Omega-3 Fatty Acid Levels and General Performance of Commercial Broilers Fed Practical Levels of Redfish Meal1

Omega-3 Fatty Acid Levels and General Performance of Commercial Broilers Fed Practical Levels of Redfish Meal 1 H. W. HULAN,2 R. G. ACKMAN,3 W.M.N. RATNAYAKE,3 and F. G. PROUDFOOT4 Agriculture Canada, Research Station, Kentville, Nova Scotia, B4N 1J5 and Canadian Institute of Fisheries Technology, Technical University of Nova Scotia, Halifax, Nova Scotia, B3J 2X4 (Received for publication November 13, 1987)

1989 Poultry Science 6 8 : 1 5 3 - 1 6 2 INTRODUCTION

Omega-3 fatty acids, especially eicosapentaenoic acid (EPA, 20:5n-3), docosahexaenoic acid (DHA, 22:6n-3) and other n-3 polyunsaturated fatty acids (n-3 PUFA), are currently of great interest in the field of cardiovascular disease (see review by Herold and Kinsella, 1986). It was demonstrated that EPA and other n-3 PUFA levels in broilers are enhanced as a result of feeding diets supplemented with menhaden oil (Edwards and May, 1965; Marion and Woodroof, 1965; Miller et al., 1967a,b, 1969) or its ethyl esters and to a lesser extent with herring oil (Miller and Robisch, 1969). The effect of dietary fish meal on broiler tissue n-3 PUFA

'Contribution Number 2009. Agriculture Canada; present address: Dr. Howard W. Hulan, Professor and Head, Department of Poultry Science, Oregon State University, Dryden Hall 208, Corvallis, Oregon 97331-3402. 'Canadian Institute of Fisheries Technology. 4 Agriculture Canada. 2

has not been documented thoroughly (Dean et al., 1969). Recently, Hulan et al. (1988) demonstrated that broiler chickens fed a diet containing 5.0% fish meal have substantial amounts of EPA, DHA, and other n-3 PUFA deposited in the total carcass and edible meat lipids, and all n-3 PUFA are significantly increased by feeding higher levels of redfish meal (15 or 30%) or redfish oil (2 or 4%). Taste panel tests in an unpublished study found "off-flavors" in the meat of birds fed 15 or 30% redfish meal or 4.2% redfish oil. The flavors detected in these samples were not described as "fishy" or as objectionable. An experiment was undertaken to determine whether smaller and more practical amounts of redfish (Sebastes sp.) meal fed to commercial broiler chickens might increase omega-3 fatty acids in the edible meat lipid. MATERIALS AND METHODS

A total of 1,200 (600 of each sex) day-old Arbor Acre broiler chickens was randomly as-

153

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

ABSTRACT A total of 1,200 day-old Arbor Acre broiler chickens was randomly assigned to 12 pens (50 males and 50 females/pen) and divided into three blocks of four pens each. Each of four different diets was fed ad libitum to one pen of birds within each block to determine the effect of feeding practical levels of redfish meal (RFM) on performance and omega-3 fatty acid content of edible meat and skin lipids of broiler chickens. The four diets included (control) 0%, 4.0%, 8.0%, and 12.0% RFM. Feeding diets containing RFM had no effect on overall mortality or feed efficiency but resulted in decreased incidence of sudden death syndrome and lower body weight (P<.01) and feed consumption (P<.05). Additions of RFM to the diets resulted in a substantial dietary enrichment of omega-3 fatty acids (especially eicosapentaenoic acid, EPA or 20:5n-3, and docosahexaenoic acid, DHA or 22:6n-3). Analyses (wt/wt%) revealed that breast meat (less skin) was lower (P<.001) in lipid and triglyceride but higher in free cholesterol (P<.001) and phospholipid (P<.001) than thigh meat (less skin). Dietary treatment had no effect on carcass lipid content or composition. Breast meat lipid contained more (P<.001) omega-3 fatty acids (especially EPA and DHA), more docosapentaenoic acid, (DPA or 22:5n-3) and more total omega-3 polyunsaturated acids (n-3 PUFA) than thigh meat lipids. Feeding additional RFM resulted in an increased (P<.001) accumulation of EPA, DPA, DHA, and total n-3 PUFA primarily at the expense of two omega-6 fatty acids, linoleic (18:2n-6) and arachidonic acid (20:4n-6). It can be calculated from the data presented that the consumption of 100 g of chicken that has been fed 12.0% RFM would contribute approximately 197 mg of omega-3 fatty acids (EPA + DPA + DHA) in contrast with the 138 mg of omega-3 fatty acids which would be realized from the consumption of 100 g of white fish such as cod. (Key words: omega-3 fatty acids, broiler chicken, thigh, breast, fish meal)

