METABOLISM AND NUTRITION. Incorporation of Different Polyunsaturated Fatty Acids into Eggs

Transcription

1 METABOLISM AND NUTRITION Incorporation of Different Polyunsaturated Fatty Acids into Eggs M. D. Baucells,*,1 N. Crespo,* A. C. Barroeta,* S. López-Ferrer,* and M. A. Grashorn *Departament de Nutrició Alimentació Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona, Bellaterra, Spain, and Department of Poultry Science, Institute for Animal Husbandry and Breeding, University of Hohenheim, Garbenstr. 17, Stuttgart, Germany ABSTRACT An experiment was carried out to examine Performance parameters were not significantly different thoroughly the relationships among different n-3 and n- 6 polyunsaturated fatty acids (PUFA) in the diet, their deposition into the eggs fat, and their effect on hens laying performance. A diet enriched with 4% fish oil (FO) was fed to the birds throughout the 14-wk laying period (Treatment 1; T1); this was the same oil source that was replaced in proportions of 25, 50, 75, or 100% with four different fat sources, resulting in 17 isocaloric dietary treatments: linseed oil (LO; T2 to T5), rapeseed oil (RO; T6 to T9), sunflower oil (SO; T10 to T13), and tallow (T; T14 to T17). Performance parameters were recorded weekly and analyzed on the basis of the replacing fat source. At the end of the 14-wk experimental period, eggs were collected, and their fatty acid (FA) profile was determined. among grouped treatments. Smaller proportions of FO in diets resulted in lower values of saturated and higher values of n-6 FA contents, regardless of the fat source used when replacing FO. The n-6 content increased mostly because of the rise in linoleic acid (LA), although the level of arachidonic acid (AA) was always higher when FO was completely suppressed. The amount of the different n-3 long-chain PUFA was lower (P < 0.001) when FO was present in lesser proportions in the diet. However, the slope of the decline of these FA changed according to the included fat. Replacing FO with LO resulted in the lowest decline of its derivatives by elongation and desaturation and an increase in the total n-3 FA in the form of linolenic acid (LNA). (Key words: egg, n-3 fatty acids, n-6 fatty acids) 2000 Poultry Science 79:51 59 INTRODUCTION There are some reports in the literature that deal with the effects of terrestrial sources of dietary n-3 fatty acids (FA) on the yolk FA composition. Enrichment of hens diets with sources rich in linolenic acid (LNA) has resulted in the production of eggs with significantly increased levels of yolk LNA and small but significantly higher increases in the 20-carbon family (long chain; LC) of n-3 FA [LC- PUFA (polyunsaturated FA)], mainly as eicosapentanoic acid (EPA) and docosahexanoic acid (DHA) (Caston and Lesson, 1990; Cherian and Sim, 1991). However, the increase in these FA in yolk has not been in a similar proportion to that of LNA. When compared with the increased levels of EPA and DHA reported by Hargis et al. (1991) in response to relatively low amounts (3%) of dietary fish oil (FO), the significantly smaller increases in the n-3 LC- PUFA observed when higher dietary fat levels in the form of flax and canola seed were supplied emphasizes the limited efficiency of conversion of LNA to EPA and DHA in birds (Aymond and Van Elswyk, 1995). Not all dietary n-3 FA are biologically equivalent (Barlow et al., 1990; Rein- Received for publication February 22, Accepted for publication August 11, To whom correspondence should be addressed: mariadolores. baucells@uab.es. hart, 1996), and what should be considered is not only the total amount of n-3 FA, but also the specific FA provided by food sources enriched with n-3 FA when developing health-modified products. To compare the beneficial effects of FO on the enrichment of n-3 LC-PUFA in eggs with the effects when other fat sources were used in egg production, an experiment with laying hens was carried out. The goal of this experiment was to improve knowledge about the relationship between precursors given in the diet and the content in egg of the different n-3 and n-6 PUFA families. Birds were fed with FO or with alternative mixtures of FO and other fat sources [linseed oil (LO), rapeseed oil (RO), sunflower oil (SO), or tallow (T)] in different proportions to attain a wide range of relative amounts of the different PUFA studied in eggs. Abbreviation Key: AA = araquidonic acid; DHA = docosahexanoic acid; DPA = docosapentanoic acid; EPA = eicosapentanoic acid; FA = fatty acid; FO = fish oil; LA = linoleic acid; LC = long chain, LNA = linolenic acid; LO = linseed oil; MUFA = monounsaturated fatty acids; PUFA = polyunsaturated FA; RO = rapeseed oil; SAT = saturated fatty acids; SO = sunflower oil; T = tallow; T1 = 100% fat added as FO; T2 = 75% FO, 25% LO; T3 = 50% FO, 50% LO; T4 = 25% FO, 75% LO; T5 = 100% LO; T2 = 75% FO, 25% LO; T3 = 50% FO, 50% LO; T4 = 25% FO, 75% LO; T5 = 100% LO; T6 = 75% FO, 25% RO; T7 = 50% FO, 50% RO; T8 = 25% FO, 75% RO; T9 = 100% RO; T10 = 75% FO, 25% SO; T11 = 50% FO, 50% SO; T12 = 25% FO, 75% SO; T13 = 100% SO; T14 = 75% FO, 25% T; T15 = 50% FO, 50% T; T16 = 25% FO, 75% T; T17 = 100% T. 51

