- Email: [email protected]
Effects of Dietary Fatty Acid Pattern on Melting Point and Composition of Adipose Tissues and Intramuscular Fat of Broiler Carcasses
METABOLISM AND NUTRITION Effects of Dietary Fatty Acid Pattern on Melting Point and Composition of Adipose Tissues and Intramuscular Fat of Broiler Carcasses C. HRDINKA,* W. ZOLLITSCH,+4 W. KNAUS,+ and F. LETTNER+ *Department of Pig and Poultry Science, Czech University of Agriculture in Prague, 165 21 Prague 6, Czech Republic, and ^Department of Animal Science, University of Agriculture, Forestry and Renewable Natural Resources, A-1180 Vienna, Austria
(Key words: fatty acids, melting point, adipose tissue, intramuscular fat, broiler) 1996 Poultry Science 75:208-215 dietary fat (Ajuyah et al, 1991; Chanmugam et al, 1992; Sheehy et al, 1993); however, the effect on the composition of the adipose tissues should be more pronounced (Yau et al, 1991). There is evidence that feeding certain dietary fatty acids will affect the levels of essential fatty acids in broiler meat. Therefore, chicken meat can be a source for these essential fatty acids for humans (Chanmugam et al, 1992; Hargis and Van Elswyk, 1993). As a factor of product quality, the melting point is an important characteristic of animal fat for both the food industry and the consumer. The melting process of triacylglycerols depends on their fatty acid composition. Because animal fats consist of mixtures of triacylglycerols, there is no true melting point (Enser, 1984a), and usually criteria like the slip point and the clarification point are used to characterize triacylglycerol consistency and melting point, respectively. It has been found in pigs that the melting (clarification) point depends on the dissolution of the more saturated glycerides and that the percentage of stearic acid will give the best prediction of the melting point. The correlation between the proportion of stearic acid and lipid consistency can be expected to be higher than 0.9 (Wood et al, 1978; Enser, 1984b).
INTRODUCTION Fatty acids are the main elements of triacylglycerols, which, by themselves, are the main components of neutral fats. Some polyunsaturated fatty acids (PUFA), such as linoleic and linolenic acid, are essential for humans and animals (Gurr, 1992) and therefore must be incorporated into the diet. Deficiencies in essential fatty acids will cause metabolic disorders (Whitehead, 1984). These essential fatty acids are important when discussing the use of various fat sources for animal nutrition. On the other hand, incorporation of fats in the diet may significantly affect several characteristics of product quality. Feeding different dietary lipids will result in different fatty acid patterns in the abdominal fat of broilers (Pinchasov and Nir, 1992; Scaife et al, 1994). Intramuscular fat is also expected to be affected by the
Received for publication April 10, 1995. Accepted for publication October 23, 1995. !To whom correspondence should be addressed: Werner Zollitsch, Department of Animal Science, Nutrition Group, Gregor Mendel Str. 33, A-1180 Vienna, Austria.
208
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
treatments caused significant changes in the fatty acid patterns for all analyzed tissues, although the differences were more pronounced for the adipose tissues. Overall, the adipose tissues contained more polyunsaturated and less saturated fatty acids than the fat from the breast and thigh portions. The melting point of the abdominal fat was significantly altered by the use of different dietary fats: RO gave a lower melting point than SO and FPl; the highest values were recorded for FP2. The data presented here indicate that the selection of certain dietary fat sources has a major impact on the composition and the melting point of broiler adipose tissues. The effect on the fatty acid composition of meat portions, however, is limited.
ABSTRACT Soybean oil (SO), rapeseed oil (RO), or two commercial fat products (FPl, FP2) were incorporated at 3.5% levels into four different corn-soybean meal mash broiler diets. Each of the four diets was fed to five replicates (pens) of broiler chickens for 42 d. After slaughtering the birds, samples of the abdominal fat, subcutaneous fat, and fat extracted from the thigh and the breast portion were collected from 16 birds per treatment. The fat samples were analyzed for their fatty acid composition using gas chromatography and the melting point of the abdominal fat was recorded. The results showed that the abdominal and subcutaneous fat had very similar fatty acid patterns and differed significantly from the composition of the fat extracted from breast and thigh. The different dietary
209
DIETARY FATTY ACIDS AND CARCASS FAT TABLE 1. Fatty acid composition of the dietary fat sources (percentage of total fatty acids) Fatty acids
Soybean oil
Rapeseed oil
Fat product 1
Fat product 2
Cl4:0 Cl6:0
<0.20 10.38 <0.20 3.71 21.62 50.56 11.75 1.31 14.09 21.62 62.31 5.96 4.22
<0.20 5.37 0.24 1.48 58.91 20.34 10.50 2.34 6.85 59.15 30.84 13.14 4.50
<0.20 10.89 <0.20 4.64 42.43 34.55 3.35 2.40 15.53 42.43 37.90 5.17 2.44
9.80 40.81 0.76 19.43 3.48 0.98 <0.20 22.63 70.04 4.24 0.98 0.07 0.01
Q&l Q&O ^18:1
Q&2 Q&3 Others SFA1 MUFA2 PUFA3 (MUFA+PUFA):SFA PUFA:SFA J
SFA = saturated fatty acids. MUFA = monounsaturated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
MATERIALS AND METHODS
extracts, 0.93% methionine plus cystine, 1.21% lysine, 1.10% calcium, 0.70% phosphorus, and 0.17% sodium. The chemical analyses of the diets showed no major differences in the calculated values and between dietary treatments. The diets were consumed ad libitum and water was available from automatic drinkers. The lighting regimen provided 24 h of continuous light/d. The fatty acid patterns of the four fat sources and of the experimental diets are given in Tables 1 and 2, respectively. Deriving from their fatty acid pattern, FP1 can be roughly characterized as a blend of animal and vegetable fat, and FP2 is a processed fat with an especially high content of myristic acid, palmitic acid, stearic acid, and behenic acid. Incorporation of the different fat sources resulted in a marked change in the fatty acid pattern of the experimental diets (Table 2).
Experimental Design and Diets
Data Collection
A total of 1,300 day-old Vedette broiler chicks of both sexes were randomly assigned to 20 litter pens (65 chickens per pen). The bird density was 22 birds per square meter. There were four dietary treatments with five replicates (pens) each. Four dietary treatments were achieved by using one out of four different fat sources for formulating the experimental diets: soybean oil (SO), rapeseed oil (RO), and two different commercial fat products (FP1,2 FP23) were incorporated into commonly used mash broiler rations, which consisted of 61.5% corn, 25.0% soybean meal, 4.0% fish meal, 2.0% meat and bone meal, 4.0% mineral and vitamin premix, and 3.5% of the respective fat source. The calculated nutrient contents per kilogram of the diets were 3,170 kcal ME, 21.5% CP, 6.6% ether
After 42 d, the birds were slaughtered and 16 carcasses per group (male:female = 1:1) were stored at a temperature of 4 C for about 16 h. Afterwards the carcasses were dissected and samples from the different portions were collected: breast meat plus attached skin, thigh muscles plus attached skin, abdominal fat, and subcutaneous fat from a position close to the coracoid bone. The samples were frozen and stored in a freezer at -20 C until further processing. The meat samples were homogenized using a blender with horizontal blades and dried at 100 C. The total lipids were extracted with ethyl ether using a Soxhlet apparatus. Samples of total lipids and of abdominal and subcutaneous fat were converted to methyl esters by transesterification with 5% BF3-MeOH4 (Morrison and Smith, 1964). The fatty acid methyl esters were analyzed using a gas chromatograph (Perkin-Elmer 8310),5 equipped with a packed column (Carbowax on Chromosorb, 80 to 100 mesh; 1/8 in x 2 m) 5 and flame ionization detection. The injector-column-detector temperatures were 240-190-250 C. Nitrogen was used as carrier gas. The melting point was measured for the abdominal fat samples only. The tissue samples were
2
Unifrutol®, Favorit Pro-Fett GmbH, A-1030 Vienna, Austria. Erbofat 90/10 Plus®, Dr. Ernst Boehlen Corp., CH-4922 Buetzlberg, Switzerland. 4MERCK 801663; Neuber, A-1060 Vienna, Austria. 5 Perkin Elmer GmbH, A-1101 Vienna, Austria. 3
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
However, data from other sources show that the sum of unsaturated fatty acids or the ratio of monounsaturated (MUFA) to saturated fatty acids (SFA) may give the best or at least a good prediction of the melting point (Elliot and Bowland, 1969; Lea et al, 1970). The above results have also been obtained with pigs. The objectives of the present study were to analyze the effects of different dietary fat sources on the composition of different adipose tissues (abdominal fat, subcutaneous fat) and of the fat from different meat portions (breast, thigh) of broiler carcasses. Also, the relationship between the melting point and the fatty acid pattern of the abdominal fat was quantified.
