammatory cytokines (Karin and Greten, 2005).The stimuli including inflammatory cytokines and immune-related stress such as LPS lead to the activation of NFκB, which can accelerate expression of COX 2 , PLA 2 , proinflammatory cytokines and acute phase response protein, etc. (Garg and Aggarwal, 2002;Bonizzi and Karin, 2004).

CONCLUSIONS
The results of these experiments give insight into a potential dietary method to ameliorate immune inflammatory responses that might occur during infectious stimulation.The signaling of immunomodulatory nutrients such as fish oil in inflammation is that PLC and COX 2 activity decreased by fish oil inhibited IP 3 and PGE 2 production, respectively.These results showed that fish oil acted as an antiinflammatory agent, which may be associated with down-regulation of production of IP 3 and PGE 2 , respectively.The results of NFκB gene expression suggested fish oil might alleviate immune stress at the level of transcription.

Table 1 .
Fatty acids composition of fish oil and maize oils

Table 2 .
Effects of feeding different oils and lipopolysaccharide stimulation on polyunsaturated fatty acids composition of splenocytes in chickens, mol%

Table 3 .
Effects of different oils and lipopolysaccharide stimulation on membrane fluidity of splenocytes in chickens

Table 4 .
Effects of dietary oils and LPS stimulation on spleen PLA2, COX and its metabolite in chickens

Table 5 .
Effects of feeding different oils and lipopolysaccharide stimulation on SMase activities of spleen in chickens, nmol/min .mg

Table 6 .
Effects of feeding different oils and lipopolysaccharide stimulation on PLC activities of spleen in chickens

Table 7 .
Effects of feeding different oils and lipopolysaccharide stimulation on IP 3 production of splenocytes in chickens, IP 3 , pmol/mg

Table 8 .
Effects of dietary oils and lipopolysaccharide stimulation on NF κ B mRNA of spleen in chickens

Table 9
MPLC and PPLC indicated as membrane phospholipase C and plasma phospholipase C respectively.An asterisk (*) indicates a significant correlation P<0.05.PGE 2 -prostaglandin E 2 ; COX 2cyclooxygenase 2. IP 3 is the abbreviation of inositol triphosphate