2 52 BAUCELLS ET AL. MATERIALS AND METHODS Animals and Diets One hundred seventy LSL-White Leghorn hens 2 at 20 wk of age were housed in individual laying cages maintained in an environmentally controlled room at the Karlshof building of the University of Hohenheim. The birds were fed a standard layer diet (17% CP; 2,850 kcal EM/ kg diet). The diet was formulated according to the requirements recommended by the National Research Council (1994) on the basis of 30% wheat, 30% corn, and 15% extruded soybean as main ingredients. The basal diet was supplied with 4% FO 3 [100%; Treatment 1 (T1); control diet; Table 1). Mixtures were made of different proportions (75, 50, 25, and 0%) of FO and the following fat sources 4 : LO, SO, RO, or T. The result was 17 experimental treatments. Pullets were randomly assigned to each treatment (n = 10 cages). The lipid profile of the experimental diets is shown in Table 2. Feed consumption was measured for each dietary treatment at the end of the trial, which lasted for 14 wk. Eggs were collected daily, and egg production was calculated on a weekly basis. Egg weight controls were performed during 3 consecutive d every 4 wk. Eggs were collected for chemical analysis during the last 3 d of the experimental period and then weighed and cracked; thereafter, yolks were separated. Five samples of 4 or 5 pooled yolks for each treatment were freezedried 5 and stored at 20 C before the FA analyses were performed. The total fat of diets and yolks was extracted according to Folch et al. (1957) and methylated with 5% boron trifluoride methanol complex in methanolic solution (Morrison and Smith, 1964). The lipid profile was determined at the Unitat de Nutrició i Alimentació Animal of the Universitat Autònoma de Barcelona by means of gas chromatography in a Shimadzu GC-14A chromatograph 6 equipped with a BPX70 capillary column (SGE capillary column; length, 30 m; I.D., 0.53 mm; 70% cyanopropyl polysilphenylene-siloxane stationary phase), film, and a flame ionization detector. The operating conditions of the gas chromatograph were as follows. The initial temperature was 75 C, increasing by 4 C/min to 148 C; from 148 to 158 C, the temperature was increased by 2.5 C/min. The temperature of the injector and the detector remained stable at 280 C. The column head pressure of the conductor gas (helium) was 1.30 g/cm 2. The FA percentage was integrated and then calculated using the CLASS-Unipac program provided by Shimadzu Europa GmbH (HPLC Group) 7 by means of direct normalization 2 Lohmann Co., Cuxhaven, Germany. 3 Nagel Co., Hamburg, Germany. 4 Tallow provided in the form of Bergafat HTL-106 by Berg and Schmidt, Hamburg, Germany. Linseed, rapeseed and sunflower oils provided by Graf Co., Nürnberg, Germany. 5 FTS Systems, Inc., New York, NY. 6 IZASA, C/ Calàbria, , Barcelona, Spain. 7 Ingeniería Analítica, S.L., Crta. Cerdanyola 65-67, Sant Cugat del Vallès, Spain. 8 Sigma Química, S.A., Apdo. Correos 161, Alcobendas, Spain. TABLE 1. Percentage composition of experimental diets Ingredients 1 (%) Corn Wheat Extruded soybean (48% CP) Limestone 8.64 Gluten feed 5.00 Oats 5.00 Added fat Dicalcium phosphate 1.18 Calcium propionate 0.40 Sodium bicarbonate 0.31 Salt 0.25 Vitamin premix Choline chlorate 0.20 L-Lysine 0.15 DL-Methionine 0.11 Mineral premix Tryptophan 0.06 L-Threonine 0.05 Antioxidant buthylhydroxytoluol 0.02 Cytranaxanthine 0.01 Yellow carophyle 0.01 Calculated nutrient content ME, kcal/kg 2,870 Available P 0.32 Met + Cys 0.71 Lysine 0.87 Chemical analyses of control diet DM 90.6 CP 17.4 Ash 11.9 Crude fat 6.5 Crude fiber 3.9 Sugars 3.0 Starch 40.6 Calcium 3.39 Phosphorus Calculated vitamin and mineral content of diets provided (per kilogram of diet): vitamin A, 12,000 IU; cholecalciferol, 3,000 IU; vitamin E, 30 mg; riboflavin, 5 mg; pantothenic acid, 14 mg; nicotinic acid, 50 mg; choline, 2,000 mg; folic acid, 1 mg; vitamin B 12,30µg; Mn, 112 mg; and Zn, 91 mg. 2 T1 = 100% fat added as fish oil. 3 Composition of vitamin premix Vit-Vorm 6/1.5 provided (per kilogram of premix): vitamin A, 6,000,000 IU; cholecalciferol, 1,500,000 IU; vitamin E, 15,000 mg; riboflavin, 3,000 mg; pantothenic acid, 7,000 mg; nicotinic acid, 25,000 mg; folic acid, 500 mg; and vitamin B 12, 15,000 µg. Supplied by Animedica (Horb, Germany). 