HULANET As-

154

the product of one composite batch from a reduction plant and contained 60.4% CP and 10.1% lipid. Three cylindrical feeders (pan diameter of 36.7 cm) and one bell-shaped waterer were provided in each pen. Initial hover temperature was 34.5 C. The temperature was reduced gradually (1 C/day) so that at 14 days all birds were at 20.5 C, which temperature was maintained for the remainder of the test. All birds received incandescent white light at 20 lx for the first 48 h. Intermittent lighting (4 h light:2 h dark) commenced on Day 3 (end of first 48 h). Starting on Day 3, light was reduced in a linear fashion

TABLE 1. Composition of diets fed Redfish meal (%) Ingredient

0

12

8

4

KW'^i Ground yellow corn Ground wheat Soybean meal (49% CP) Redfish meal (RFM 60% CP) 1 Poultry grease Salt (NaCl) Ground limestone Dibasic calcium phosphate Vitamin-mineral premix 2 DL-Methionine L-Lysine HC1 Analysis Protein, % Gross energy, kcal/kg Calculated composition, % Arginine Histidine Isoleucine Leucine Lysine Methionine Methionine + cystine Phenylalanine Threonine Tryptophan Valine Calcium Phosphorus (total) Phosphorus (available) Sodium ME, kcal/kg 1

504.5 50.0 349.3 47.9

552.9 50.0 291.4 40.0 30.9

38.4 657.0 165.0 80.0 37.0

722.3 103.2 120.0 30.0

2.2 8.3

1.3 4.8

6.7

3.8

2.9

17.4 13.2 10.0

13.1 10.0

10.0

10.0

2.8 1.1

2.3

2.1

1.7

6.5

21.78 4,112

22.21 4,081

22.34 4,132

21.93 4,127

1.26

1.28

1.22

1.22

.50 .91

.49 .94

.44 .97

1.77 1.23

1.82 1.23

1.58 1.23

1.00 1.59 1.28

.60 .82 .97 .77 .23

.60 .83 .96 .78 .23

.60 .89 .95 .75 .26

.60 .90 .95 .76 .26

1.02

1.04

1.04

.95 .65 .40 .25

.95 .63 .40 .25

.94 .65 .42 .25

1.06 1.08

3,075

3,075

3,079

.43

.77 .56 .25

3,078

For amino acid composition of redfish meal, see Allan (1987).

'Supplied per kilogram of diet: 10,000 IU vitamin A, 2,000 ICU vitamin D 3 , 8 mg riboflavin, 15 mg d-calcium pantothenate, 15 Mg vitamin B 1 2 , 4 mg menadione sodium bisulfite, 35 mg niacin, 2 mg folic acid, 20 IU vitamin E, 1,000 mg choline, 300 Mg biotin, 5 mg pyridoxine-HCl, 3 mg thiamine, 187.5 mg amprolium, 200 mg ethoxyquin, 80 mg manganese, 70 mgzinc, 8 mg copper, 90 mg iron, 350 Mg iodine, 100 Mg selenium.

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

signed to 12 pens, each 2.5 m x 3.25 m. The sexes were divided equally so that each pen housed 50 males and 50 females. The twelve pens were divided into three blocks of four pens each. Each diet (Table 1) was fed ad libitum to one pen of birds within each block. Diet 1 served as the control. Diets 2, 3, and 4 contained 4.0, 8.0, and 12.0% redfish meal (RFM), respectively. The protein content of the RFM and diets (Table 1) was determined analytically (Association of Official Analytical Chemists, 1984). The gross energy of the diets was determined by adiabatic calorimetry (Parr Instrument Co., Moline, IL, Manual No. 153). The RFM was

