1st Page

Journal of Animal Science - Article



This article in

  1. Vol. 87 No. 1, p. 285-295
    Received: June 03, 2008
    Accepted: July 21, 2008
    Published: December 5, 2014

    2 Corresponding author(s):


Dietary conjugated linoleic acid changes belly and bacon quality from pigs fed varied lipid sources1

  1. S. T. Larsen*,
  2. B. R. Wiegand†2,
  3. F. C. Parrish*,
  4. J. E. Swan* and
  5. J. C. Sparks*
  1. Iowa State University, Ames 50011; and
    University of Missouri, Columbia 65211


The objective of this experiment was to determine the effects of dietary lipid source with or without the addition of CLA on bacon composition and quality. Forty-eight barrows at a beginning BW of 55 kg ± 2.2 were fed 1 of 6 diets for 56 d. These diets consisted of: 1) normal corn (NC), 2) NC + 1.25% CLA-60 oil (NC + CLA), 3) high-oil corn (HOC), 4) HOC + 1.25% CLA-60 oil (HOC-CLA), 5) NC + choice white grease (CWG; NC + CWG), and 6) NC + CWG + 1.25% CLA-60 oil (NC + CWG + CLA). The CLA-60 contains 60% CLA isomers in the oil, and therefore, 1.25% oil was needed to achieve 0.75% CLA in the diet. Soy oil replaced CLA in control diets. Choice white grease and high-oil corn were selected as fat sources for this study because of their utility in energy density for growing-finishing pigs, especially in hot weather. Pigs were slaughtered at an average BW of 113 kg ± 4.1, and carcasses were fabricated at 24 h postmortem. Statistical analysis was performed using the mixed model procedure of SAS, and the main effects tested were dietary lipid source, CLA, and 2-way interaction. The addition of CLA to each basal diet improved (P < 0.05) belly firmness measured either lean side down or fat side down from the belly bar firmness test [4.39 cm vs. 7.01 cm (lean down) and 5.75 cm vs. 10.54 cm (fat down)] for 0 and 0.75% dietary CLA, respectively. The compression test used on bacon slabs showed that bacon from CLA-supplemented pigs was approximately 20% firmer than that from controls. Pigs fed the HOC diets had softer bellies compared (P < 0.05) with pigs fed the NC diet as measured by the belly bar test [6.94 cm vs. 9.26 cm (fat down)], respectively. Conjugated linoleic acid did not, however, improve bacon sliceability. No differences were observed for moisture, protein, or lipid percentages between any treatments. Overall, there was a CLA effect (P < 0.04) for lipid oxidation, in which the addition of CLA decreased bacon oxidation (0.1498 CLA vs. 0.1638 no CLA). Dietary CLA increased the percentage of SFA in tissues from pigs supplemented with CLA. Dietary inclusion of CLA increased the concentration of all measured isomers of CLA in bacon. Sensory scores of bacon showed no differences for any of the sensory attributes measured between any of the treatments. Our results indicate that inclusion of dietary CLA will improve belly firmness, extend the shelf life stability of bacon, and increase the degree of fat saturation.


Soft bellies are a problem in bacon production, because they have decreased yields, poor sliceability, and potentially decreased shelf life (Nordstrom et al., 1972; Shackelford et al., 1990; Correa et al., 2008). Leaner genetic lines of pigs tend to have leaner bellies and deposit more unsaturated fat compared with fatter pigs (Wood et al., 1985; Correa et al., 2008). Additionally, belly softness can be attributed to the variety of unsaturated lipids fed to pigs. The use of high-oil corn and choice white grease in the diet of swine yields greater percentages of soft bellies (Nordstrom et al., 1972; Rentfrow et al., 2003). We hypothesized that by modifying these diets by using dietary CLA, the production of firmer, greater-quality bellies could be accomplished. Thiel-Cooper et al. (2001) observed that the production of firmer bellies resulted from CLA-fed pigs. Several reports indicate a shift to more SFA profiles in pork tissues with dietary CLA inclusion (Thiel-Cooper et al., 2001; Wiegand et al., 2002; Ostrowska et al., 2003; Dunshea et al., 2005). Altering the gross composition of pork is another advantage of feeding dietary CLA to pigs. Dugan et al. (1997) and Swan et al. (2001) showed a decrease in fat and an increase in lean of CLA-fed pigs.