210
HRDINKA ET AL. TABLE 2. Fatty acid composition of the experimental diets (percentage of total fatty acids) Diet 1 Fatty acids Cl4:0
Q&0 ^16:1 ^18:0 Cl8:l
Q&2 ^18:3
Others SFA2 MUFA3 PUFA* (MUFA+PUFA):SFA PUFA:SFA
SO 0.44 15.55 0.40 3.83 25.76 46.89 5.72 0.58 19.82 26.16 52.60 3.97 2.65
RO 0.40 12.92 0.42 2.90 42.96 34.45 5.18 0.30 16.22 43.38 39.63 5.12 2.44
FP1 0.49 15.68 0.42 4.21 34.29 40.98 2.88 0.54 20.38 34.71 43.86 3.86 2.15
FP2 4.32 28.17 0.63 10.48 16.93 28.97 2.04 7.31 42.97 17.56 31.01 1.13 0.72
J
Diet containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2. SFA = saturated fatty acids. 3 MUFA = monounsaturated fatty acids. 4 PUFA = polyunsaturated fatty acids. 2
Statistical Analysis All data expressed as percentage (p) were transformed using the equation: p' = arcsin b/~(p/100)]. The relationship between the fatty acid content of the diets and the fatty acid pattern of the various tissues as well as the relations between the fatty acid content of the abdominal fat and its clarification point and slip point were analyzed using the GLM (General Linear Models) procedure of SAS® (SAS Institute, 1989). Differences between group least squares means were analyzed using the LSMEANS option and the BONFERRONI-HOLM test procedure (Holm, 1979). Statistical differences were considered to be significant when P < 0.05.
RESULTS AND DISCUSSION The fatty acid pattern of the diets typically reflected the composition of the different dietary fat sources (Table 1 and 2). The predominant SFA were palmitic and stearic acid, the predominant MUFA was the oleic acid, and linoleic and linolenic acids had the highest percentage among PUFA. The high amounts of arachidonic acid (C20:4) and behenic acid (C22:o) mainly account for the high percentage of fatty acids, which are summarized as "others" for FP2 and diet FP2.
Composition of Abdominal and Subcutaneous Fat The different dietary fats caused typical differences in the fatty acid pattern of the abdominal and subcutaneous fat (Table 3). This is in good agreement with data from various authors, who reported significant changes in the composition of the abdominal fat after feeding different dietary fat sources or different amounts of certain fatty acids (Pinchasov and Nir, 1992; Zollitsch et al, 1992; Scaife et al, 1994). As compared to the other diets, FP2 caused a significantly higher percentage of SFA (particularly myristic acid and palmitic acid) in the adipose tissues. Oleic acid is the major fatty acid of carcass fat (Ajuyah et al, 1991) and intramuscular fat (Olomu and Baracos, 1991). In this experiment, oleic acid was also the predominant fatty acid in all tissues (Table 3 and 5). With the exception of the linoleic acid percentage of the subcutaneous fat, there were significant differences between all dietary treatments in the content of linoleic acid, linolenic acid, and PUFA in the adipose tissues. These differences typically reflect the specific composition of the dietary fat sources of the different experimental diets (Table 1 and 2). Pinchasov and Nir (1992) also described a direct relationship between the dietary content of PUFA and its percentage in the adipose tissues.
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
melted at 100 C, placed in a U-shaped capillary glass tube, and stored below 10 C for 24 h. Slip point and clarification point were measured according to Naumann and Bassler (1983): the capillary glass tube with the fat sample was placed in a water bath with a stirring unit and a thermometer. The water bath was placed in front of a dark background and was well lighted. The initial water temperature was 18 C. The water was initially heated at a rate of 2 C/min; the heating rate was decreased to 1 C/ min once the difference between the expected melting point (from preliminary analysis) and the water bath was less than 10 C. The slip point was defined as the temperature at which the fat sample started to slip down in the capillary glass tube. The clarification point was defined as the temperature at which clouding of the sample was no longer observable. The measurements of the melting points were replicated three times. Values from the second and the third replicate were used for statistical analysis.
DIETARY FATTY ACIDS AND CARCASS FAT
211
TABLE 3. The effect of dietary fatty acids on the fatty acid pattern of subcutaneous and abdominal fat Diet 1 Fat
Fatty acids
SO
RO
Subcutaneous fat
^14:0
0.77b 24.94b 4.27b 5.36* 34.14" 26.00» 2.89" 0.52 31.38b 38.41= 28.88" 2.15b 0.93" 0.68= 23.35b 4.06b 5.23" 34.32d 27.67" 3.20" 0.56"b 29.57b 38.37= 30.87" 2.36" 1.06"
0.76b 23.92b 4.39b 4.69b 44.14" 17.68' 2.70" 0.66 29.61b 48.53" 20.37= 2.34" 0.69b 0.80b= 23.81b 4.72b 4.41b 44.51" 17.80= 2.68b 0.58"b 29.27b 49.23" 20.48= 2.42" 0.71b
RpciHiial
FP1
FP2
SD
P value
0.88b 24.37b 4.03b 5.76" 39.80b 22.07b 1.36b 0.66 31.28b 43.83b 23.43 b 2.17b 0.76b 0.89b 24.26b 4.06b 5.38" 39.74b 22.88b 1.40= 0.44b 30.82b 43.81b 24.28b 2.23" 0.80b
2.52" 30.67" 6.92" 5.47" 36.91= 14.37" 1.08b 0.69 38.92" 43.83b 15.46" 1.53= 0.40= 2.40" 29.30" 7.26" 5.37" 37.92= 14.62" 1.17d 0.76" 37.34" 45.18b 15.79" 1.64b 0.42=
0.24 1.88 0.74 0.68 2.06 2.08 0.38 0.42 2.21 2.33 2.25 0.21 0.10 0.19 2.32 0.94 0.70 1.79 2.37 0.31 0.26 2.63 2.30 2.57 0.28 0.14
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.651 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.013 <0.001 <0.001 <0.001 <0.001 <0.001
C°'i1 {•-)
Q&O ^16:1 ^-18:0 ^18:1
Q&2 ^-18:3
Others SFA2 MUFA 3 PUFA* (MUFA+PUFA):SFA PUFA:SFA Abdominal fat
Cl4:0 ^16:0
Q6:l Q&l Cl8:2 Q&3 Others SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA
"-"Least squares means in a row with no common superscript differ significantly according to Bonferroni-Holm test (P < 0.05). ^ i e t containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2. 2 SFA = saturated fatty acids. 3 MUFA = monounsaturated fatty acids. 4 PUFA = polyunsaturated fatty acids.
TABLE 4. Regression coefficients (for regression equation y = b 0 + bj x x + b 2 x x2) of the fatty acids of subcutaneous and abdominal fat on the dietary fatty acids Fat Subcutaneous fat
Fatty acid
b0
b]
C
0.607 17.681 -0.847 1.038 48.395 -4.617 -0.031 24.423 65.160 -4.082 1.317 0.227 0.604 18.056 -1.446 0.980 56.606 -7.094 1.320 24.189 72.051 -6.671 1.441 0.52
0.444 0.459 12.293 1.568 -1.109 0.652 0.515 0.335 -1.845 0.625 0.209 0.236 0.418 0.394 43.849 1.484 -1.368 0.736 -0.285 0.302 -2.286 0.707 0.206 -0.242
14:0 16:0 C 16:l C 18:0 C 18:l C 18:2 C 18:3
C
Abdominal fat
SFA1 MUFA2 PUFA3 (MUFA+PUFA):SFA PUFA:SFA Cl4:0 Q&0 Cl&l ^18:0 C 18:l Cl8:2 C 18:3
SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA J
SFA = saturated fatty acids. MUFA = monounsaturated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
b2
-0.109 0.024
0.034
-0.102 0.028 0.107 0.041
0.158
0.91 0.69 0.71 0.27 0.71 0.82 0.80 0.75 0.62 0.82 0.70 0.67 0.95 0.51 0.66 0.27 0.75 0.82 0.89 0.62 0.66 0.83 0.55 0.60
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
^18:0
212
HRDINKA ET AL. TABLE 5. The effect of dietary fatty acids on the fatty acid pattern of thigh and breast Diet*
(""arrays portion
Fatty acids
SO
RO
Rp^iHiial
FP1
FP2
SD
P value
1.10b 31.41b 3.65b 7.39b 41.57b 10.18 0.51 2.57 40.39b* 45.22b 10.69 1.42*b 0.28* 0.96b 28.41*b 3.68b 0.31 6.80* 41.72* 14.84*b 0.77 1.65 36.48*b 45.40b 15.61 1.73* 0.46
2.81* 34.01b 6.59* 6.00* 37.14* 10.07 0.44 1.69 43.04b 43.72b 10.51 1.27b 0.25*b 2.51* 30.77* 6.71* 0.26 5.92b 36.99b 13.59*b 0.96 1.20 39.45* 43.70* 14.55 1.48b 0.37
0.31 3.90 0.87 0.89 2.52 3.48 0.41 1.20 4.63 2.70 3.77 0.28 0.11 0.26 3.65 0.79 0.20 0.81 2.00 4.69 0.62 0.98 4.40 2.18 5.12 0.33 0.18
<0.001 <0.001 <0.001 <0.001 <0.001 0.015 0.143 0.049 <0.001 <0.001 0.023 <0.001 0.025 <0.001 0.034 <0.001 0.161 <0.001 <0.001 0.031 0.077 0.319 0.005 <0.001 0.072 0.004 0.262
/%) Thigh
Cl4:0 Cl6:0 ^16:1 ^18:0 ^18:1 ^18:2
Cl8:3 Others SFA* MUFA3 PUFA* (MUFA+PUFA):SFA PUFA:SFA Breast
Others SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA
6.96<> 37.49b 16.59a 1.15 1.48 38.21* 41.66*1 17.74 1.60b 0.49
0.99b 30.66b 4.10b 5.99* 46.12* 7.66 0.62 2.66 38.00* 50.22* 8.29 1.58* 0.23*b 0.85b 27.15b 4.32b 0.21 5.67b 46.40* 11.63b 1.32 1.72 33.88b 50.72* 12.95 1.92* 0.40
a-dLeast squares means in a row with no common superscript differ significantly according to Bonferroni-Holm test (P < 0.05). 1 Diet containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2. 2 SFA = saturated fatty acids. 3 MUFA = monounsaturated fatty acids. 4 PUFA = polyunsaturated fatty acids.