OMEGA-3 FATTY ACIDS IN BROILER CHICKENS

were measured using a Perkin-Elmer LCI-100 (Perkin-Elmer Co., Ltd.) computing integrator. These areas were converted to weight percentages using a computer program that incorporated the appropriate FID response factors for all of the fatty acids (Ackman and Eaton, 1978). The total fatty acids of the diets were recovered by direct saponification with KOH/ ethanol in the presence of an added internal standard, heptadecanoic acid (Serdary Research Laboratories Inc., London, Ontario). The internal standard was used to calculate the fatty acid composition expressed as milligrams of fatty acid/100 g diet. About 20 g of the diet was accurately weighed into a 250-mL round-bottom flask. To this was added 20 mg of accurately weighed internal standard. This mixture was saponified by refluxing for 2 h with 2 mL of 50% KOH and 25 mL of 96% ethanol. The unsaponifiable matter was extracted before fatty acid recovery according to the American Oil Chemists Society (1983) method. Fatty acids were converted to methyl esters and analyzed by the GC as described above. An ANOVA was calculated on pen means for percentage of mortality (transformed to arcsin angles for analysis) and for body weight according to the split-plot design for complete blocks (Snedecor and Cochran, 1980). The sum of squares for diets (the main plot factor) was partitioned into a set of orthogonal contrasts: control vs. RFM, linear, and quadratic polynomial regression components on the nonzero levels of RFM in diet. The subplot factor (sex) interaction sums of squares with diets was partitioned by the same set of orthogonal contrasts. The feed consumption and feed efficiency data could not be identified for each sex because the sexes were intermingled but were calculated for each pen and analyzed as a complete randomized block. For the analysis of the composite meat samples, the full factorial ANOVA was calculated; the four-factor interaction mean square was used as the error term for testing the significance of the main effects and lower order interactions. Diet effects were partitioned by the same orthogonal contrasts as the performance traits. RESULTS AND DISCUSSION

The composition of the diets is given in Table 1. The RFM was incorporated into the diets primarily at the expense of soybean meal. The diets were essentially kept isoenergic (3,075

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

so that all birds were at .2-lx intensity by 14 days of age; this intensity was continued to the end of the trial (42 days). The diets were mixed weekly and stored at 0 C. During the hover period, diets were fed fresh daily, whereas for the remainder of the test, feed was added to the feeders each morning from cold storage. All diets were fed in mash form. Percentage mortality and live body weight were measured at 21 days and 42 days. Feed consumption was recorded, and feed efficiency at 21 days and 42 days was calculated as units live weight gain per unit feed consumed. One male and one female bird were randomly selected from each pen at 42 days, killed by exsanguination, and samples of one whole breast (both sides) and one whole thigh and skin were taken for subsequent lipid analyses. The skin was removed from the breast and thigh prior to analysis. Fat lumps attached to the skin or meat were removed and discarded before extraction to attain uniformity because this study was designed to ascertain the distribution of omega-3 fatty acids in edible portions, and excess adipose tissue was not considered as part of the edible portion. In each case the total sample (breast or thigh) was homogenized and a 100-g aliquot was taken for lipid extraction. Lipids were extracted from the tissues according to the method of Bligh and Dyer (1959). Lipid composition was determined of these extracts by Iatroscan thin layer chromatography/ flame ionization detection (TLC/FID, Ackman, 1981) on a "Chromarod-SII" as previously described (Hulan et al., 1984). Rods were developed in a tank containing hexane/diethyl ether/formic acid (97/3/1, vol/vol/vol). Area percentages of peaks were converted to weight percentages using appropriate conversion factors developed with authentic standards. A small portion of each extracted lipid sample was converted to methyl esters by transesterification with 5% BF3-MeOH (Morrison and Smith, 1964). The resulting fatty acid methyl esters were analyzed on a Sigma 3B gas chromatograph (GC; Perkin-Elmer Co., Ltd., Montreal, Quebec) equipped with a SUPELCOWAX-10 fused-silica capillary column (30 m X .25 mm i.d., 175 C/10 psig) and FID. The injector/detector temperatures were 200 C/ 260 C; the carrier gas was helium. Temperature programming was used: Temperature 1 (190 C) was held for 8 min; the temperature was then increased 3 C/min until Temperature 2 (240 C) was attained and held for 10 min. Peak areas

155

156

HULAN ETAL.