In addition to improved bacon composition and quality, another potential benefit of feeding CLA is the increased incorporation of CLA in bacon (Dunshea et al., 2005), which may make bacon a more healthful product if consumed with other CLA-rich foods. Conjugated linoleic acid has been shown to decrease risk factors that cause heart disease (Lee et al., 1994), protect against certain types of cancer (Park et al., 1998), and enhance the immune system (Wang et al., 1998; Winchell et al., 1998) in rodent models and cell culture studies. Because little literature exists specifically on the characteristics of processed pork with CLA supplementation in conjunction with highly unsaturated fat sources in pigs, the objective of this experiment was to determine the effect of dietary lipid source with or without CLA on bacon composition and quality.


All procedures of this project were in accordance with the guidelines of the Iowa State University Animal Care and Use Committee.

Animals and Diet

Forty-eight crossbred barrows, penned individually and randomly assigned, weighing 55 kg ± 2.2, were fed 1 of 6 diets for 56 d. One pig was removed from the test due to health reasons unrelated to treatment. These diets were as follows: normal corn (NC, n = 8), NC + CLA-60 oil (Conlinco, Detroit Lakes, MN; now available from BASF Corporation, Floram Park, NJ; NC +CLA, n = 8), high-oil corn (HOC, n = 8), HOC + CLA-60 oil (HOC + CLA, n = 8), NC + choice white grease (CWG; NC + CWG, n = 8), and NC + CWG + CLA-60 oil (NC + CWG + CLA, n = 7). The CLA-60 contains 60% CLA isomers in the oil, and therefore, 1.25% oil was needed to achieve 0.75% CLA in the diet. The concentration of 0.75% CLA was selected as an optimum for growth, carcass, and meat quality factors based on previous work with the same genetic line of pigs (Swan et al., 2001; Thiel-Cooper et al., 2001). Pigs were allowed ad libitum access to feed and water. Experimental diets were formulated to have equal ratios of ME to lysine between the NC and the HOC and NC + CWG diets. Composition and calculated analysis of the diets are shown in Tables 1 and 2. Table 3 shows the fatty acid profile of the CLA supplement. The pigs were removed from the test, fasted 12 h, and then humanely slaughtered at the Iowa State University Meat Laboratory at a BW of 113 kg ± 4.1.

Belly and Bacon Quality Measurements

After a 24-h chill at 2°C, carcasses were fabricated and each pair of pork bellies (Institutional Meat Purchase Specifications 408) was collected and spare ribs removed. Bellies were skinned using a Townsend Model 9000 belly skinner (Townsend Engineering, Des Moines, IA), and a green belly weight was recorded using a model D5-410 CW digital scale (Digi Matex Inc., San Diego, CA). Before processing, belly firmness was measured using a standard bar test (Thiel-Cooper et al., 2001). Briefly, each belly was suspended longitudinally across a 1.27-cm-diameter stainless steel bar, and the distance between the shoulder and ham ends of the belly was measured. Each belly was measured fat side down and lean side down. Then each belly was brine-pumped to a 12% target concentration using a Townsend 1400 injector (Townsend Engineering Inc.). The composition of the brine was 78.69% water, 12.25% sodium chloride, 4.25% sucrose, 4.25% sodium phosphate, 0.458% sodium erythorbate, and 0.102% sodium nitrite. After injection, bellies were allowed to equilibrate at 4°C and then were weighed to determine pumped belly weight.

Smoking and thermal processing of the bellies were accomplished in a Maurer (Maurer, Reichenau, Germany) thermal processing unit. A final endpoint temperature of 55°C was attained, and immediately after their removal from the thermal processing unit, bacon was weighed for a hot yield weight. Then, bacon was placed in a cooler at 2°C to chill for 24 h.

After a 24-h chill, the bacon sides were weighed to determine a chilled final processing weight. The bacon side was then squared and a trimmed weight was recorded. The bacon was then placed in a US Berkel model 1170 GS (US Slicing Machine Company Inc., Laporte, IN) slicer (18 to 20 slices per pound). Bacon was completely sliced, and the slices were separated on the basis of visual quality. A slice of bacon in the acceptable category needed to be intact (not shattered or the fat could not be separated from the lean) and it needed to be a full slice of bacon. A slice of bacon in the unacceptable category needed to be shattered or the fat was separated from the lean. Shatter classification was performed as described by Mandigo (1998). Each of these categories was weighed to determine the portions of acceptable and unacceptable bacon slices.

Slice Selection

The bacon slices were separated into acceptable and unacceptable categories. When the bacon slices were removed, they were laid in the same order that they were sliced from the belly. Then 2 slices were selected from each end of the belly and from the middle portion of the belly to help minimize variations within bellies.