4 Composition of mineral premix SpürElevor SG1 provided as follows per kilogram of premix: Mn, 120,000 mg; Zn, 80,000 mg; Fe, 90,000 mg; Cu, 15,000 mg; I, 1,600 mg; Se, 500 mg; and Co, 600 mg. Supplied by Animedica (Horb, Germany). of the peak areas. Each FA was identified in the form of a methyl ester by comparing the retention times with the standard acquired at Sigma Quimica S.A. 8 Statistical Analysis Seventeen experimental treatments were designed according to a series of mixtures of different proportions of the studied fats (FO, LO, RO, SO, and T). Performance parameters and FA percentages were subjected to AN- OVA by using the GLM procedure of SAS (SAS Institute, 1996). Means were compared for significant difference (P < 0.05) by using the least significant difference method of the same statistical package. The comparative response

3 POLYUNSATURATED FATTY ACIDS IN EGGS 53 TABLE 2. Fatty acid composition of experimental diets 1,2 Experimental diet Fatty acid T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12 T13 T14 T15 T16 T17 (% of total methyl esters of fatty acids) C 18:1 n C 18:2 n C 18:3 n C 20:4 n C 20:5 n C 22:5 n C 22:6 n Total SAT Total MUFA Total PUFA Total n Total n n-6:n The values presented are means of duplicate determinations. 2 T1 = 100% fat added as fish oil (FO); T2 = 75% FO, 25% linseed oil (LO); T3 = 50% FO, 50% LO; T4 = 25% FO, 75% LO; T5 = 100% LO; T6 = 75% FO, 25% rapeseed oil (RO); T7 = 50% FO, 50% RO; T8 = 25% FO, 75% RO; T9 = 100% RO; T10 = 75% FO, 25% sunflower oil (SO); T11 = 50% FO, 50% SO; T12 = 25% FO, 75% SO; T13 = 100% SO; T14 = 75% FO, 25% tallow (T); T15 = 50% FO, 50% T; T16 = 25% FO, 75% T; T17 = 100% T; SAT = saturated fatty acids; MUFA = monounsaturated fatty acids; PUFA = polyunsaturated fatty acids. Fish oil was provided by Nagel Co., Hamburg, Germany; linseed, rapeseed and sunflower oils were provided by Graf Co., Nürnberg, Germany, cold-pressed; tallow was provided in the form of Bergafat HTL-106 by Berg and Schmidt, Hamburg, Germany. of yolk FA in proportion to changes in dietary FA percentages was assessed by regression analysis. Data were adjusted to linear functions and subjected to a test of slope comparison. RESULTS AND DISCUSSION Egg Production Parameters Performance parameters are shown in Table 3; they are grouped according to the percentage of replacement of FO with each source of fat studied (0, 25, 50, 75, or 100% replacement), as well as according to different orthogonal contrasts performed: the type of replacing fat (FO, LO, RO, SO, or T) and the global percentage of FO present in the fat added to the experimental diets. There were no statistically significant differences among treatments. Hen-day egg productions higher than 80% were recorded throughout the entire experimental period for all treatments. In general, under our conditions (4% of added fat), the use of FO at any percentage in diets never led to poorer performance of the layers. No unpalatability of the diets was observed in relation to feed intake, which is in disagreement with the findings of Hulan et al. (1989), who observed palatability problems when including marine oils in diets and recorded poorer broiler chicken performance. Similarly, egg weight was not influenced by dietary treatment in the current study; mean weight was approximately 58.3 ± 1.23 g, and feed conversion per dozen eggs and per kg of eggs was 1.41 ± and 2.02 ± 0.088, respectively. In conclusion, no problems concerning the efficiency of production of the hens were observed, as was already demonstrated by Hargis et al. (1991), Jiang and Sim (1993), and Nash et al. (1995), when using several different added fats in the diet. When performance parameters were analyzed according to the relative percentage of PUFA or the linoleic acid (LA) amount in the diet (data not shown), there were no statistically significant differences in the studied parameters among treatments. These results are in accordance with the LA requirements determined by the NRC (1994). Fatty Acid Composition of Total Yolk Lipids The FA composition