vitamin E should not have been a factor in the present study, as the diets contained more than twice the recommended level of vitamin E (National Research Council, 1984). Furthermore, none of the birds in the present research showed any evidence of the classical symptoms of vitamin E deficiency, indicating that the diet contained an ample amount of biologically active vitamin E. Intake and nutritional value of fish meals can be adversely affected by overheating (Scott et al., 1982). A high level of phospholipid (PL), which in fish is rich in polyunsaturated fatty acids, was found in the fish meal used in both the present and accompanying studies (Hulan et al., 1988). This evidence substantiates the effectiveness of the ethoxyquin added to the fish meal and to the diet (Table 1), but does not indicate if low palatability was responsible for the reduced performance of the RFM-fed birds. The major and selected nutritionally important fatty acids of chicken diets expressed as milligrams per 100 g of diet are given in Table 3. Diets 1 and 2 were slightly enriched with n-6 PUFA, especially that of 18:2n-6, which is derived mainly from the corn and soybean meal, but in part from the poultry grease. The control (Diet 1) showed nutritionally important amounts of n-3 PUFA (EPA or 20:5n-3; DPA or 22:5n-3; and DHA or 22:6n-3), also derived from poultry grease. The major n-3 PUFA, however, was 18:3n-3, derived mainly from soybean meal. The total of n-3 PUFA in the diets was comparatively low; Diet 4 showed the largest amounts of EPA and DHA in proportion to the fish meal in the diet. For total lipid (Table 4) there was a significant (P<.05) deviation from the linear response for diet X sample (meat). To assess the effect of RFM on the total lipid deposition in breast, thigh, and skin, logarithmic transformation was taken prior to analysis to stabilize the variance among the different deposition levels. The SE are expressed as a percentage of the means. Lipid content of the breast was .9%, of thigh meat, 2.2%; and of skin, 30.6%. The RFM fed at 8% generally gave higher lipid levels than when fed at 4 or 12%. Additions of RFM did not significantly change the lipid levels in breast meat or in thigh meat over those of their respective controls, but increased the level in the skin. Breast meat contained less (P<.001) total lipid and triglyceride (TG) but more (P<.001) free cholesterol (CH) and PL than did thigh meat,

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

kcal/kg AME) and isonitrogenous (22.0% CP) by making necessary adjustments in the amount of corn (wheat), poultry grease, and dibasic calcium phosphate resulting from the formulation of the diets by linear programming. Substitution of wheat for corn was confounded with the fish meal addition among the 4, 8, and 12% RFM levels. Throughout the experiment, the mortality (Table 2) of males was greater (21 days, P<.01; 42 days, P<.05) than that of females, due mainly to a greater incidence of sudden death syndrome (SDS), determined post-mortem (Hulan et al., 1980). This difference was not surprising, as it has been documented that the incidence of SDS is generally greater for male than for female broiler chickens (Brigden and Riddell, 1975; Hulan et al., 1980). The amount of RFM fed had no effect on the incidence of overall mortality, which confirmed the data of a previous study wherein large amounts of mis fish meal were fed (Hulan et al., 1988). However, the incidence of SDS decreased as the level of RFM in the diet was increased. The incidence (number) of birds succumbing to SDS on the different diets was: control (8), 4% RFM (6), 8% RFM (3), 12% RFM (2). As expected, at both 21 days and 42 days of age, males were heavier (P<.001) than females (Table 2). Mean live body weights of both sexes were reduced (21 days, P<.01; 42 days of age, P<.001) by the inclusion of RFM in the diet, confirming the data of aprevious study (Hulan etal., 1988). Addition of RFM to the diets resulted in reduced feed consumption (Table 2). This effect was linear (P<.01) at 42 days, but additions of RFM to the diet had no effect (P<.05) on feed efficiency at either 21 days or 42 days. The adverse effect on feed consumption observed in this study when practical amounts of RFM were fed is in agreement with results of a similar study (Hulan et al., 1988) that included larger amounts of RFM. Others (Waldroup et al., 1965; Proudfootefa/., 1971) reported decreased body weight and poorer feed efficiency when larger amounts of fish meal were fed to broiler chickens. The observed unpalatability of large amounts of RFM in this study is congruent with the observations of Opstvedt (1973a,b); Opstvedt postulated that when chicks were fed large amounts of fish lipids, or fish meals containing a high residual oil content, chicks' requirements for vitamin E may increase, resulting in poor performance. However, a deficiency in