The lower differences of the oleic acid content in the adipose tissues than in the diets that were found in this experiment are confirmed by the data of Valencia et al. (1993), who reported minimal effects of feeding different fat sources on the MUFA content of the abdominal fat, whereas SFA and PUFA were affected more severely. In this experiment, the greatest differences between dietary treatments were found for the percentages of linoleic and linolenic acid (Table 3). The dependency between the dietary content and the percentage of the respective fatty acid in the adipose tissues can be described by regression equations and are quantified by the r2 values (Table 4). The closest relationship was found for myristic acid. The reason for this is the high utilization of myristic acid (highest among SFA and higher than for MUFA; Ketels and DeGroote, 1987), because the absorption from the digestive tract decreases with increasing chain length of the fatty acids (Freeman, 1984). In this experiment, the dietary content of the essential linoleic and linolenic acid highly affected the percentages of these fatty acids in the adipose tissues (Table 4). This was represented by a linear or quadratic regression and is in good agreement with the results published by Pin-
chasov and Nir (1992), who found both linear and quadratic relationships between the dietary amount of PUFA and the PUFA content in the abdominal and carcass fat. In contrast, Valenicia et al. (1993) reported linear relationships only. The composition of the dietary fat affects its absorption and utilization and there can be synergistic effects between different fatty acids (Freeman, 1984). These could be the reasons for the reported differences in the shape of the response lines (curves), because in this experiment as well as in the work of Pinchasov and Nir (1992), data from feeding different fat sources were combined for the statistical analysis of the effects of dietary fat composition, whereas Valencia et al. (1993) only analyzed the effect of a single fat source. There were no significant differences in the fatty acid composition between abdominal and subcutaneous fat for any of the dietary treatments. This is in good agreement with the results of previous work (Marion and Woodroof, 1963) and is of importance for the consumer demand for a desired fatty acid pattern of foodstuffs: the composition of the abdominal fat, from which samples can be taken easily, more accurately represents the fatty acid pattern of the subcutaneous fat that is consumed than does the abdominal fat pad.
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
^14:0 ^16:0 Ci&i ^17:0 '-18:0 ^18:1 Cl8:2 ^18:3
1.11b 38.05* 3.51b 8.34a 37.17* 6.82 0.29 2.80 48.24* 40.68* 7.12 1.02* 0.16b 0.96b 29.92* 4.17b 0.37
213
DIETARY FATTY ACIDS AND CARCASS FAT
Fat Composition of the Meat Portions
TABLE 6. Regression coefficients (for regression equation y = b 0 + b t x x + b 2 x x2) of the fatty acids from thigh and breast on the dietary fatty acids Carcass portion
Fatty acid
bo
bi
Thigh
C
0.867 31.822 -1.669 -0.490 41.84 13.045 -0.567 -3.818 59.038 14.597 1.628 0.008 0.739 25.659 -0.940 1.116 40.724 6.316 0.636 11.316 57.262 8.283 1.343 0.340
0.450 0.946 13.115 2.950 -0.547 -0.115 0.642 3.500 -1.367 -0.130 -0.399 0.444
14:0
Q&O Cl6:l ^18:0
Q&l Cl8:2 ^1831
SFA MUFA2 PUFA3 (MUFA+PUFA):SFA PUFA:SFA Breast
Cl4:0 Cl6:0 ^16:1 Cl8:0 Cl8:l Cl8:2 ^183
SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA
!SFA = saturated fatty acids. MUFA = monounsarurated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
0.410 0.188 12.106 2.029 -0.468 0.207 0.105 1.841 -1.219 0.166 0.970 0.045
b2
-0.222 0.015 -0.084 -0.056 0.027 0.076 -0.147
-0.150 0.014 -0.027 0.025
r* 0.86 0.14 0.68 0.42 0.68 0.04 <0.01 0.17 0.60 0.06 0.18 0.14 0.88 0.09 0.64 0.29 0.79 0.08 0.07 0.18 0.72 0.06 0.16 0.04
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
The different dietary fat sources also significantly influenced the percentage of fatty acids of the fat extracted from the breast and thigh portion (Table 5). In general, differences between dietary treatments, when represented as relative differences in the contents of SFA, MUFA, and PUFA, were more pronounced for the thigh portion than for the breast. This is in agreement with the data presented by Sheehy et al. (1993), who concluded that the fatty a d d composition of muscles tended to reflect the composition of the dietary lipids and that after feeding different fat sources, the changes in the fatty acid pattern were more pronounced for the thigh than for the breast. On the other hand, in this experiment the fatty acid pattern of the breast appeared to be more similar to the subcutaneous and abdominal fat than was the composition of the thigh. This could be explained by the fact that in this experiment, the breast and the thigh should represent the usually consumed portions and therefore were analyzed including the attached skin. Because the thigh portion contains more fat than the breast, it could be expected that the subcutaneous fat would influence the fatty a d d pattern of the breast muscle more severely than the thigh, therefore causing greater differences. However, the latter has not been the case (Table 5). Overall, the breast and thigh portion showed a similarity in their fatty acid pattern and clearly differed from the fat composition of the adipose tissues: for all fatty acids with significant differences between breast and abdominal fat, significant differences
between thigh and abdominal fat were also recorded. The content of MUFA (particularly palmitoleic a d d and oleic acid) was quite similar for the adipose tissues and the meat portions. Regarding the relationship between dietary fatty acids and the fat composition of the meat portions, the regression analysis showed that, similar to the adipose tissues, the closest relationship was recorded for myristic acid (Table 6). The r 2 values for the regression equations of dietary palmitoleic and oleic acid on the content of these fatty a d d s in the meat portions are within a medium range. Despite significant differences between dietary treatments in the percentage of a number of fatty acids in the meat portions, there-was n o close relationship between the dietary content of palmitic, linoleic, and linolenic acid and the percentage of these fatty acids in the meat. This was quantified by low r 2 values and could be due to the oxidation of fatty acids, which are functioning as a source of energy in tine muscle (Pethick et al., 1984). It was expected that the breast contained more SFA and less PUFA than the thigh (Ajuyah et al, 1991; Sheehy et al, 1993). However, this is in contrast to the data presented here (Table 5). A probable reason is that there were different tissues included in the samples: whereas in this work both the breast and the thigh portion included the attached subcutaneous fat, the latter has been removed in the experiments of Ajuyah et al. (1991) and Sheehy et al. (1993). When comparing the fatty acid pattern of the abdominal fat with the composition of the fat extracted from the meat portions, the main differences between these tissues
214
HRDINKA ET AL. TABLE 7. Effect of dietary treatment on the melting points of the abdominal fat tissue Diet 1 Melting point
SO
RO
Slip point Clarification point
24.9C 26.81*
22.4d 25.0=
FP1
FP2
Residual SD
P value
27.9b 28.2b
31.9' 34.0*
2.82 3.35
<0.001 <0.001
-(C)-
a-dMeans within rows with no common superscript differ significantly (P < 0.05). iDiet containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2.