Proximate, Cooking, and Texture Analysis

A total of 6 slices, 2 slices from each end and 2 slices from the middle section of each bacon slab, were vacuum-packaged using a Multipac double chamber-packaging machine (model AG800, Multipac, Hamburg, Germany) in Cryovac (7 × 12 oxygen-impermeable) clear vacuum bags (Sealed Air Corporation, Simpsonville, SC) and stored for 5 mo or until the slices were needed for analysis at 2°C. These 6 slices were subsequently used for proximate analysis for moisture, fat, and protein (AOAC, 1993). A homogenous mixture was acquired by freezing the slices in liquid nitrogen and immediately pulverizing them in a Waring blender (Dynamics Corporation of America, New Hartford, CT). A package of 6 slices was used to measure moisture, lipid, and protein percentages from a pulverized sample (AOAC, 1993).

Another 6 slices were used to determine lipid oxidation. The vacuum-packaged bacon, held at 2°C, was analyzed once a month for 5 consecutive months, beginning 1 mo after belly processing. Thiobarbituric acid reactive substances were determined using the procedures of Zipser and Watts (1962). Slices were frozen in liquid nitrogen and then ground in a Waring blender to obtain a homogenized ground sample. The homogenized ground sample was then immediately used for thiobarbituric acid reactive substances analysis. Lipid oxidation values are reported in milligrams of malonaldehyde per kilogram of sample and were read from a spectrophotometer at 532 nm (Beckman Instruments Inc., Schaumburg, IL).

Six more slices were used for a sensory panel evaluation. An 8-member consumer sensory panel was used to determine the sensory attributes of sliced bacon. The members of the sensory panel consisted of graduate students, staff, and faculty at Iowa State University. Two methods were used to cook the slices, frying and microwave methods. These slices were cooked according to recommendations by AMSA (1985; i.e., the slices were cooked beyond the limp stage or until the desired crispness was reached). An electric skillet (The West Bend Co., West Bend, WI) set at 177°C cooked the slices for the frying method. The slices were cooked on one side for 2 min and 30 s, turned a second time and cooked for another 1 min and 30 s, and then turned one more time and cooked for 30 s.

A Radarange microwave oven powered with 1,500 max watts (Amana Refrigeration Inc., Amana, IA) cooked the slices for the microwave method. The 6 slices were arranged on a microwavable tray, placed in the microwave and cooked at full power for 2 min and 20 s, rotated 180°, and cooked for another 2 min and 20 s. For both cooking methods, the slices were blotted by using a paper towel, cut in half, and immediately served to the sensory panelists in a random order of both diet and cooking method.

Each panelist evaluated a half slice of bacon cut into 2 pieces. The panelists evaluated aroma, lean color intensity, flavor intensity, off-flavor intensity, and brittleness on a scale of 1 through 8. The score of 1 described the least aroma, lightest color intensity in the lean, least flavorful, most off-flavor, and most brittle piece of bacon. The score of 8 described the most aroma, darkest color intensity in the lean, most flavorful, least off-flavor, and least brittle piece of bacon. Lean color was evaluated by visually appraising the lean portion of the bacon piece. Flavor and off-flavor were evaluated by placing the sample in the mouth and chewing to evaluate the taste and off-flavor. Brittleness was evaluated on the first bite to determine how well the bacon broke into pieces.

Another 6 slices were analyzed to determine cooking yields. The 2 methods used for sensory evaluation were also used to cook the slices for cooking yield determination (AMSA, 1985). After cooking, the slices were immediately blotted with a paper towel to remove excess grease and weighed using a Mettler PM2000 (Mettler Instrument Corp., Hightstown, NJ) to determine a final cooking loss.

The opposite bacon side was partially sliced, and three 10.16-cm slabs were taken at 25, 50, and 75% of the length of the belly from the shoulder end and used for compression tests. Compression tests were performed at 25, 50, and 75% of the width from the navel edge of the bacon slab by using a TA-XT2 Texture Analyzer (Texture Technologies Corp., Scarsdale, NY). The Texture Analyzer was equipped with an 8-mm stainless steel spherical probe. The conditions for the compression test were as follows: 2.0 mm/s pretest speed, 1.0 mm/s test speed, 1.0 mm/s posttest speed, 1.0% rupture test distance, 3,000 × g force, 5 s, and a 5 count. Compression firmness was defined as the force necessary to compress a slab of bacon 25% of the original thickness of the slab. Values are reported in kilograms of force.