1

_

• ** ***

*** *

*•* •• *• NS

1,950 1,864 1,813 1,803 14.8

705 684 684 672 5.9

NS NS

1,984 1,731 10.5

712 660 4.2

42 days

Live weight 21 days

Feed consumption

NS NS

*

NS

45.8 45.1 44.8 44.5 .29

(g)

1 to 21 days

Feed efficiency ratio = gain/feed. The males and females were intermingled, therefore these data could not be separ

_

NS NS NS

NS NS NS

*

NS

NS

**

21.7 (13.7) 17.1 (8.6) 22.1 (14.1) 23.7 (16.1) 1.88

(7.8) (4.6) (7.3) (7.1)

The sex X diet interaction was not significant.

•••P<.001.

**P<.01.

*P<.05.

2

_

Diet Contrasts within diets RFM Linear Deviations

Sex

21.2 (13.1) 16.8 (8.4) 1.33

(9.4) (4.1)

(%)

42 days

Mortality (Angle)

21 days

(%) Sex Males 17.9 Females 11.7 SEM 1.29 Diet Control 16.2 4.0% RFM 12.4 8.0% RFM 15.7 12.0% RFM 15.5 SEM 1.82 ANOVA, level of significance2

Source of variation

TABLE 2. Effect of feeding different levels ofredfish meal (RFM) on mortality, mean live body weight, feed consu

m http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

158

HULAN £TAL.

(Marion and Woodroof, 1965; Hulan et al., 1988). Of the three edible tissues studied, the PL content was greater (P<.001) in breast meat than in thigh meat; skin was devoid of this lipid class (Table 4). The higher proportion of TG in the lipid of skin tissue resulted in total fatty acids that were richer in monoenes (Table 5) and poorer in PUFA of both types, than the fatty acids of breast lipid. The level (wt/wt %) of the omega-3 fatty acids-EPA (20:5n-3), DPA (22:5n-3), and DHA (22:6n-3)-in lipids associated with breast meat was at least double that of thigh meat lipids (Table 6), and DHA exceeded EPA in both breast and thigh meat lipids by a factor of two or more. For human dietary considerations, it can be calculated from the data presented in Tables 4 to 6 that an average meal of 100 g (without skin) of chicken would contain 10.7% lipid and 46.2% PL (Table 4). Assuming 70% fatty acid and containing 1.51% EPA, 1.17% DPA, and 3.00% DHA (Table 6), chicken that had been fed 12.0% RFM would contribute 52.3 mg EPA, 40.5 mg DPA, and 103.8 mg of DHA for a total of 196.6 mg of these three omega-3 fatty acids. Cod flesh, for comparison, based on the data of Addison et al. (1968), contains .6% lipid, with about 70% PL estimated to be 70% fatty acids or 294 mg/100 (100 g X .006 lipid x .70 PL X .70 FA). The EPA and DHA are 17 and 30% of these fatty acids, so the total for these two n-3 PUFA will then be about 138 mg/100 g fish,

TABLE 3. Major and nutritionally important fatty acids of chicken diets Redfish meal (%) Fatty acid1

0

4

8

12

(mg/100 g of diet) Saturates

1,181.5

1,328.4

1,501.0

Monoenes

3,103.9

2,173.9

2,501.3

1,477.9 2,503.5

18:2n-6 18:3n-3 20:4n-6 20:5n-3 (EPA) 22:5n-3 (DPA) 22:6n-3 (DHA)

2,661.9 149.9 19.9 10.4 6.4 15.9

2,139.8 114.5 13.2 37.4 7.8 47.0

1,748.5 124.9 18.0 57.8 10.6 69.0

1,871.8 146.1 18.1 102.8 14.9 108.6

n-6 PUFA n-3 PUFA

2,713.3 196.2

2,173.3 221.7

1,785.1 284.8

1,910.7 404.8

PUFA Total FA, g/100 g diet

2,932.3 7.8^

2,405.2 5.91

2,092.4 6.09

2,335.8 6.32

'EPA = Eicosapentaenoic acid; DPA = docosapentaenoic acid; DHA = docosahexaenoic acid;PUFA = polyunsaturated fatty acids; FA = fatty acids.