of total yolk lipids reflected that of the laying hen diets (Table 4). However, the magnitude of change was different according to the FA analyzed. Although significant statistically, there were no large differences in the degree of yolk saturation. The wide range in the degree of saturation of the experimental diets (resulting from the progressive substitution of FO with the other fat sources) hardly affected the respective levels of saturated fatty acids (SAT) of the yolk. In addition, as already mentioned by Jiang et al. (1991) and others, the ability of the laying hens to increase the monounsaturated fatty acids (MUFA) content (mainly as oleic acid) in yolk seems rather limited. That is, a slightly higher content (approximately 5 percentage units) of MUFA was found in yolk total lipids when the FO of the diet (T1; control diet) was completely replaced with RO (T9) or T (T17), whereas the respective MUFA content in these diets was 90% higher or 60% lower, than the amount present in the control diet (100% FO). These results clearly show the proven tendency of birds for keeping the degree of saturation and monounsaturation in yolk within very narrow margins. The highest percentage variation in the PUFA content was found with the progressive replacement of FO with SO and, to a lesser extent, with LO. However, in the first

4 54 BAUCELLS ET AL. TABLE 3. Performance parameters 1 of hens Hen s Hen s Feed Treatment 2 Intake initial weight final weight Eggs/hen/d Egg weight efficiency (g/hen/d) (g) (g) (kg feed/kg egg) T ,522 1, T ,457 1, T ,481 1, T ,517 1, T ,477 1, Pooled SEM , , T ,522 1, T ,454 1, T ,570 1, T ,434 1, T ,443 1, Pooled SEM , , T ,522 1, T ,475 1, T ,461 1, T ,455 1, T ,545 1, Pooled SEM , , T ,522 1, T ,452 1, T ,478 1, T ,483 1, T ,476 1, Pooled SEM , , FO 3 T ,522 1, LO T2 T ,483 1, RO T6 T ,475 1, SO T10 T ,484 1, T T14 T ,473 1, Pooled SEM , , % FO 4 T ,522 1, % FO T2, T6, T10, T ,460 1, % FO T3, T7, T11, T ,498 1, % FO T4, T8, T12, T ,472 1, % FO T5, T9, T13, T ,485 1, Pooled SEM , , Values are means of 10 observations per treatment and their pooled SEM. 2 T1 = 100% fat added as fish oil (FO); T2 = 75% FO, 25% linseed oil (LO); T3 = 50% FO, 50% LO; T4 = 25% FO, 75% LO; T5 = 100% LO; T6 = 75% FO, 25% rapeseed oil (RO); T7 = 50% FO, 50% RO; T8 = 25% FO, 75% RO; T9 = 100% RO; T10 = 75% FO, 25% sunflower oil (SO); T11 = 50% FO, 50% SO; T12 = 25% FO, 75% SO; T13 = 100% SO; T14 = 75% FO, 25% tallow (T); T15 = 50% FO, 50% T; T16 = 25% FO, 75% T; T17 = 100% T. 3 Values are means of either 10 or 40 observations per treatment (for FO and every other fat group, respectively) and their pooled SEM. 4 Values are means of either 10 or 40 observations per treatment (for 100% FO and every other FO percentage group, respectively) and their pooled SEM. case, the higher PUFA content was mainly due to a progressive increase in LA and a slight but constant increase in the arachidonic acid (AA) content. In the case of LO, the rise in PUFA can be explained because of an important increase in the LNA content of the yolk. It is nothing new that increasing levels of LA or LNA from different vegetable fat sources result in increases in their concentrations in the yolk lipid (Leskanich and Noble, 1997). The longer chain metabolites of LNA, such as EPA, docosapentaenoic (DPA), and DHA, decreased with any kind of withdrawal of FO from the diet. In previous findings (López-Ferrer et al., 1999), the simple substitution of FO for vegetable oil during the last week prior to slaughter of the broiler chickens resulted in a remarkable decline of the n-3 LC-PUFA in the meat of the birds. Nevertheless, when comparing the different fat sources used to replace FO, the highest values in these n-3 LC-PUFA in yolk were observed when LO was added to the diet and, to a lesser extent, when RO was added. These fat sources had the highest LNA content. On the other hand, it is interesting to note that, although the FO used in this study contained a higher level of EPA than DHA, the resulting level of DHA in the yolk was much higher than the EPA content as was previously found by other researchers (Oh et al., 1994; Marshall et al., 1994). Given that humans seem to have a rather limited