Fatty Acid Analysis

Six slices (2-wk vacuum storage) were taken for lipid extraction using the Folch method (Folch et al., 1957), with chloroform:methanol (2:1, vol/vol). Lipids extracted from all of the samples were methylated (sodium methoxide) by the following procedure: 2 g of sample was dissolved in hexane (1 mL) in a capped test tube, 0.5 M sodium methoxide in anhydrous methanol (2 mL) was added, the solution was maintained at 50°C for 10 min, and glacial acetic acid (0.1 mL) was added followed by deionized water (5 mL). The fatty acid methyl esters were extracted into hexane (2 × 3 mL) and dried over anhydrous sodium sulfate. All of the fatty acid methyl esters were analyzed by using a gas chromatograph (HP 6890, autosampler 6890; Hewlett-Packard Co., Avondale, PA) equipped with an HP 19091J-413 column (30 m, 0.320-μm i.d., 0.25-μm film thickness: Hewlett-Packard Co.) and operated at 180°C for 0.5 min (temperature programmed 2°C/min to 230°C and held for 4.5 min). The injector and flame-ionization detector temperatures were both set at 300°C. Fatty acid methyl esters were identified by comparison of their retention times with an authentic standard (UC-59 MX, Nu-Chek-Prep, Elysian, MN) and were verified with mass spectrophotometry. The ionization potential of mass selective detector (model 5973, Hewlett-Packard Co.) was 70 eV, scan range was 50 to 550 m/z, and scan velocity was 2.94 scans/s. The identification of volatiles was achieved by comparing mass spectral databased on their retention times with those of the Wiley library (Hewlett-Packard Co.). Fat and CLA extractions were run in duplicate to determine percentage of lipids and reported as an average. Average CLA values are reported as a percentage of total fatty acids.

Statistical Analysis

This experiment was a complete randomized design consisting of a 3 (NC, HOC, and NC + CWG) × 2 (0 or 0.75% CLA) factorial arrangement. The individual pig served as the experimental unit. One pig was removed from the test due to health reasons. Statistical analysis was performed for all measurements using PROC MIXED (SAS Inst. Inc., Cary, NC). Main effects tested were dietary lipid source NC, HOC, or CWG in combination with CLA. Least squares means were used to determine level of significance at an α level of P < 0.05 with an adjustment for all pairwise comparisons by the Tukey-Kramer procedure. Repeated measures were used to analyze thiobarbituric acid and taste panel data with day serving as the repeated measure. Each cooking method for bacon was utilized for each pig. Least squares means were used to determine level significance at P < 0.05.


Carcass data are shown in Table 4 and indicate no differences for carcass weight, loin muscle area, or internal (leaf) fat. Subcutaneous fat depots at the first and last rib did not differ by treatment. However, subcutaneous fat measured at the last lumbar vertebrae and at the 10th and 11th rib junction were greater (P < 0.04) for pigs fed HOC regardless of CLA supplementation. It is reasonable to expect these differences, because although diets were equivalent for ME:lysine, they were not isocaloric and the additional energy availability from HOC resulted in fatter pigs at these depots.

Belly bar firmness values of fresh bellies are shown in Table 5. The bellies from CLA-fed pigs had increased (P < 0.05) belly firmness when measured fat side down and lean side down. Thiel-Cooper et al. (2001) also showed an increase in belly firmness from pigs supplemented with CLA. However, there was a decrease (P < 0.05) in belly firmness from the feeding of HOC compared with the NC diet when measured fat side down. These results agree with other data indicating a softening of bellies as a result of feeding high-oil corn of a similar fat percentage (Rentfrow et al., 2003). Typically, de novo synthesis of SFA (firmer fat) occurs in pigs (O’Hea and Leveille, 1969). However, pigs with genetic predisposition for less subcutaneous fat also can be expected to produce carcasses with more unsaturated fatty acids (Wood et al., 1985; Correa et al., 2008). These changes toward more unsaturated fatty acids are presumably a function of less de novo fatty acid synthesis and greater uptake of dietary fatty acids. Subcutaneous fat depth at the 10th rib for carcasses from this study ranged from 2.08 to 2.59 cm, which do not likely fall among the leanest pigs in US pork production. Therefore, increased belly firmness can likely be explained, in part, by the increase in the SFA:unsaturated fatty acid ratio as a result from feeding CLA as shown by others in rats (Szymczyk et al., 2000) and pigs (Wiegand et al., 2002; Lauridsen et al., 2005). Data in Table 5 show compression values of force to determine firmness of bacon slabs from pigs fed various fat sources. Bacon from CLA-fed pigs had (P < 0.05) greater bacon slab firmness compared with bacon from pigs not fed CLA. Also, there were no differences between the NC, the HOC, and the NC + CWG diets for bacon slab firmness. This observation reinforces the belly firmness test in the current study and also in studies showing an increase in belly firmness from pigs supplemented with CLA (Thiel-Cooper et al., 2001). Eggert et al. (2001) suggested that the firmer belly from CLA feeding would make the belly easier to process for bacon because of improved physical attributes. Miller et al. (1993) reported a decrease in fat firmness causing bellies to be unacceptable for bacon production when unsaturated fats, such as CWG, were fed in swine diets. Unacceptable bellies for bacon production from CWG-fed pigs were not seen in this study or in previous studies of feeding CWG in swine diets (Engel et al., 2001; Rentfrow et al., 2003). However, CWG concentration, CWG fatty acid composition, and days on feed should be considered when comparing studies, because each of these factors could have dramatic effects on fatty acid profiles of nonruminant animals.