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

confirming an earlier observation (Hulan etal., 1988). Lipids associated with the skin were made up entirely of TG. Lesser amounts (P<.001) of saturated fatty acids but greater amounts (P<.05) of omega-3 PUFA were found in female carcass lipids (Table 5). Compared with thigh meat lipids, breast meat lipids contained more (P< .001) total saturates, n-6 PUFA, n-3 PUFA, and total PUFA but less (P<.001) total monoenes. Similar relationships were found for thigh and skin. Additions of RFM to the diet resulted in a linear (P<.001) increase in total saturates, monoenes, and n-3 PUFA and a linear (P<.001) decrease in n-6 PUFA and total PUFA as previously reported (Hulan et al., 1988). Edible meat lipids of female broilers contained more (P<.05) 18:3n-3 and 22:5n-3 than edible meat lipids of males (Table 6). Lipids associated with breast meat contained less (P< .001) 18:3n-3, but more (P<.001) 20:4n-6, 20:5n-3, 22:5n-3, and 22:6n-3 than lipids associated with thigh meat. Similar differences were observed between thigh meat and skin. Additions of RFM to the diet linearly (P<.001) decreased 18:2n-6, 18:3n-3, and 20:4n-6, but linearly (P<.001) increased the omega-3 PUFA (20:5n-3, 22:5n-3, and 22:6n-3) of the edible meat lipids. In chickens, as well as in many other animals, long-chain PUFA are usually concentrated in the PL and to a lesser extent in neutral lipids

OMEGA-3 FATTY ACIDS IN BROILER CHICKENS

n-3 PUFA consumption. Not only did DPA in chicken meat lipids increase (P<.001) with increasing RFM in the diet, but the RFM dietary proportions relative to EPA or DHA (Table 3) were substantially augmented in broiler total fatty acids (Table 6). Chicken lipid fatty acids can be made to more closely resemble those of the seal-rich diet of the Eskimo than the fish lipids now in vogue (Ackman, 1988). The increase of n-3 PUFA in the tissues with increasing amounts of fish meal in the diet was at the expense of n-6 PUFA. It is apparent from the fatty acid profile in Table 6 that an inverse relation exists between the distribution of these two classes of PUFA; in general, the accumula-

TABLE 4. Effect of different dietary levels ofredfish meal (RFM) on the lipid content and composition of broiler chickens Lipid composition 1 Source of variation

Total lipid

TG

CH

PL

wt %) • Sex Male Female SEM Sample Breast Thigh Skin SEM Diet 0% RFM 4.0% RFM 8.0% RFM 12.0% RFM SEM ANOVA, level of significance Sex Sample Diet Contrasts within diets RFM Linear Deviations4

11.6 10.8 (2.8%)2

52.4 57.0 2.00

1.65 1.69 .134

45.7 41.1 1.95

.9 2.2 30.6 (3.5%)

36.6 72.8 (1O0.0)3 2.00

2.42 .92 tr .134

60.6 26.2 tr 1.95

10.2 11.1 12.8 10.7 (4.1%)

59.2 52.1 56.0 51.5 2.83

1.66 1.68 1.52 1.83 .190

39.1 46.0 42.0 46.2 2.76 NS

NS

NS

NS

• •*

• **

***

#•*

NS

NS

NS

NS

NS NS

NS NS NS

NS NS NS

NS NS NS

*

1 Trace amounts (<.05%) of cholesterol esters were found in all samples. TG = triglyceride; CH = free cholesterol; PL = phospholipids; tr = trace amount (<.05). 2

SEM expressed as percentage of mean.

3

Lipids associated with the skin consisted entirely of TG with only trace amounts (<.01) of diglycerides (DG), cholesterol esters (CE), CH, and PL detected; therefore the data for skin was not included in the analysis. 4 Principally due to the deviation effects of RFM on the diet X sample interaction (% SEM = 7.0%); for breast, results were .88, .95, .91, and .86; for thigh, results were 2.77, 1.95, 2.25, and 2.20; for skin, results were 26.87, 30.26, 35.10, and 29.20, for dietary RFM inclusion levels of 0, 4, 8, and 12%, respectively.

*P<.05. ***P<.001.