Melting Point of Abdominal Fat The melting points (slip point and clarification point) of the abdominal fat are given in Table 7. The melting point is
dependent on the actual content of fatty acids in the triacylglycerols, which account for about 95% of the adipose tissues (Enser, 1984a). In this experiment, the differences in the fatty acid patterns of the abdominal fat that were caused by feeding different dietary fat sources resulted in significant differences in the slip point and slight differences in the clarification point. A difference of 9.5 C in the slip point (between FP2 and RO) will certainly have a major impact on the consistency of the abdominal fat pad. From the results in Table 8 it can be concluded that SFA and (MUFA+PUFA):SFA give the best estimation of both the slip point and the clarification point. Also, the percentage of stearic acid (Wood et ah, 1978; Enser, 1984b) or the MUFA:SFA ratio (Lea et al, 1970) may be the best predictors for the melting point of adipose tissue. However, in this experiment lower r2 values for the stearic acid than for several other criteria were found (Table 8). For the data presented here, r 2 values even for the criteria that give the best prediction of the melting point are only in a medium range of about 0.5. This result means that only 50% of the variance of the melting point can be explained by the variance of the respective character. In general, r2 values are slightly higher for the slip point than for the clarification point (Table 8). This higher r2 seems to be due to the greater probability for inaccuracies when determining the clarification point than when determining the slip point, which results in a greater variation of the data for the clarification point. In conclusion, the results of this study showed that on the one hand, abdominal fat and subcutaneous fat have an
TABLE 8. Effect of fatty acid content on the melting points of abdominal fat measured by P value and coefficient of determination Clarification point
Slip point 2
Fatty acids
P value
r
Cl8:0
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.24 0.53 0.19 0.42 0.35 0.53 0.30
SFAi Cl8:0 : Q8:l
SFA:C 18:1 SFA:MUFA2 (MUFA+PUFA):SFA3 PUFA:SFA
!SFA = saturated fatty acids. MUFA = monounsaturated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
P value
r2
<0.001 <0.001 0.003 <0.001 <0.001 <0.001 <0.001
0.17 0.48 0.13 0.35 0.28 0.47 0.31
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
occurred with regards to palmitic acid and PUFA. A reason for this result could be that the PUFA may have been metabolized at a high rate in the muscle tissue, whereas they have been stored as a source of energy in the adipose tissues. In contrast to this, the content of MUFA is quite similar for both the adipose tissues and the meat portions. In this experiment, there was also a trend towards a higher content of stearic acid in the thigh than in the breast portion, and the lowest values were recorded for the abdominal fat (Table 3 and 5). This can be explained by the higher content of phospholipids in the muscles (Salmon and Neil, 1973). The data presented here also show that for commonly consumed portions of the carcass (breast and thigh), the content of PUFA and the ratios (MUFA+PUFA):SFA and PUFA:SFA, which are frequently discussed as having a major impact on the nutritive quality of foodstuffs for human nutrition (Gurr, 1992), are influenced only to a rather small extent by the fatty acid content of the animal's diet. Differences between adipose and other tissues have been reported earlier (Marion and Edwards, 1963; Marion and Woodroof, 1963). The sex of the chickens was expected to have no effect on the fatty acid pattern of both the adipose tissues and the meat portions (Olomu and Baracos, 1991). This expectation was confirmed by the data from this experiment.
DIETARY FATTY ACIDS AND CARCASS FAT almost identical fatty acid pattern. On the other hand, composition of the adipose tissues differ significantly from the fat extracted from the meat portions. This is the case for the fatty acid pattern itself as well as for the observed effect of the dietary fat on the composition of the animal fat. Contrary to the composition of the adipose tissue, which closely follows the dietary lipids, the fat in the meat portions can be influenced by different dietary fat sources only to a small extent. This is especially the case for the content of PUFA, which is often discussed in connection with consumer demand for "healthy" foodstuffs.
ACKNOWLEDGMENTS
REFERENCES Ajuyah, A. O., K. H. Lee, R. T. Hardin, and J. S. Sim, 1991. Changes in the yield and in the fatty acid composition of whole carcass and selected meat portions of broiler chickens fed full-fat oil seeds. Poultry Sci. 70:2304-2314. Chanmugam, P., M. Boudreau, T. Boutte, R. S. Park, J. Hebert, L. Berrio, and D. H. Hwang, 1992. Incorporation of different types of n-3 fatty acids into tissue lipids of poultry. Poultry Sci. 71:516-521. Elliot, J. I., and J. P. Bowland, 1969. Correlation of melting point with the sum of the unsaturated fatty acids in samples of porcine depot fat. Can. J. Anim. Sci. 49:397-398. Enser, M., 1984a. The chemistry, biochemistry and nutritional importance of animal fats. Pages 23-51 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Enser, M., 1984b. The relationship between the composition and consistency of pig backfat. Pages 53-57 in: Fat Quality in Lean Pigs. A Workshop in the CEC Programme of Coordination of Research on Animal Husbandry, Brussels, Belgium. Freeman, C. P., 1984. The digestion, absorption and transport of fats—Non-ruminants. Pages 105-122 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Gurr, M. I., 1992. Role of Fats in Food and Nutrition. 2nd ed. Elsevier Science Publishers Ltd., Barking, UK. Hargis, P. S., and M. E. Van Elswyk, 1993. Manipulating the fatty acid composition of poultry meat and eggs for the health conscious consumer. World's Poultry Sci. J. 49: 251-264. Holm, S., 1979. A simple sequentially rejecrive multiple test procedure. Scand. J. Stat. 6:65-70. Ketels, E., and G. De Groote, 1987. Effect of fat source and level of fat inclusion on the utilization of fatty acids in broilers diets. Arch. Gefltigelkd. 51:127-132.
Lea, C. H., P.A.T. Swoboda, and D. P. Gatherum, 1970. A chemical study of soft fat in cross-bred pigs. J. Agric. Sci. 74:279-284. Marion, J. E., and H. M. Edwards, 1963. The effect of dietary fatty acid cholesterol on growth and fatty acid composition of the chicken. J. Nutr. 79:53-61. Marion, J. E., and J. G. Woodroof, 1963. The fatty acid composition of breast, thigh and skin tissue of chicken broilers as influenced by dietary fats. Poultry Sci. 42: 1202-1207. Morrison, W. R., and L. M. Smith, 1964. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron trifluorid—methanol. J. Lipid Res. 5:600-608. Naumann, K., and R. Bassler, 1983. Die chemische Untersuchung von Futtermitteln. Verlag J. NeumannNeudamm, Melsungen, Berlin, Germany. Olomu, J. M., and V. E. Baracos, 1991. Influence of dietary flaxseed oil on the performance, muscle protein deposition and fatty acid composition of broiler chicks. Poultry Sci. 70:1403-1411. Pethick, D. W., A. W. Bell, and E. F. Annison, 1984. Fats as energy sources in animal tissues. Pages 225-248 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Pinchasov, Y., and I. Nir, 1992. Effect of dietary polyunsaturated fatty acid concentration on performance, fat deposition and carcass fatty acid composition in broiler chickens. Poultry Sci. 71:1504-1512. Salmon, R. E., and J. B. O'Neil, 1973. The effect of the level and source and of a change of source of dietary fat on the fatty acid composition of the depot fat and the thigh and breast meat of turkeys as related to age. Poultry Sci. 52:302-314. SAS Institute, 1989. SAS/STAT® User's Guide. Release 6.03 Edition. SAS Institute Inc., Cary, NC. Scaife J. R., J. Moyo, M. Galbraith, M. Michie, and V. Campbell, 1994. Effect of different dietary supplemental fats and oils on the tissue fatty acid composition and growth of female broilers. Br. Poult. Sci. 35:107-118. Sheehy, P.J.A., P. A. Morrissey, and A. Flynn, 1993. Influence of heated vegetable oils and alphatocopherol acetate supplementation on alphatocopherol, fatty acids and lipid peroxidation in chicken muscle. Br. Poult. Sci. 34:367-381. Valencia, M. E., E. S. Watkins, A. L. Waldroup, P. W. Waldroup, and D. L. Fletcher, 1993. Utilization of crude and refined palm and palm kernel oils in broiler diets. Poultry Sci. 72:2200-2215. Whitehead, C. C, 1984. Essential fatty acids in poultry nutrition. Pages 153-166 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Wood, J. D., H. B. Enser, H.J.H. Mc Gie, W. C. Smith, J. P. Chadwick, M. Ellis, and R. Laird, 1978. Fatty acid composition of backfat in Large White pig selected for low backfat thickness. Meat Sci. 2:289-300. Yau, J. C, J. H. Denton, C. A. Bailey, and A. R. Sams, 1991. Customizing the fatty acid content of broiler tissues. Poultry Sci. 70:167-172. Zollitsch, W., W. Wetscherek, and F. Lettner, 1992. Use of rapeseed oil in a broiler diet. Arch. Gefltigelkd. 56: 182-186.