No differences were observed between treatments for belly processing weights, belly processing yields, and bacon sliceability weights (acceptable or unacceptable bacon slices) from pigs fed various fat sources (Table 6). Gatlin et al. (2002) showed an increase in the green weight of bellies from CLA supplementation, and that difference was maintained after pumping and smoking. Rentfrow et al. (2003) reported less bacon slice fracturing with dietary inclusion of HOC. The current study does not support this even though the oil content from HOC was similar for both studies. However, days on feed were 98 d for Rentfrow et al. (2003) and only 56 d for the current study, which may have resulted in a greater incorporation of PUFA in the bacon tissue for Rentfrow et al. (2003).

Thiobarbituric acid values as a measure of bacon freshness (lack of oxidation) from pigs fed various fat sources showed a slight difference (P < 0.04) between treatments at 2 mo of vacuumed and refrigerated bacon storage (Table 6). This could be due to the nitrite in the bacon formulation being recognized as an anti-oxidant and would cause significant reduction in lipid oxidation (Shahidi et al., 1991), and these samples were vacuum-packaged, causing limited lipid oxidation. However, this difference was an overall CLA effect (P < 0.04), in which the addition of CLA decreased bacon lipid oxidation. However, the difference was so small (0.1498 CLA vs. 0.1638 no CLA) and would doubtfully be considered significant or detectable by consumers. Previous studies have also observed an increase in shelf life stability with the supplementation of CLA in swine diets (Joo et al., 2002). The combination of vacuum-packaging and the nitrite in the bacon formulation probably helped extend the shelf life of the HOC and the NC + CWG diets. Previous studies have demonstrated that with an increased concentration of unsaturated fatty acids fed in animal diets, that the products from those animals are more susceptible to lipid oxidation (Kouba et al., 2003; Elmore et al., 2005), especially bacon (Rogers and Etzler, 2000; Sheard et al., 2000). However, the values throughout the shelf life study are very small and were well below the suggested threshold of 0.5 to 1.0 for oxidative rancidity reported by Tarladgis et al. (1960).

Furthermore, no statistical differences were observed for cooking yields of fried and microwave bacon slices from pigs fed various fat sources (Table 7). Also, there were no significant differences of protein, moisture, and lipid percentages from raw, fried, and microwave bacon slices (Table 7). This result is in contrast to other studies showing an increase of lean (protein and moisture) and a decrease of fat (lipid) when CLA was added to pig diets (Ostrowska et al., 1999; Swan et al., 2001). The lack of composition changes in bacon for the current study may not be surprising, because other fat and muscle depots were not affected by CLA either (i.e., subcutaneous fat and loin muscle area).