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

made up 50 mg EPA (294 x .17) and 88 mg DHA (294 X .30), as DPA is so low as not to be a factor (Ackman, 1980). In the definitive 20-yr study of Kromhout et al. (1985), eating fish three times a week or a mean intake of approximately 150 mg of EPA/day reduced cardiovascular mortality by 50%. Both thigh and breast meat lipids showed moderate amounts of DPA (Table 6). This fatty acid occupies an intermediate position in the interconversion of EPA and DHA, and is possibly a temporary storage form for these two fatty acids (von Schacky and Weber, 1985). Von Schacky and Weber suggested that DPA might be included in the potential health benefits of

159

HULAN ETAL.

160

tion of n-6 PUFA in the tissues reflects the levels of these PUFA in the diet. The all-vegetable-protein diet (Diet 1) contained the largest amount of n-6 PUFA (primarily 18:2n-6) whereas the high fish content diet showed smaller amounts of this fatty acid, with about the same 20:4n-6 content. Others (Mohrhauer and Holman, 1963; Miller et al., 1967a, 1969) have demonstrated that fatty acids of the n-3 family interfere with the synthesis of 20:4n-6 from 18:2n-6. The data presented here are consistent with the findings

of these investigators. From the data presented, it is concluded that the edible meat of chickens fed a diet containing 8% RFM contains amounts of n-3 PUFA, comparable to those provided by white fish muscle (Addison et al, 1968; Ackman, 1986). Therefore, consumers could benefit from chicken meat enriched in these acids via the addition of more fish meal to the diets. Chicken could become a supplemental or alternative source of n-3 PUFA. Compared with the well-publicized

Source of variation

Saturates

Monoenes

n-6 PUFA

n-3 PUFA

PUFA

(wt/wt % total FA) Sex Male Female SEM Sample Breast Thigh Skin SEM Diet 0% RFM 4.0% RFM 8.0% RFM 12.0% RFM SEM ANOVA, level of significance2 Sex Sample Diet Contrasts within diets RFM Linear

34.0 33.0 .18

42.5 42.9 .37

18.7 18.9 .22

4.8 5.3 .16

23.5 24.2 .34

34.3 33.4 32.8 .22

36.1 42.3 49.7 .45

21.3 19.7 15.4 .27

8.3 4.7 2.1 .20

29.7 24.4 17.5 .42

31.3 33.5 34.3 34.9 .25

41.5 41.7 43.4 43.8 .52

23.7 20.4 16.4 14.7 .31

3.5 4.5 5.6 6.5 .23

27.3 24.8 22.0 21.3 .48

*** *** ***

NS

NS

** « ** *

* * ** ** •

NS

*** ***

• **

*• ** *

* **

*** * **

*** #**

***

1

PUFA = Polyunsaturated fatty acids.

2

No significant deviations or interactions were found.

*P<.05. *'P<.01. *'*P<.001.

***

* ** ***

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

TABLE 5. Effect of different levels ofredfisb meal (RFM) on selected classes of fatty acids (FA) in broiler chickens1

OMEGA-3 FATTY ACIDS IN BROILER CHICKENS

161

TABLE 6. Effect of different dietary levels ofredfish meal (RFM) on the level of selected nutritionally important fatty acids (FA) of broiler chickens Source of variation

1

18:3n-3

20:4n-6

20:5n-3

22:5n-3

22:6n-3

14.4 15.0 .23

.61 .64 .010

2.01 1.97 .071

.98 1.05 .045

.78 1.00 .056

2.08 2.19 .083

15.0 15.0 14.2 .28

.43 .65 .80 .013

3.29 2.32 .35 .087

1.66 .87 .53 .056

1.60 .90 .16 .069

4.17 1.93 .29 .101

18.0 15.4 13.3 12.1 .32

.76 .59 .57 .55 .014

2.87 2.17 1.52 1.39 .100

.52 .91 1.13 1.51 .064

.64 .73 1.02 1.17 .080

1.16 1.87 2.51 3.00 .117

NS NS

NS

NS

** * * **

*** * **

** *** ** *

NS

** *

* • *• ***

** * ***

**• * **

*** ** *

*** * **

*** ***

*• * ***

*#* ***

No significant deviations or interactions were found.

*P<.05. •*P<01. *"P<.001.

fish and shellfish sources, chicken is popular, free of religious limitations, and already freely consumed world-wide. ACKNOWLEDGMENT

The authors would like to express appreciation to K. B. McRae, Regional Statistician, for advice on statistical design and analysis and to National Sea Products Limited, Lunenburg, Nova Scotia, for preparing and supplying the redfish meal.