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
The authors thank the Austrian Science Foundation and the Austrian Federal Ministry of Science and Research (Oesterreichischer Akademischer Austauschdienst) for financial support of this research.
215
(Key words: fatty acids, melting point, adipose tissue, intramuscular fat, broiler) 1996 Poultry Science 75:208-215 dietary fat (Ajuyah et al, 1991; Chanmugam et al, 1992; Sheehy et al, 1993); however, the effect on the composition of the adipose tissues should be more pronounced (Yau et al, 1991). There is evidence that feeding certain dietary fatty acids will affect the levels of essential fatty acids in broiler meat. Therefore, chicken meat can be a source for these essential fatty acids for humans (Chanmugam et al, 1992; Hargis and Van Elswyk, 1993). As a factor of product quality, the melting point is an important characteristic of animal fat for both the food industry and the consumer. The melting process of triacylglycerols depends on their fatty acid composition. Because animal fats consist of mixtures of triacylglycerols, there is no true melting point (Enser, 1984a), and usually criteria like the slip point and the clarification point are used to characterize triacylglycerol consistency and melting point, respectively. It has been found in pigs that the melting (clarification) point depends on the dissolution of the more saturated glycerides and that the percentage of stearic acid will give the best prediction of the melting point. The correlation between the proportion of stearic acid and lipid consistency can be expected to be higher than 0.9 (Wood et al, 1978; Enser, 1984b).
INTRODUCTION Fatty acids are the main elements of triacylglycerols, which, by themselves, are the main components of neutral fats. Some polyunsaturated fatty acids (PUFA), such as linoleic and linolenic acid, are essential for humans and animals (Gurr, 1992) and therefore must be incorporated into the diet. Deficiencies in essential fatty acids will cause metabolic disorders (Whitehead, 1984). These essential fatty acids are important when discussing the use of various fat sources for animal nutrition. On the other hand, incorporation of fats in the diet may significantly affect several characteristics of product quality. Feeding different dietary lipids will result in different fatty acid patterns in the abdominal fat of broilers (Pinchasov and Nir, 1992; Scaife et al, 1994). Intramuscular fat is also expected to be affected by the
Received for publication April 10, 1995. Accepted for publication October 23, 1995. !To whom correspondence should be addressed: Werner Zollitsch, Department of Animal Science, Nutrition Group, Gregor Mendel Str. 33, A-1180 Vienna, Austria.
208
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
treatments caused significant changes in the fatty acid patterns for all analyzed tissues, although the differences were more pronounced for the adipose tissues. Overall, the adipose tissues contained more polyunsaturated and less saturated fatty acids than the fat from the breast and thigh portions. The melting point of the abdominal fat was significantly altered by the use of different dietary fats: RO gave a lower melting point than SO and FPl; the highest values were recorded for FP2. The data presented here indicate that the selection of certain dietary fat sources has a major impact on the composition and the melting point of broiler adipose tissues. The effect on the fatty acid composition of meat portions, however, is limited.
ABSTRACT Soybean oil (SO), rapeseed oil (RO), or two commercial fat products (FPl, FP2) were incorporated at 3.5% levels into four different corn-soybean meal mash broiler diets. Each of the four diets was fed to five replicates (pens) of broiler chickens for 42 d. After slaughtering the birds, samples of the abdominal fat, subcutaneous fat, and fat extracted from the thigh and the breast portion were collected from 16 birds per treatment. The fat samples were analyzed for their fatty acid composition using gas chromatography and the melting point of the abdominal fat was recorded. The results showed that the abdominal and subcutaneous fat had very similar fatty acid patterns and differed significantly from the composition of the fat extracted from breast and thigh. The different dietary
209
DIETARY FATTY ACIDS AND CARCASS FAT TABLE 1. Fatty acid composition of the dietary fat sources (percentage of total fatty acids) Fatty acids
Soybean oil
Rapeseed oil
Fat product 1
Fat product 2
Cl4:0 Cl6:0
<0.20 10.38 <0.20 3.71 21.62 50.56 11.75 1.31 14.09 21.62 62.31 5.96 4.22
<0.20 5.37 0.24 1.48 58.91 20.34 10.50 2.34 6.85 59.15 30.84 13.14 4.50
<0.20 10.89 <0.20 4.64 42.43 34.55 3.35 2.40 15.53 42.43 37.90 5.17 2.44
9.80 40.81 0.76 19.43 3.48 0.98 <0.20 22.63 70.04 4.24 0.98 0.07 0.01
Q&l Q&O ^18:1
Q&2 Q&3 Others SFA1 MUFA2 PUFA3 (MUFA+PUFA):SFA PUFA:SFA J
SFA = saturated fatty acids. MUFA = monounsaturated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
MATERIALS AND METHODS
extracts, 0.93% methionine plus cystine, 1.21% lysine, 1.10% calcium, 0.70% phosphorus, and 0.17% sodium. The chemical analyses of the diets showed no major differences in the calculated values and between dietary treatments. The diets were consumed ad libitum and water was available from automatic drinkers. The lighting regimen provided 24 h of continuous light/d. The fatty acid patterns of the four fat sources and of the experimental diets are given in Tables 1 and 2, respectively. Deriving from their fatty acid pattern, FP1 can be roughly characterized as a blend of animal and vegetable fat, and FP2 is a processed fat with an especially high content of myristic acid, palmitic acid, stearic acid, and behenic acid. Incorporation of the different fat sources resulted in a marked change in the fatty acid pattern of the experimental diets (Table 2).
Experimental Design and Diets
Data Collection
A total of 1,300 day-old Vedette broiler chicks of both sexes were randomly assigned to 20 litter pens (65 chickens per pen). The bird density was 22 birds per square meter. There were four dietary treatments with five replicates (pens) each. Four dietary treatments were achieved by using one out of four different fat sources for formulating the experimental diets: soybean oil (SO), rapeseed oil (RO), and two different commercial fat products (FP1,2 FP23) were incorporated into commonly used mash broiler rations, which consisted of 61.5% corn, 25.0% soybean meal, 4.0% fish meal, 2.0% meat and bone meal, 4.0% mineral and vitamin premix, and 3.5% of the respective fat source. The calculated nutrient contents per kilogram of the diets were 3,170 kcal ME, 21.5% CP, 6.6% ether
After 42 d, the birds were slaughtered and 16 carcasses per group (male:female = 1:1) were stored at a temperature of 4 C for about 16 h. Afterwards the carcasses were dissected and samples from the different portions were collected: breast meat plus attached skin, thigh muscles plus attached skin, abdominal fat, and subcutaneous fat from a position close to the coracoid bone. The samples were frozen and stored in a freezer at -20 C until further processing. The meat samples were homogenized using a blender with horizontal blades and dried at 100 C. The total lipids were extracted with ethyl ether using a Soxhlet apparatus. Samples of total lipids and of abdominal and subcutaneous fat were converted to methyl esters by transesterification with 5% BF3-MeOH4 (Morrison and Smith, 1964). The fatty acid methyl esters were analyzed using a gas chromatograph (Perkin-Elmer 8310),5 equipped with a packed column (Carbowax on Chromosorb, 80 to 100 mesh; 1/8 in x 2 m) 5 and flame ionization detection. The injector-column-detector temperatures were 240-190-250 C. Nitrogen was used as carrier gas. The melting point was measured for the abdominal fat samples only. The tissue samples were
2
Unifrutol®, Favorit Pro-Fett GmbH, A-1030 Vienna, Austria. Erbofat 90/10 Plus®, Dr. Ernst Boehlen Corp., CH-4922 Buetzlberg, Switzerland. 4MERCK 801663; Neuber, A-1060 Vienna, Austria. 5 Perkin Elmer GmbH, A-1101 Vienna, Austria. 3
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
However, data from other sources show that the sum of unsaturated fatty acids or the ratio of monounsaturated (MUFA) to saturated fatty acids (SFA) may give the best or at least a good prediction of the melting point (Elliot and Bowland, 1969; Lea et al, 1970). The above results have also been obtained with pigs. The objectives of the present study were to analyze the effects of different dietary fat sources on the composition of different adipose tissues (abdominal fat, subcutaneous fat) and of the fat from different meat portions (breast, thigh) of broiler carcasses. Also, the relationship between the melting point and the fatty acid pattern of the abdominal fat was quantified.