Effects of CLA on fatty acid profiles of bacon from pigs fed various fat sources are shown in Table 8. The percentage of palmitic acid (16:0), stearic acid (18:0), and palmitoleic acid (16:1) increased (P < 0.05) in bacon tissues from CLA-fed pigs. Ostrowska et al. (2003) reported similar increases in 16:0 and 16:1 for intramuscular and subcutaneous fat depots in pigs. Similar to their study, the current study replaced soy oil with CLA, which resulted in a linear decrease of dietary palmitic acid in the CLA diet. The shift to more palmitic acid in tissues is therefore driven by dietary CLA as discussed by Ostrowska et al. (2003). Conversely, the percentage of oleic (18:1), linolenic (18:3), and arachidonic acid (20:4) decreased (P < 0.05) in bacon tissues from CLA-fed pigs. Wiegand et al. (2002) showed similar results of the change in fatty acid composition from CLA-fed pigs. Additionally, Chin et al. (1994) and Park et al. (1997) demonstrated an increase in SFA and a decrease in unsaturated fatty acid from feeding CLA to rats and mice. Research suggests that feeding CLA decreases the Δ9-desaturase index and stearoyl CoA desaturase (SCD1) enzyme activity (Lee et al., 1998; Smith et al., 2002). The significance of a SCD1 reduction lies in the fact that SCD1 is the rate-limiting enzyme in the synthesis of MUFA where a double-bond is added at Δ9. A reduction of 45 and 75% in SCD1 mRNA concentrations in liver tissue was observed in mice supplemented 0.5% dietary CLA in conjunction with high-carbohydrate or 5.0% corn oil diets, respectively (Lee et al., 1998). Smith et al. (2002) reported a decrease in Δ9-desaturase index and inhibition of SCD1 activity when1.75% CLA was fed to pigs. Therefore, a reasonable assumption would be an increase in SFA as a result of feeding CLA. In contrast, Wynn et al. (2006) fed a rumen-protected CLA to sheep, and although they observed a decrease in 16:1 in subcutaneous fat and an increase in 18:0 in perirenal fat, there was not an inhibition of SCD1 activity. Hwang and Kang (2007) reported a decrease in 16:1 and a subsequent increase in 18:0 in liver tissues of mice fed beef tallow + CLA and fish oil + CLA. The SCD1 activity was not inhibited in this model. In fact, SCD1 activity actually increased with concurrent decreases in total fat deposition, tria-cylglyceride concentration, and Δ9-desaturase index in the beef tallow + CLA mice. Overall, the total SFA increased (P < 0.05) and the total unsaturated fatty acids decreased (P < 0.05) in bacon tissues from CLA-fed pigs in the current study.

An interesting observation in this and other studies with CLA feeding of pigs is the decrease in arachidonic acid (20:4) in certain tissues. A significant (P < 0.05) decrease from 1.39 to 1.14% of 20:4 was observed for bacon from the CLA-supplemented pigs in the current study. This sample was an aggregate of lean and fat from intact bacon slices, so we cannot differentiate if the decrease in 20:4 was in adipose or muscle portion of the pork belly. Szymczyk et al. (2000) and Ramsay et al. (2001) reported a decrease in 20:4 in subcutaneous fat layers, whereas Wiegand et al. (2002) showed a linear decrease for 20:4 in loin muscle tissues, but no change of 20:4 in subcutaneous adipose tissue. Joo et al. (2002) reported a decrease in 20:4 in intramuscular fat tissue. Ramsay et al. (2001) indicated a greater sensitivity to fatty acid composition change in adipose tissue compared with muscle tissue with dietary inclusion of CLA. Depending on the magnitude of 20:4 loss in muscle and adipose tissue, one might expect some effects on other aspects of metabolism, specifically bone remodeling. Mater et al. (1999) discussed how n-6 PUFA are precursors to PG. A specific example of this is that arachidonic acid is an upstream precursor to PGE2 (Watkins and Seifert, 2000). Because PGE2 (at increased concentrations) is known to stimulate bone catabolism, it seems possible that a decrease in PGE2 could result in greater bone mass (Watkins and Seifert, 2000). Li and Watkins (1998) reported that CLA alters bone fatty acid composition by decreasing ex vivo PGE2 biosynthesis in rats when fed with menhaden fish oil, but not when fed with soybean oil.

The percentage of palmitic acid (16:0) increased (P < 0.05) in bacon tissues from NC-fed pigs compared with HOC and NC + CWG-fed pigs. Palmitoleic acid (16:1) increased (P < 0.05) in bacon tissues from the NC and NC + CWG-fed pigs compared with HOC-fed pigs. Engel et al. (2001) observed that the supplementation of CWG increased MUFA (17:1 and 20:1). This study did not analyze those specific MUFA, but there was an increase in the 16:1 (palmitoleic) MUFA. Feeding MUFA may increase those fatty acids in the tissues of these pigs (Engel et al., 2001). Linoleic acid (18:2) increased (P < 0.05) in bacon tissues from the HOC-fed pigs compared with the NC and NC + CWG-fed pigs. Linolenic acid (18:3) increased (P < 0.05) in bacon tissues from the NC + CWG-fed pigs compared with the NC and HOC-fed pigs. Also, linolenic acid (18:3) increased (P < 0.05) in bacon tissues from the normal corn-fed pigs compared with the HOC-fed pigs. Increases in unsaturated fatty acids in pork tissues by feeding unsaturated fats have been observed regularly (Shackelford et al., 1990; Fontanillas et al., 1998; Rentfrow et al., 2003). Therefore, the increase in the degree of unsaturation from feeding HOC in the current study is not surprising. Direct incorporation of C18 and longer PUFA would be expected, because these are obtained mainly through dietary intake, whereas C16 and C18 SFA and MUFA are generally the outcome of metabolic synthesis.