5 POLYUNSATURATED FATTY ACIDS IN EGGS 55 TABLE 4. Fatty acid composition of eggs 1 Treatment 2 C 18:2 n-6 C 18:3 n-3 C 20:4 n-6 C 20:5 n-3 C 22:5 n-3 C 22:6 n-3 SAT 3 MUFA PUFA Total n-6 Total n-3 n-6:n-3 T a 0.59 a 0.92 c d b a a 5.06 a 2.42 T b 0.64 ab 0.63 b cd ab b b 5.72 ab 2.46 T b 0.71 ab 0.66 b c b bc b 6.11 b 2.29 T c 0.77 bc 0.33 a b ab bc b 6.44 bc 2.28 T d 0.91 c 0.25 a a a c b 7.07 c 2.12 Pooled SEM P NS *** ** *** NS ** * NS ** * ** NS T a 0.44 a 0.59 a 0.92 d 0.47 c 3.18 d c a a 5.06 e 2.43 a T ab 0.51 ab 0.73 b 0.64 c 0.44 c 2.84 d bc a ab 4.48 d 2.90 a T bc 0.58 bc 0.93 c 0.43 b 0.33 c 2.48 c b a bc 3.86 c 3.67 b T c 0.66 cd 1.05 c 0.18 a 0.16 a 1.76 b a b c 2.82 b 5.19 c T c 0.73 d 1.43 d 0.08 a 0.09 a 1.01 a a b c 1.95 a 7.72 d Pooled SEM P * *** *** *** *** *** *** *** NS ** *** *** T a 0.44 c 0.59 a 0.92 e 0.47 e 3.18 e c c a a 5.06 e 2.43 a T b 0.32 b 0.90 b 0.58 d 0.34 d 2.72 d b c a b 4.02 d 3.69 ab T c 0.31 b 1.15 b 0.39 c 0.25 c 2.30 c bc b b c 3.31 c 5.49 b T d 0.26 ab 1.64 c 0.17 b 0.14 b 1.61 b a b c d 2.22 b c T d 0.18 a 2.20 d 0.02 a 0.06 a 0.52 a b a c e 0.82 a d Pooled SEM P *** ** *** *** *** *** *** *** *** *** *** *** T b 0.59 a 0.92 e 0.47 c 3.18 e b a c a 5.06 e 2.43 a T a 0.72 a 0.69 d 0.35 c 2.64 d b a bc a 4.01 d 3.04 ab T a 0.79 a 0.44 c 0.23 ab 2.24 c b b ab a 3.21 c 3.77 bc T a 1.01 b 0.25 b 0.17 a 1.88 b a c ab a 2.61 b 4.74 c T a 1.71 c 0.04 a 0.07 a 0.49 a a c a b 0.87 a d Pooled SEM P NS *** *** *** ** *** ** *** ** * *** *** FO 3 T ab 0.44 a 0.59 a 0.92 c 0.47 c 3.18 b a b b a 5.07 c 2.43 ab LO T2 T b 3.26 b 0.76 a 0.46 b 0.33 bc 2.22 a b b c b 6.37 d 2.28 a RO T6 T b 0.61 a 1.01 b 0.34 ab 0.26 ab 2.08 a c a b b 3.35 b 4.72 ab SO T10 T c 0.27 a 1.47 c 0.29 a 0.20 a 1.79 a b c d c 2.59 a c T T14 T a 0.25 a 1.06 b 0.36 ab 0.21 a 1.81 a a a a a 2.67 ab 6.78 b Pooled SEM P *** *** *** *** *** ** *** *** *** *** *** *** 100% FO 4 T b c 0.92 a 0.47 a 3.18 a a c 5.06 a 2.43 b 75% FO T2, T6, T10, T b c 0.64 b 0.37 a 2.75 b a c 4.49 a 3.05 b 50% FO T3, T7, T11, T ab c 0.48 c 0.30 b 2.41 c a bc 4.12 a 3.80 b 25% FO T4, T8, T12, T a b 0.23 d 0.19 c 1.81 d b ab 3.52 ab 5.60 b 0% FO T5, T9, T13, T a a 0.10 e 0.13 c 0.89 e b a 2.72 b a Pooled SEM P * NS *** *** *** *** *** NS NS ** * *** a,b,c,d,e Values within the same column and section of five treatments with no common superscript are different: *P < 0.05, **P < 0.01, or ***P < Values are means of five observations per treatment and their pooled SEM. 2 T1 = 100% fat added as fish oil (FO); T2 = 75% FO, 25% linseed oil (LO); T3 = 50% FO, 50% LO; T4 = 25% FO, 75% LO; T5 = 100% LO; T6 = 75% FO, 25% rapeseed oil (RO); T7 = 50% FO, 50% RO; T8 = 25% FO, 75% RO; T9 = 100% RO; T10 = 75% FO, 25% sunflower oil (SO); T11 = 50% FO, 50% SO; T12 = 25% FO, 75% SO; T13 = 100% SO; T14 = 75% FO, 25% tallow (T); T15 = 50% FO, 50% T; T16 = 25% FO, 75% T; T17 = 100% T. 3 SAT = Saturated fatty acids; MUFA = monounsaturated fatty acids; PUFA = polyunsaturated fatty acids. 4 Values are means of five or 20 observations per treatment (for FO group and for every other fat group, respectively) and their pooled SEM. 5 Values are means of five or 20 observations per treatment (for 100% FO and for every other FO percentage group, respectively) and their pooled SEM. ability to desaturate the last step of formation of DHA from LNA (Sanders, 1993), eggs can be considered an interesting dietary source of such FA. It must be recalled that DHA is obtained in yolk from a double path; on the one hand, because of the direct deposit of DHA from the diet and, on the other, as a final result of the de novo synthesis from its precursors (LNA, EPA, and DPA) given in the diet (Leskanich and Noble, 1997; Meluzzi et al., 1997). Regression analysis showed a clear relationship between the proportions of PUFA and n-3 FA family in the dietary fat and in the lipid yolk deposits (Table 5). This observation supports the established idea that the amount of FA in the diet is directly responsible for its deposit in the egg s fat. Concerning the C 20 and C 22 series of the n-3 FA, it can be observed that 85 or 78% of all variability of the content in EPA or DHA in egg, respectively, can be attributed to their proportion in the diet. That fact supports, once again, the idea that the FA composition of yolk can be manipulated to varying degrees by changing the FA composition of the dietary lipid. Clear improvements in the determination coefficient were observed when the level of precursor in diet as second independent variable was added to the regression model (Table 6), thus suggesting that, in some cases [as in the use of sources rich in LNA (e.g., LO)], a limited but effective synthesis of n-3 LC-PUFA takes place from precursors. The content of AA, mainly generated through elongation and desaturation from LA, was found to be higher in