REFERENCES Ackman, R. G., 1980. Fish Lipids, Part I. Pages 86-110 in: Advances in Fish Science and Technology. J. J. Connell, ed. Fishing News Books Ltd., Farnham, UK. Ackman, R. G., 1981. Problems in introducing new chromatographic techniques for lipid analyses. Chem. Industry, October, 17:715-722. Ackman, R. G., 1986. Perspectives on eicosapentaenoic acid (EPA), n-3 News 1:1-4. Ackman, R. G., 1988. Some possible effects on lipid biochemistry of difference in the distribution on glycerol of long chain (n-3) fatty acids in the fats of marine fish and marine mammals. Atherosclerosis

Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

Sex Male Female SEM Sample Breast Thigh Skin SEM Diet 0% RFM 4.0% RFM 8.0% RFM 12.0% RFM SEM ANOVA, level of significance1 Sex Sample Diet Contrasts within diets RFM Linear

18:2n-6

162

HULAN £TAZ,. Miller, D., E. H. Gruger, Jr., K. C. Leong, and G. M. Knobl, Jr., 1967a. Effect of refined menhaden oils on flavor and fatty acid composition of broiler flesh. J. Food Sci. 32:342-345. Miller, D., E. H. Gruger, Jr., K. C. Leong, and G. M. Knobl, Jr., 1967b. Dietary effect of menhaden oil ethyl esters on the fatty acid pattern of broiler muscle lipids. Poultry Sci. 46:438^144. Miller, D., K. C. Leong, and P. Smith, Jr., 1969. Effect of feeding and withdrawal of menhaden oil on the
Downloaded from http://ps.oxfordjournals.org/ at University of Otago Science Library on April 22, 2015

7:171-173. Ackman, R. G., andC. A. Eaton, 1978. Some contemporary applications of open-tubular gas-liquid chromatography in analyses of methyl esters of longer-chain fatty acids. Fette Seifen Anstrichm. 80:21-37. Addison, R. F., R. G. Ackman, and J. Hingley, 1968. Distribution of fatty acids in cod flesh lipids. J. Fish Res. Board Can. 25:2083-2090. Allen, R. D., 1987. Feedstuffs ingredient analysis table. Feedstuffs 59(#31):22-30. American Oil Chemists Society, 1983. Official and Tentative Methods. Tentative method Ca 66-53. 3rd ed. Am. Oil Chem. Soc., Champaign, IL. Association of Official Analytical Chemists, 1984. Methods of Analysis. 14th ed. Assoc. Off. Anal. Chem., Washington, DC. Bligh, E. G., and W. J. Dyer, 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911-917. Bridgen, J. L., and C. Riddell, 1975. A survey of mortality in four broiler flocks in Western Canada. Can. Vet. J. 16:194-200. Dean, P., W. F. Lamoreux, J. R. Aitken, and F. G. Proudfoot, 1969. Flavor associated with fish meal in diets fed to broiler chickens. Can. J. Anim. Sci. 49:11-15. Edwards, H. M., Jr., and K. N. May, 1965. Studies with menhaden oil in practical-type broiler rations. Poultry Sci. 44:685-689. Herald, P. M., and J. E. Kinsella, 1986. Fish oil consumption and decreased risk of cardiovascular disease: a comparison of findings from animal and human feeding trials. Am. J. Clin. Nutr. 43:566-598. Hulan, H. W., R. G. Ackman, W.M.N. Ratnayake, and F. G. Proudfoot, 1988. Omega-3 fatty acid levels and performance of broiler chickens fed redfish meal or redfish oil. Can. J. Anim. Sci. 68:543-547. Hulan, H. W., F. G. Proudfoot, and K. B. McRae, 1980. Effect of vitamins on the incidence of mortality and acute death syndrome ("flip-over") in broiler chickens. Poultry Sci. 59:927-931. Hulan, H. W., F. G. Proudfoot, and D. M. Nash, 1984. The effects of different dietary fat sources on general performance and carcass fatty acid composition and broiler chickens. Poultry Sci. 63:324-332. Kromhout, D., E. B. Bosschieter, and C. de L. Coulander, 1985. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. N. Engl. J. Med. 312:1205-1209. Marion, J. E., and J. G. Woodroof, 1965. Lipid fractions of chicken broiler tissues and their fatty acid composition. J. Food Sci. 30:38-43.