210
HRDINKA ET AL. TABLE 2. Fatty acid composition of the experimental diets (percentage of total fatty acids) Diet 1 Fatty acids Cl4:0
Q&0 ^16:1 ^18:0 Cl8:l
Q&2 ^18:3
Others SFA2 MUFA3 PUFA* (MUFA+PUFA):SFA PUFA:SFA
SO 0.44 15.55 0.40 3.83 25.76 46.89 5.72 0.58 19.82 26.16 52.60 3.97 2.65
RO 0.40 12.92 0.42 2.90 42.96 34.45 5.18 0.30 16.22 43.38 39.63 5.12 2.44
FP1 0.49 15.68 0.42 4.21 34.29 40.98 2.88 0.54 20.38 34.71 43.86 3.86 2.15
FP2 4.32 28.17 0.63 10.48 16.93 28.97 2.04 7.31 42.97 17.56 31.01 1.13 0.72
J
Diet containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2. SFA = saturated fatty acids. 3 MUFA = monounsaturated fatty acids. 4 PUFA = polyunsaturated fatty acids. 2
Statistical Analysis All data expressed as percentage (p) were transformed using the equation: p' = arcsin b/~(p/100)]. The relationship between the fatty acid content of the diets and the fatty acid pattern of the various tissues as well as the relations between the fatty acid content of the abdominal fat and its clarification point and slip point were analyzed using the GLM (General Linear Models) procedure of SAS® (SAS Institute, 1989). Differences between group least squares means were analyzed using the LSMEANS option and the BONFERRONI-HOLM test procedure (Holm, 1979). Statistical differences were considered to be significant when P < 0.05.
RESULTS AND DISCUSSION The fatty acid pattern of the diets typically reflected the composition of the different dietary fat sources (Table 1 and 2). The predominant SFA were palmitic and stearic acid, the predominant MUFA was the oleic acid, and linoleic and linolenic acids had the highest percentage among PUFA. The high amounts of arachidonic acid (C20:4) and behenic acid (C22:o) mainly account for the high percentage of fatty acids, which are summarized as "others" for FP2 and diet FP2.
Composition of Abdominal and Subcutaneous Fat The different dietary fats caused typical differences in the fatty acid pattern of the abdominal and subcutaneous fat (Table 3). This is in good agreement with data from various authors, who reported significant changes in the composition of the abdominal fat after feeding different dietary fat sources or different amounts of certain fatty acids (Pinchasov and Nir, 1992; Zollitsch et al, 1992; Scaife et al, 1994). As compared to the other diets, FP2 caused a significantly higher percentage of SFA (particularly myristic acid and palmitic acid) in the adipose tissues. Oleic acid is the major fatty acid of carcass fat (Ajuyah et al, 1991) and intramuscular fat (Olomu and Baracos, 1991). In this experiment, oleic acid was also the predominant fatty acid in all tissues (Table 3 and 5). With the exception of the linoleic acid percentage of the subcutaneous fat, there were significant differences between all dietary treatments in the content of linoleic acid, linolenic acid, and PUFA in the adipose tissues. These differences typically reflect the specific composition of the dietary fat sources of the different experimental diets (Table 1 and 2). Pinchasov and Nir (1992) also described a direct relationship between the dietary content of PUFA and its percentage in the adipose tissues.
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
melted at 100 C, placed in a U-shaped capillary glass tube, and stored below 10 C for 24 h. Slip point and clarification point were measured according to Naumann and Bassler (1983): the capillary glass tube with the fat sample was placed in a water bath with a stirring unit and a thermometer. The water bath was placed in front of a dark background and was well lighted. The initial water temperature was 18 C. The water was initially heated at a rate of 2 C/min; the heating rate was decreased to 1 C/ min once the difference between the expected melting point (from preliminary analysis) and the water bath was less than 10 C. The slip point was defined as the temperature at which the fat sample started to slip down in the capillary glass tube. The clarification point was defined as the temperature at which clouding of the sample was no longer observable. The measurements of the melting points were replicated three times. Values from the second and the third replicate were used for statistical analysis.
DIETARY FATTY ACIDS AND CARCASS FAT
211
TABLE 3. The effect of dietary fatty acids on the fatty acid pattern of subcutaneous and abdominal fat Diet 1 Fat
Fatty acids
SO
RO
Subcutaneous fat
^14:0
0.77b 24.94b 4.27b 5.36* 34.14" 26.00» 2.89" 0.52 31.38b 38.41= 28.88" 2.15b 0.93" 0.68= 23.35b 4.06b 5.23" 34.32d 27.67" 3.20" 0.56"b 29.57b 38.37= 30.87" 2.36" 1.06"
0.76b 23.92b 4.39b 4.69b 44.14" 17.68' 2.70" 0.66 29.61b 48.53" 20.37= 2.34" 0.69b 0.80b= 23.81b 4.72b 4.41b 44.51" 17.80= 2.68b 0.58"b 29.27b 49.23" 20.48= 2.42" 0.71b
RpciHiial
FP1
FP2
SD
P value
0.88b 24.37b 4.03b 5.76" 39.80b 22.07b 1.36b 0.66 31.28b 43.83b 23.43 b 2.17b 0.76b 0.89b 24.26b 4.06b 5.38" 39.74b 22.88b 1.40= 0.44b 30.82b 43.81b 24.28b 2.23" 0.80b
2.52" 30.67" 6.92" 5.47" 36.91= 14.37" 1.08b 0.69 38.92" 43.83b 15.46" 1.53= 0.40= 2.40" 29.30" 7.26" 5.37" 37.92= 14.62" 1.17d 0.76" 37.34" 45.18b 15.79" 1.64b 0.42=
0.24 1.88 0.74 0.68 2.06 2.08 0.38 0.42 2.21 2.33 2.25 0.21 0.10 0.19 2.32 0.94 0.70 1.79 2.37 0.31 0.26 2.63 2.30 2.57 0.28 0.14
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.651 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.013 <0.001 <0.001 <0.001 <0.001 <0.001
C°'i1 {•-)
Q&O ^16:1 ^-18:0 ^18:1
Q&2 ^-18:3
Others SFA2 MUFA 3 PUFA* (MUFA+PUFA):SFA PUFA:SFA Abdominal fat
Cl4:0 ^16:0
Q6:l Q&l Cl8:2 Q&3 Others SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA
"-"Least squares means in a row with no common superscript differ significantly according to Bonferroni-Holm test (P < 0.05). ^ i e t containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2. 2 SFA = saturated fatty acids. 3 MUFA = monounsaturated fatty acids. 4 PUFA = polyunsaturated fatty acids.
TABLE 4. Regression coefficients (for regression equation y = b 0 + bj x x + b 2 x x2) of the fatty acids of subcutaneous and abdominal fat on the dietary fatty acids Fat Subcutaneous fat
Fatty acid
b0
b]
C
0.607 17.681 -0.847 1.038 48.395 -4.617 -0.031 24.423 65.160 -4.082 1.317 0.227 0.604 18.056 -1.446 0.980 56.606 -7.094 1.320 24.189 72.051 -6.671 1.441 0.52
0.444 0.459 12.293 1.568 -1.109 0.652 0.515 0.335 -1.845 0.625 0.209 0.236 0.418 0.394 43.849 1.484 -1.368 0.736 -0.285 0.302 -2.286 0.707 0.206 -0.242
14:0 16:0 C 16:l C 18:0 C 18:l C 18:2 C 18:3
C
Abdominal fat
SFA1 MUFA2 PUFA3 (MUFA+PUFA):SFA PUFA:SFA Cl4:0 Q&0 Cl&l ^18:0 C 18:l Cl8:2 C 18:3
SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA J
SFA = saturated fatty acids. MUFA = monounsaturated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
b2
-0.109 0.024
0.034
-0.102 0.028 0.107 0.041
0.158
0.91 0.69 0.71 0.27 0.71 0.82 0.80 0.75 0.62 0.82 0.70 0.67 0.95 0.51 0.66 0.27 0.75 0.82 0.89 0.62 0.66 0.83 0.55 0.60
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
^18:0
212
HRDINKA ET AL. TABLE 5. The effect of dietary fatty acids on the fatty acid pattern of thigh and breast Diet*
(""arrays portion
Fatty acids
SO
RO
Rp^iHiial
FP1
FP2
SD
P value
1.10b 31.41b 3.65b 7.39b 41.57b 10.18 0.51 2.57 40.39b* 45.22b 10.69 1.42*b 0.28* 0.96b 28.41*b 3.68b 0.31 6.80* 41.72* 14.84*b 0.77 1.65 36.48*b 45.40b 15.61 1.73* 0.46
2.81* 34.01b 6.59* 6.00* 37.14* 10.07 0.44 1.69 43.04b 43.72b 10.51 1.27b 0.25*b 2.51* 30.77* 6.71* 0.26 5.92b 36.99b 13.59*b 0.96 1.20 39.45* 43.70* 14.55 1.48b 0.37
0.31 3.90 0.87 0.89 2.52 3.48 0.41 1.20 4.63 2.70 3.77 0.28 0.11 0.26 3.65 0.79 0.20 0.81 2.00 4.69 0.62 0.98 4.40 2.18 5.12 0.33 0.18
<0.001 <0.001 <0.001 <0.001 <0.001 0.015 0.143 0.049 <0.001 <0.001 0.023 <0.001 0.025 <0.001 0.034 <0.001 0.161 <0.001 <0.001 0.031 0.077 0.319 0.005 <0.001 0.072 0.004 0.262
/%) Thigh
Cl4:0 Cl6:0 ^16:1 ^18:0 ^18:1 ^18:2
Cl8:3 Others SFA* MUFA3 PUFA* (MUFA+PUFA):SFA PUFA:SFA Breast
Others SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA
6.96<> 37.49b 16.59a 1.15 1.48 38.21* 41.66*1 17.74 1.60b 0.49
0.99b 30.66b 4.10b 5.99* 46.12* 7.66 0.62 2.66 38.00* 50.22* 8.29 1.58* 0.23*b 0.85b 27.15b 4.32b 0.21 5.67b 46.40* 11.63b 1.32 1.72 33.88b 50.72* 12.95 1.92* 0.40
a-dLeast squares means in a row with no common superscript differ significantly according to Bonferroni-Holm test (P < 0.05). 1 Diet containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2. 2 SFA = saturated fatty acids. 3 MUFA = monounsaturated fatty acids. 4 PUFA = polyunsaturated fatty acids.