6 56 BAUCELLS ET AL. TABLE 5. Simple regression equations: Y = percentage fatty acid in egg; X = percentage fatty acid in feed Fatty acid Equation 1 R 2 RSD 2 P 3 C 18:3 n-3 y = 0.17x *** C 20:5 n-3 y = 0.08x *** C 22:5 n-3 y = 0.30x *** C 22:6 n-3 y = 0.42x *** C 18:2 n-6 y = 0.26x *** C 20:4 n-6 y = 2.16x *** PUFA y = 0.26x *** Total n-3 y = 0.21x *** Total n-6 y = 0.29x *** 1 n = 85 samples (five samples 17 treatments). 2 RSD = Residual SD. 3 Significance of the model: P < PUFA = Polyunsaturated fatty acids. almost all treatments when compared with the control diet. An abundance of AA and, especially, C 20:3 n-3 and DHA in the diet markedly decreased the -6 desaturation of LA, less markedly influenced the -5 desaturation of C 20:3 n-6 to AA, and did not alter the -6 desaturation of LNA (Bézard et al., 1994). This result explains why the biosynthesis of AA generated from LA was restrained in all of those treatments that included important amounts of FO. Thus, treatments supplied with SO (T10 to T13) resulted in a higher AA content in yolk than the other treatments, probably because of the high level in its precursor (LA) and the low amount in the LNA and all of the other n-3 FA content in these diets. In fact, as shown in Table 5, the dietary content in AA explained less than 50% of the variation of its content in the yolk lipid; these variables were inversely correlated. Thus, a higher AA dietary content does not guarantee a higher AA deposit in the yolk fat. The inclusion of its precursor (LA) as a second and positive independent variable made the percentage variation increase to 67%. It is pointed out that, contrary to what has been observed in the case of the n-3 LC-PUFA content (in which the inclusion of LA as an independent variable was reported as nonsig- FIGURE 1. Simple regression of the EPA (eicosapentanoic acid; C 20:5 n-3) content according to the percentage of fish oil (FO) in the fat added to the hen s feed. nificant; data not shown), the inclusion of LNA as a third variable in the regression model allowed for a better adjustment (R 2 = 0.80), because both variables were inversely related (Table 6). As we have previously mentioned, the fact that both LA and LNA are substrates of the -6 desaturase, allowing LNA to establish a competitive inhibition of the LA desaturation (Bézard et al., 1994), could easily explain these findings. Table 7 shows adjustments of all regressions to the model made according to the kind of fat used when replacing FO. A decrease in the content of DPA and DHA in egg when withdrawing FO from the hens diets is different according to the substitution that has been carried out. Therefore, differences depended on whether the replacement was performed with sources rich in LNA (LO and, to a lesser extent, RO) or poor in LNA (SO or T). For all TABLE 6. Multiple regression equations: Y = percentage fatty acid in egg; Xn = percentage fatty acid in feed Fatty acid Fatty acid dependent variable independent variable Equation 1 R 2 RSD 2 P 3 C 20:5 n-3 x1 = C 20:5 n-3 y = 0.09 x x *** x2 = C 18:3 n-3 C 22:5 n-3 x1 = C 22:5 n-3 y = 0.33 x x *** x2 = C 18:3 n-3 C 22:6 n-3 x1 = C 22:6 n-3 y = 0.44 x x *** x2 = C 18:3 n-3 C 20:4 n-6 x1 = C 20:4 n-6 y = 1.59 x x *** x2 = C 18:2 n-6 C 20:4 n-6 x1 = C 20:4 n-6 y = 1.95 x x x *** x2 = C 18:2 n-6 x3 = C 18:3 n-3 1 n = 85 samples (five samples 17 treatments). 2 RSD = Residual SD. 3 Significance of the model: ***P <

7 POLYUNSATURATED FATTY ACIDS IN EGGS 57 TABLE 7. Regression equations according to diet: Y = percentage fatty acid in egg; X = percentage fatty acid in feed Fatty acid Dietary fat 1 Equation R 2 RSD 2 P 3 C 20:5 n-3 LO 4 y = 0.069x *** SO, RO, and T 5 y = 0.083x *** C 22:5 n-3 LO 4 y = 0.148x ** SO, RO, and T 5 y = 0.339x *** C 22:6 n-3 LO 4 y = 0.284x *** RO 6 y = 0.360x *** SO and T 7 y = 0.439x *** C 20:4 n-6 LO 4 y = 0.598x *** SO 6 y = 3.438x *** RO and T 7 y = 2.111x *** 1 LO = Linseed oil; SO = sunflower oil; RO = rapeseed oil; T = tallow. 2 RSD = Residual SD. 3 Significance of the model: **P < 0.01; ***P < n = 25 samples (five samples five treatments). Diets containing LO included T2 [75% fish oil (FO), 25% LO], T3 (50% FO, 50% LO), T4 (25% FO, 75% LO), and T5 (100% LO). The control diet [T1 (100% FO)] was also included. 5 n = 65 samples (five samples 13 treatments). Diets containing SO included T10 (75% FO, 25% SO), T11 (50% FO, 50% SO), T12 (25% FO, 75% SO), and T13 (100% SO). Diets containing RO included T6 (75% FO, 25% RO), T7 (50% FO, 50% RO), T8 (25% FO, 75% RO), and T9 (100% RO). Diets containing T included T14 (75% FO, 25% T), T15 (50% FO, 50% T), T16 (25% FO, 75% T), and T17 (100% T). For all dietary fats, the control diet was also considered. 