Three detectable CLA isomers were incorporated into bacon tissues of CLA-fed pigs; those isomers were trans-9, trans-11; cis-9, trans-11; and trans-10, cis-12 isomers (Table 8). We recognize the fact that the gas chromatography column length in the current study may have limited our ability to detect additional isomers of CLA in the bacon product. Many isomers, including the 3 in this study, have been shown to be biologically active in decreasing fat, increasing lean, protecting against factors that cause heart disease and cancer, and enhancing the immune system in rodent models (Ha et al., 1990; Lee et al., 1994; Wang et al., 1998). The difference in the percentages of the isomers found in the bacon tissues seem to be correlated linearly to the amount of isomers fed to pigs (Cook et al., 1998; Thiel-Cooper et al., 2001; Dunshea et al., 2005). However, the efficiency of uptake to pork tissues seems to differ according to various CLA isomers (Ostrowska et al., 2003). These authors indicated that the cis-9, trans-11 isomer was most efficiently deposited in subcutaneous adipose. Differential uptake of dietary CLA can be a factor when considering CLA-enriched products via metabolic uptake and deposition by food-producing animals.

Sensory scores of bacon cooked by the fried and microwaved methods from pigs fed various diets showed no differences for any of the sensory attributes measured between any of the treatments (Table 9). These results agree with previous studies, in which no differences were observed in sensory characteristics with CLA supplementation (Dugan et al., 1999; Wiegand et al., 2001). In a meta-analysis of 15 studies investigating CLA and pork quality, Dunshea et al. (2005) pointed out that CLA seems to have the ability to increase marbling by 11% on average while decreasing undesirable backfat by 6%. Dunshea et al. (2005) also discussed a decrease, albeit small, of flavor, juiciness, and tenderness in response to CLA supplementation to pigs despite the increase in marbling. These changes have been reported in various loci of the fresh LM, and, although important, might not bear heavily on the discussion of sensory attributes of further-processed pork products.

There was a cooking method difference for lean color and flavor (P < 0.02) of bacon regardless of dietary treatment. In each of those sensory measurements, the frying method gave a darker lean color (5.44 vs. 5.10, P < 0.05) and a more flavorful (5.57 vs. 5.33, P < 0.05) piece of bacon compared with the microwave method, respectively. This is likely due to the bacon being fried on a hot metal frying pan compared with being cooked on a plastic bacon tray in the microwave oven.

In summary, dietary inclusion of CLA at 0.75% to growing-finishing pigs will alter the fatty acid profile of pork fat toward a greater degree of saturation, specifically in belly fat. This shift toward saturation corresponds to a decrease in lipid oxidation of the pork fat. Although the saturation of fat creates a firmer pork belly, this firmness does not result in improved sliceability of the subsequent bacon product.

View Full Table | Close Full ViewTable 1.

Composition (as fed) and calculated analysis of experimental diets fed to growing-finishing pigs for d 1 through 28

Ingredient, %
    Corn 76.6 76.6 72.4 72.4
    HOC 76.9 76.9
    Soybean meal 19.6 19.6 19.5 19.5 21.1 21.1
    CWG 3.0 3.0
    Soy oil 1.25 1.25 1.25
    CLA 1.25 1.25 1.25
    Dicalcium phosphate 0.57 0.57 0.57 0.57 0.57 0.57
    Calcium carbonate 0.77 0.77 0.77 0.77 0.77 0.77
    Salt 0.25 0.25 0.25 0.25 0.25 0.25
    Vitamin premix7 0.20 0.20 0.20 0.20 0.20 0.20
    Mineral premix8 0.05 0.05 0.05 0.05 0.05 0.05
    Tylosin premix9 0.05 0.05 0.05 0.05 0.05 0.05
Calculated analysis
    ME, kcal/kg 3,318 3,318 3,468 3,468 3,468 3,468
    CP, % 15.0 15.0 15.6 15.6 15.6 15.6
    Crude fat, % 4.24 4.24 6.64 6.64 7.10 7.10
    Lys, % 0.80 0.80 0.84 0.84 0.84 0.84
    Met + Cys, % 0.54 0.54 0.58 0.58 0.55 0.55
Analyzed, %
    Lys 0.73 0.72 0.75 0.73 0.75 0.73

View Full Table | Close Full ViewTable 2.