The lower differences of the oleic acid content in the adipose tissues than in the diets that were found in this experiment are confirmed by the data of Valencia et al. (1993), who reported minimal effects of feeding different fat sources on the MUFA content of the abdominal fat, whereas SFA and PUFA were affected more severely. In this experiment, the greatest differences between dietary treatments were found for the percentages of linoleic and linolenic acid (Table 3). The dependency between the dietary content and the percentage of the respective fatty acid in the adipose tissues can be described by regression equations and are quantified by the r2 values (Table 4). The closest relationship was found for myristic acid. The reason for this is the high utilization of myristic acid (highest among SFA and higher than for MUFA; Ketels and DeGroote, 1987), because the absorption from the digestive tract decreases with increasing chain length of the fatty acids (Freeman, 1984). In this experiment, the dietary content of the essential linoleic and linolenic acid highly affected the percentages of these fatty acids in the adipose tissues (Table 4). This was represented by a linear or quadratic regression and is in good agreement with the results published by Pin-
chasov and Nir (1992), who found both linear and quadratic relationships between the dietary amount of PUFA and the PUFA content in the abdominal and carcass fat. In contrast, Valenicia et al. (1993) reported linear relationships only. The composition of the dietary fat affects its absorption and utilization and there can be synergistic effects between different fatty acids (Freeman, 1984). These could be the reasons for the reported differences in the shape of the response lines (curves), because in this experiment as well as in the work of Pinchasov and Nir (1992), data from feeding different fat sources were combined for the statistical analysis of the effects of dietary fat composition, whereas Valencia et al. (1993) only analyzed the effect of a single fat source. There were no significant differences in the fatty acid composition between abdominal and subcutaneous fat for any of the dietary treatments. This is in good agreement with the results of previous work (Marion and Woodroof, 1963) and is of importance for the consumer demand for a desired fatty acid pattern of foodstuffs: the composition of the abdominal fat, from which samples can be taken easily, more accurately represents the fatty acid pattern of the subcutaneous fat that is consumed than does the abdominal fat pad.
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
^14:0 ^16:0 Ci&i ^17:0 '-18:0 ^18:1 Cl8:2 ^18:3
1.11b 38.05* 3.51b 8.34a 37.17* 6.82 0.29 2.80 48.24* 40.68* 7.12 1.02* 0.16b 0.96b 29.92* 4.17b 0.37
213
DIETARY FATTY ACIDS AND CARCASS FAT
Fat Composition of the Meat Portions
TABLE 6. Regression coefficients (for regression equation y = b 0 + b t x x + b 2 x x2) of the fatty acids from thigh and breast on the dietary fatty acids Carcass portion
Fatty acid
bo
bi
Thigh
C
0.867 31.822 -1.669 -0.490 41.84 13.045 -0.567 -3.818 59.038 14.597 1.628 0.008 0.739 25.659 -0.940 1.116 40.724 6.316 0.636 11.316 57.262 8.283 1.343 0.340
0.450 0.946 13.115 2.950 -0.547 -0.115 0.642 3.500 -1.367 -0.130 -0.399 0.444
14:0
Q&O Cl6:l ^18:0
Q&l Cl8:2 ^1831
SFA MUFA2 PUFA3 (MUFA+PUFA):SFA PUFA:SFA Breast
Cl4:0 Cl6:0 ^16:1 Cl8:0 Cl8:l Cl8:2 ^183
SFA MUFA PUFA (MUFA+PUFA):SFA PUFA:SFA
!SFA = saturated fatty acids. MUFA = monounsarurated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
0.410 0.188 12.106 2.029 -0.468 0.207 0.105 1.841 -1.219 0.166 0.970 0.045
b2
-0.222 0.015 -0.084 -0.056 0.027 0.076 -0.147
-0.150 0.014 -0.027 0.025
r* 0.86 0.14 0.68 0.42 0.68 0.04 <0.01 0.17 0.60 0.06 0.18 0.14 0.88 0.09 0.64 0.29 0.79 0.08 0.07 0.18 0.72 0.06 0.16 0.04
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
The different dietary fat sources also significantly influenced the percentage of fatty acids of the fat extracted from the breast and thigh portion (Table 5). In general, differences between dietary treatments, when represented as relative differences in the contents of SFA, MUFA, and PUFA, were more pronounced for the thigh portion than for the breast. This is in agreement with the data presented by Sheehy et al. (1993), who concluded that the fatty a d d composition of muscles tended to reflect the composition of the dietary lipids and that after feeding different fat sources, the changes in the fatty acid pattern were more pronounced for the thigh than for the breast. On the other hand, in this experiment the fatty acid pattern of the breast appeared to be more similar to the subcutaneous and abdominal fat than was the composition of the thigh. This could be explained by the fact that in this experiment, the breast and the thigh should represent the usually consumed portions and therefore were analyzed including the attached skin. Because the thigh portion contains more fat than the breast, it could be expected that the subcutaneous fat would influence the fatty a d d pattern of the breast muscle more severely than the thigh, therefore causing greater differences. However, the latter has not been the case (Table 5). Overall, the breast and thigh portion showed a similarity in their fatty acid pattern and clearly differed from the fat composition of the adipose tissues: for all fatty acids with significant differences between breast and abdominal fat, significant differences
between thigh and abdominal fat were also recorded. The content of MUFA (particularly palmitoleic a d d and oleic acid) was quite similar for the adipose tissues and the meat portions. Regarding the relationship between dietary fatty acids and the fat composition of the meat portions, the regression analysis showed that, similar to the adipose tissues, the closest relationship was recorded for myristic acid (Table 6). The r 2 values for the regression equations of dietary palmitoleic and oleic acid on the content of these fatty a d d s in the meat portions are within a medium range. Despite significant differences between dietary treatments in the percentage of a number of fatty acids in the meat portions, there-was n o close relationship between the dietary content of palmitic, linoleic, and linolenic acid and the percentage of these fatty acids in the meat. This was quantified by low r 2 values and could be due to the oxidation of fatty acids, which are functioning as a source of energy in tine muscle (Pethick et al., 1984). It was expected that the breast contained more SFA and less PUFA than the thigh (Ajuyah et al, 1991; Sheehy et al, 1993). However, this is in contrast to the data presented here (Table 5). A probable reason is that there were different tissues included in the samples: whereas in this work both the breast and the thigh portion included the attached subcutaneous fat, the latter has been removed in the experiments of Ajuyah et al. (1991) and Sheehy et al. (1993). When comparing the fatty acid pattern of the abdominal fat with the composition of the fat extracted from the meat portions, the main differences between these tissues
214
HRDINKA ET AL. TABLE 7. Effect of dietary treatment on the melting points of the abdominal fat tissue Diet 1 Melting point
SO
RO
Slip point Clarification point
24.9C 26.81*
22.4d 25.0=
FP1
FP2
Residual SD
P value
27.9b 28.2b
31.9' 34.0*
2.82 3.35
<0.001 <0.001
-(C)-
a-dMeans within rows with no common superscript differ significantly (P < 0.05). iDiet containing fat source: SO = soybean oil, RO = rapeseed oil, FP1, FP2 = fat product 1, 2.