6 n = 25 samples (five samples five treatments). 7 n = 45 samples (five samples nine treatments). of the n-3 LC-PUFA, mainly present in diets with FO, the slope of their regression line is lower (EPA: P < 0.1; DPA and DHA: P < 0.001) in the case of all the treatments that included LO, which clearly indicates that the losses in the relative amount of EPA, DPA, and DHA in eggs are lower when the alternative fat source used to replace FO is LO compared with the other dietary fats used in the present experiment. The level of LNA is higher in LO treatments than in the rest of the fats used. That fact reinforces once again the theory of a slight de novo synthesis of these LC- PUFA from their precursors. The replacement of FO always resulted in a significant increase of the AA content, which was higher when the replacement was performed with a source rich in its precursor. In that case, the slope of the treatments with increasing amounts of SO was significantly higher than TABLE 8. Regression equations according to planning of replacement of fish oil in the added fat of the diet: Y = percentage fatty acid in egg; X = percentage fish oil in added fat (from 100 to 0%) Fatty acid Dietary fat 1 Equation 2 R 2 RSD 3 P 4 C 20:5 n-3 LO 5 (1) y = 0.007x *** SO, RO, and T 6 (2) y = 0.008x *** C 22:5 n-3 LO 5 (3) y = 0.002x ** SO, RO, and T 6 (4) y = 0.004x *** C 22:6 n-3 LO 5 (5) y = 0.017x *** SO, RO, and T 6 (6) y = 0.026x *** C 20:4 n-6 LO 5 (7) y = 0.003x *** RO and T 7 (8) y = 0.010x *** SO 8 (9) y = 0.016x *** 1 LO= Linseed oil; SO = sunflower oil; RO = rapeseed oil; T = tallow. 2 Equations (1) to (9); corresponding graphics presented in Figures 1 to 4. 3 RSD = Residual SD. 4 Significance of the model: **P < 0.01; ***P < n = 25 samples (five samples five treatments). Diets containing LO included T2 [75% fish oil (FO), 25% LO], T3 (50% FO, 50% LO), T4 (25% FO, 75% LO), and T5 (100% LO). The control diet [T1 (100% FO)] was also included. 6 n = 65 samples (five samples 13 treatments). Diets containing SO included T10 (75% FO, 25% SO), T11 (50% FO, 50% SO), T12 (25% FO, 75% SO), and T13 (100% SO). Diets containing RO included T6 (75% FO, 25% RO), T7 (50% FO, 50% RO), T8 (25% FO, 75% RO), and T9 (100% RO). Diets containing T included T14 (75% FO, 25% T), T15 (50% FO, 50% T), T16 (25% FO, 75% T), and T17 (100% T). For all dietary fats, the control diet was also considered. 7 n = 45 samples (five samples nine treatments). 8 n = 25 samples (five samples five treatments).

8 58 BAUCELLS ET AL. FIGURE 2. Simple regression of the DPA (docosapentanoic acid; C 22:5 n-3 ) content according to the percentage of fish oil (FO) in the fat added to the hen s feed. in the rest of the fats used in the present experiment (P < 0.001). Taking as a basis the results obtained in the present experiment, and knowing that the extensive use of FO in diets somehow implies many sensory problems in the eggs (Koehler and Bearse, 1975; Van Elswyk et al., 1992), limiting the inclusion of FO in diets and preserving the nutritional benefits it allows is a real need of the producer of poultry products. Table 8 and Figures 1 to 4 show the percentages of the main FA derived from LNA and LA in egg as related to FIGURE 4. Simple regression of the AA (arachidonic acid; C 20:4 n-6 ) content according to the percentage of fish oil (FO) in the fat added to the hen s feed. the percentage of FO included in the mixture of the fat added to the layers diets (from 0 to 100%). Taking DHA, for instance, to obtain 150 mg DHA per egg (3/4 dietary allowances established by the International Life Sciences Institute, 1995), under our conditions (4% added fat), a lower proportion of FO would be necessary in the fat added to the diet (37.5%) when it is mixed with LO compared with any other kind of mixture used (RO, SO, or T; 52.3%). In addition, the relationship of n-6 FA and n- 3 FA in egg (Figure 5) is always kept more constant, as well as smaller, when LO is used than when any other fat used to replace FO in the diet. FIGURE 3. Simple regression of the DHA (docosahexanoic acid; C 22:6 n-3) content according to the percentage fish oil (FO) in the added fat to the hen s feed. FIGURE 5. Simple regression of the n-6:n-3 relationship according to the percentage of fish oil (FO) in the fat added to the hen s feed.

9 POLYUNSATURATED FATTY ACIDS IN EGGS 59 The current work clearly shows that when LO is added to FO in the hens diets, it is slightly more efficient in maintaining the n-3 FA in egg compared with RO, SO, or T. The poultry producer is then in the fortunate position of being able to respond to changes in consumer demand as nutritional advice changes in response to research. ACKNOWLEDGMENTS The authors are grateful to Alfred Blanch for his valuable support and to Olga Baños for her skilled technical assistance throughout the experimental analyses. REFERENCES Aymond, W. M., and M. E. Van Elswyk, Yolk thiobarbituric acid reactive substances and n-3 fatty acids in response to whole and ground flaxseed. Poultry Sci. 74: Barlow, S. M., F.V.K. Young, and I. F. Duthie, Nutritional recommendations for omega-3 polyunsaturated fatty acids and the challenge to the food industry. Proc. Nutr. Soc. 49: Bézard, J., J. P. Blond, A. Bernard, and P. Clouet, The metabolism and availability of essential fatty acids in animal and human tissues. Reprod. Nutr. Dev. 34: Caston, L., and S. Leeson, Research note: Dietary flax and egg composition. Poultry Sci. 69: Cherian, G., and J. S. Sim, Effect of feeding full fat flax and canola seeds to laying hens on the fatty acid composition of eggs, embryos and newly hatched chicks. Poultry Sci. 70: Folch, J., M. Lees, and G. H. Sloane Stanley, A simple method for isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: Hargis, P. S., M. E. Van Elswyk, and B. M. Hargis, Dietary modification of yolk lipid with menhaden oil. Poultry Sci. 70: Hulan, H. W., R. G. Ackman, W.M.N. Ratnayake, and F. G. Proudfoot, Omega-3 fatty acid levels and general performance of commercial broilers fed practical levels of redfish meal. Poultry Sci. 68: International Life Sciences Institute, Dietary Fat: Some Aspects of Nutrition and Health and Production Development. ILSI Press, Washington, DC. Jiang, Z., D. U. Ahn, and J. S. Sim, Effects of feeding flax and two types of sunflower seeds on fatty acid compositions of yolk lipid classes. Poultry Sci. 70: Jiang, Z., and J. S. Sim, Consumption of n-3 polyunsaturated fatty acid-enriched eggs and changes in plasma lipids of human subjects. Nutrition 9: Koehler, H. H., and G. E. Bearse, Egg flavor quality as affected by fish meals or fish oils in laying rations. Poultry Sci. 54: Leskanich, C. O., and R. C. Noble, Manipulation of the n- 3 polyunsaturated fatty acid composition of avian eggs and meat. World s Poult. Sci. J. 53: López-Ferrer, S., M. D. Baucells, A. C. Barroeta, and M. A. Grashorn, n-3 Enrichment of chicken meat using fish oil: Alternative substitution with rapeseed and linseed oils. Poultry Sci. 78: Marshall, A. C., K. S. Kubena, K. R. Hinton, P. S. Hargis, and M. E. Van Elswyk, n-3 Fatty acid-enriched table eggs A survey of consumer acceptability. Poultry Sci. 73: Meluzzi, A., N. Tallarico, F. Sirri, and A. Franchini, Changes in the acidic profile of the egg as a response to the dietary administration of refined fish oils. Proc. ASPA Natl. Cong.: Morrison, W. R., and M. L. Smith, Preparation of fatty acid methyl esters and dimethylacetates from lipid with boron trifluoride methanol. J. Lipid. Res. 5: Nash, D. M., R.M.G. Hamilton, and H. W. Hulan, The effect of dietary herring meal on the omega-3 fatty acid content of plasma and egg yolk lipids of laying hens. Can. J. Anim. Sci. 75: National Research Council, Nutrient Requirements of Poultry. 9th rev. ed. National Academy Press, Washington, DC. Oh, S. Y., C.-H. H. Lin, J. Ryue, and D. E. Bell, Eggs enriched with omega-3 fatty acids as a wholesome food. J. Appl. Nutr. 46: Reinhart, G. A., Pages in: Review of Omega-3 Fatty Acid and Dietary Influences on Tissue Concentrations. Recent Advances in Canine and Feline Nutritional Research. Proc Iams Int. Nutr. Symp. Iams Co., ed. Sanders, T.A.B., Marine oils: Metabolic effects and role in human nutrition. Proc. Nutr. Soc. 52: SAS Institute, SAS User s Guide: Statistics. SAS Institute Inc., Cary, NC. Van Elswyk, M. E., A. R. Sams, and P. S. Hargis, Composition, functionality, and sensory evaluation of eggs from hens fed dietary menhaden oil. J. Food Sci. 57: , 349.