Melting Point of Abdominal Fat The melting points (slip point and clarification point) of the abdominal fat are given in Table 7. The melting point is
dependent on the actual content of fatty acids in the triacylglycerols, which account for about 95% of the adipose tissues (Enser, 1984a). In this experiment, the differences in the fatty acid patterns of the abdominal fat that were caused by feeding different dietary fat sources resulted in significant differences in the slip point and slight differences in the clarification point. A difference of 9.5 C in the slip point (between FP2 and RO) will certainly have a major impact on the consistency of the abdominal fat pad. From the results in Table 8 it can be concluded that SFA and (MUFA+PUFA):SFA give the best estimation of both the slip point and the clarification point. Also, the percentage of stearic acid (Wood et ah, 1978; Enser, 1984b) or the MUFA:SFA ratio (Lea et al, 1970) may be the best predictors for the melting point of adipose tissue. However, in this experiment lower r2 values for the stearic acid than for several other criteria were found (Table 8). For the data presented here, r 2 values even for the criteria that give the best prediction of the melting point are only in a medium range of about 0.5. This result means that only 50% of the variance of the melting point can be explained by the variance of the respective character. In general, r2 values are slightly higher for the slip point than for the clarification point (Table 8). This higher r2 seems to be due to the greater probability for inaccuracies when determining the clarification point than when determining the slip point, which results in a greater variation of the data for the clarification point. In conclusion, the results of this study showed that on the one hand, abdominal fat and subcutaneous fat have an
TABLE 8. Effect of fatty acid content on the melting points of abdominal fat measured by P value and coefficient of determination Clarification point
Slip point 2
Fatty acids
P value
r
Cl8:0
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
0.24 0.53 0.19 0.42 0.35 0.53 0.30
SFAi Cl8:0 : Q8:l
SFA:C 18:1 SFA:MUFA2 (MUFA+PUFA):SFA3 PUFA:SFA
!SFA = saturated fatty acids. MUFA = monounsaturated fatty acids. 3 PUFA = polyunsaturated fatty acids. 2
P value
r2
<0.001 <0.001 0.003 <0.001 <0.001 <0.001 <0.001
0.17 0.48 0.13 0.35 0.28 0.47 0.31
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
occurred with regards to palmitic acid and PUFA. A reason for this result could be that the PUFA may have been metabolized at a high rate in the muscle tissue, whereas they have been stored as a source of energy in the adipose tissues. In contrast to this, the content of MUFA is quite similar for both the adipose tissues and the meat portions. In this experiment, there was also a trend towards a higher content of stearic acid in the thigh than in the breast portion, and the lowest values were recorded for the abdominal fat (Table 3 and 5). This can be explained by the higher content of phospholipids in the muscles (Salmon and Neil, 1973). The data presented here also show that for commonly consumed portions of the carcass (breast and thigh), the content of PUFA and the ratios (MUFA+PUFA):SFA and PUFA:SFA, which are frequently discussed as having a major impact on the nutritive quality of foodstuffs for human nutrition (Gurr, 1992), are influenced only to a rather small extent by the fatty acid content of the animal's diet. Differences between adipose and other tissues have been reported earlier (Marion and Edwards, 1963; Marion and Woodroof, 1963). The sex of the chickens was expected to have no effect on the fatty acid pattern of both the adipose tissues and the meat portions (Olomu and Baracos, 1991). This expectation was confirmed by the data from this experiment.
DIETARY FATTY ACIDS AND CARCASS FAT almost identical fatty acid pattern. On the other hand, composition of the adipose tissues differ significantly from the fat extracted from the meat portions. This is the case for the fatty acid pattern itself as well as for the observed effect of the dietary fat on the composition of the animal fat. Contrary to the composition of the adipose tissue, which closely follows the dietary lipids, the fat in the meat portions can be influenced by different dietary fat sources only to a small extent. This is especially the case for the content of PUFA, which is often discussed in connection with consumer demand for "healthy" foodstuffs.
ACKNOWLEDGMENTS
REFERENCES Ajuyah, A. O., K. H. Lee, R. T. Hardin, and J. S. Sim, 1991. Changes in the yield and in the fatty acid composition of whole carcass and selected meat portions of broiler chickens fed full-fat oil seeds. Poultry Sci. 70:2304-2314. Chanmugam, P., M. Boudreau, T. Boutte, R. S. Park, J. Hebert, L. Berrio, and D. H. Hwang, 1992. Incorporation of different types of n-3 fatty acids into tissue lipids of poultry. Poultry Sci. 71:516-521. Elliot, J. I., and J. P. Bowland, 1969. Correlation of melting point with the sum of the unsaturated fatty acids in samples of porcine depot fat. Can. J. Anim. Sci. 49:397-398. Enser, M., 1984a. The chemistry, biochemistry and nutritional importance of animal fats. Pages 23-51 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Enser, M., 1984b. The relationship between the composition and consistency of pig backfat. Pages 53-57 in: Fat Quality in Lean Pigs. A Workshop in the CEC Programme of Coordination of Research on Animal Husbandry, Brussels, Belgium. Freeman, C. P., 1984. The digestion, absorption and transport of fats—Non-ruminants. Pages 105-122 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Gurr, M. I., 1992. Role of Fats in Food and Nutrition. 2nd ed. Elsevier Science Publishers Ltd., Barking, UK. Hargis, P. S., and M. E. Van Elswyk, 1993. Manipulating the fatty acid composition of poultry meat and eggs for the health conscious consumer. World's Poultry Sci. J. 49: 251-264. Holm, S., 1979. A simple sequentially rejecrive multiple test procedure. Scand. J. Stat. 6:65-70. Ketels, E., and G. De Groote, 1987. Effect of fat source and level of fat inclusion on the utilization of fatty acids in broilers diets. Arch. Gefltigelkd. 51:127-132.
Lea, C. H., P.A.T. Swoboda, and D. P. Gatherum, 1970. A chemical study of soft fat in cross-bred pigs. J. Agric. Sci. 74:279-284. Marion, J. E., and H. M. Edwards, 1963. The effect of dietary fatty acid cholesterol on growth and fatty acid composition of the chicken. J. Nutr. 79:53-61. Marion, J. E., and J. G. Woodroof, 1963. The fatty acid composition of breast, thigh and skin tissue of chicken broilers as influenced by dietary fats. Poultry Sci. 42: 1202-1207. Morrison, W. R., and L. M. Smith, 1964. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron trifluorid—methanol. J. Lipid Res. 5:600-608. Naumann, K., and R. Bassler, 1983. Die chemische Untersuchung von Futtermitteln. Verlag J. NeumannNeudamm, Melsungen, Berlin, Germany. Olomu, J. M., and V. E. Baracos, 1991. Influence of dietary flaxseed oil on the performance, muscle protein deposition and fatty acid composition of broiler chicks. Poultry Sci. 70:1403-1411. Pethick, D. W., A. W. Bell, and E. F. Annison, 1984. Fats as energy sources in animal tissues. Pages 225-248 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Pinchasov, Y., and I. Nir, 1992. Effect of dietary polyunsaturated fatty acid concentration on performance, fat deposition and carcass fatty acid composition in broiler chickens. Poultry Sci. 71:1504-1512. Salmon, R. E., and J. B. O'Neil, 1973. The effect of the level and source and of a change of source of dietary fat on the fatty acid composition of the depot fat and the thigh and breast meat of turkeys as related to age. Poultry Sci. 52:302-314. SAS Institute, 1989. SAS/STAT® User's Guide. Release 6.03 Edition. SAS Institute Inc., Cary, NC. Scaife J. R., J. Moyo, M. Galbraith, M. Michie, and V. Campbell, 1994. Effect of different dietary supplemental fats and oils on the tissue fatty acid composition and growth of female broilers. Br. Poult. Sci. 35:107-118. Sheehy, P.J.A., P. A. Morrissey, and A. Flynn, 1993. Influence of heated vegetable oils and alphatocopherol acetate supplementation on alphatocopherol, fatty acids and lipid peroxidation in chicken muscle. Br. Poult. Sci. 34:367-381. Valencia, M. E., E. S. Watkins, A. L. Waldroup, P. W. Waldroup, and D. L. Fletcher, 1993. Utilization of crude and refined palm and palm kernel oils in broiler diets. Poultry Sci. 72:2200-2215. Whitehead, C. C, 1984. Essential fatty acids in poultry nutrition. Pages 153-166 in: Fats in Animal Nutrition. J. Wiseman, ed. Butterworths, London, UK. Wood, J. D., H. B. Enser, H.J.H. Mc Gie, W. C. Smith, J. P. Chadwick, M. Ellis, and R. Laird, 1978. Fatty acid composition of backfat in Large White pig selected for low backfat thickness. Meat Sci. 2:289-300. Yau, J. C, J. H. Denton, C. A. Bailey, and A. R. Sams, 1991. Customizing the fatty acid content of broiler tissues. Poultry Sci. 70:167-172. Zollitsch, W., W. Wetscherek, and F. Lettner, 1992. Use of rapeseed oil in a broiler diet. Arch. Gefltigelkd. 56: 182-186.
Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on March 14, 2015
The authors thank the Austrian Science Foundation and the Austrian Federal Ministry of Science and Research (Oesterreichischer Akademischer Austauschdienst) for financial support of this research.
215