Composition (as fed) and calculated analysis of the experimental diets fed to growing-finishing pigs for d 29 through 56

Ingredient, %
    Corn 87.95 87.95 82.46 82.46
    HOC 87.88 87.88
    Soybean meal 10.75 10.75 8.86 8.86 10.7 10.7
    CWG 3.59 3.59
    Soy oil 1.25 1.25 1.25
    CLA 1.25 1.25 1.25
    Dicalcium phosphate 0.57 0.57 0.57 0.57 0.57 0.57
    Calcium carbonate 0.77 0.77 0.77 0.77 0.77 0.77
    Salt 0.25 0.25 0.25 0.25 0.25 0.25
    Vitamin premix7 0.20 0.20 0.20 0.20 0.20 0.20
    Mineral premix8 0.05 0.05 0.05 0.05 0.05 0.05
    Tylosin premix9 0.05 0.05 0.05 0.05 0.05 0.05
Calculated analysis
    ME, kcal/kg 3,333 3,333 3,499 3,499 3,499 3,499
    CP, % 10.6 10.6 11.4 11.4 11.1 11.1
    Crude fat, % 4.54 4.54 7.24 7.24 7.93 7.93
    Lys, % 0.60 0.60 0.64 0.64 0.64 0.64
    Met + Cys, % 0.46 0.46 0.48 0.48 0.44 0.44
Analyzed, %
    Lys 0.58 0.54 0.62 0.60 0.59 0.59

View Full Table | Close Full ViewTable 3.

Fatty acid composition of the CLA source used for dietary supplementation

Fatty acid
    16:0 (palmitic) 4.6
    18:0 (stearic) 3.4
    18:1 (oleic) 23.8
    18:2 (linoleic) 7.7
Total CLA isomers
    cis-9, trans-11 and trans-9, cis-11 12.3
    trans-10, cis-12 14.0
    trans-9, trans-11 and trans-10, trans-12 5.2
    cis-8, trans-10 and trans-8, cis-10 8.3
    cis-9, cis-11 1.3
    cis-11, trans-13 and trans-11, cis-13 12.2
    cis-8, cis-10 0.5
    cis-10, cis-12 2.1
    cis-11, cis-13 0.4
Other 4.2

View Full Table | Close Full ViewTable 4.

Carcass measurements of pigs fed varied fat sources

Carcass wt, kg 82.8 82.4 82.3 80.8 84.7 82.2 NS2
Fat depth, cm
    First rib 3.78 3.81 4.01 4.11 3.68 3.86 NS
    Last rib 2.69 2.44 2.92 2.69 2.67 2.64 NS
    Last lumbar 2.21 2.03 2.49 2.44 1.98 2.44 <0.043
    10th rib 2.31 2.16 2.51 2.59 2.08 2.57 <0.043
LM area, cm2 40.58 42.32 39.42 38.58 39.94 44.32 NS
Leaf fat, kg 1.16 1.17 1.41 1.34 1.10 1.54 NS

View Full Table | Close Full ViewTable 5.

Belly bar firmness test of bellies and compression values (force) of bacon from pigs fed various fat sources

LD,4 cm 4.39a 7.01b 6.12 5.33 5.65 0.58
FD,5 cm 5.75a 10.54b 9.26c 6.94d 8.24cd 0.75
Force, kg 9.41a 11.12b 10.96 9.73 10.12 0.62
n 24 23 16 16 15

View Full Table | Close Full ViewTable 6.

Belly and bacon processing characteristics of bellies from pigs fed various fat sources

Processing yield
    Green wt,7 kg 5.45 5.73 5.54 5.17 5.69 5.83 0.18 0.19
    Pump wt, kg 6.15 7.00 6.77 6.07 6.88 6.70 0.26 0.07
    Hot wt, kg 5.64 6.05 5.93 5.41 6.09 6.04 0.18 0.08
    Chilled wt, kg 5.47 5.79 5.74 5.28 5.87 5.87 0.17 0.12
Bacon slicing
    Green wt,8 kg 5.24 5.30 5.44 5.13 5.34 5.53 0.13 0.38
    Acceptable, kg 3.71 3.96 4.04 3.81 3.80 4.05 0.12 0.29
    Unacceptable, kg 0.31 0.27 0.25 0.27 0.27 0.41 0.10 0.24
Lipid oxidation9
    TBA, 1 mo 0.1575 0.1675 0.1800 0.1562 0.1850 0.1585 0.014 0.61
    TBA, 2 mo 0.1500 0.1525 0.1625 0.1400 0.1837 0.1585 0.009 0.04
    TBA, 3 mo 0.1612 0.1387 0.1487 0.1562 0.1700 0.1457 0.013 0.61
    TBA, 4 mo 0.1425 0.1412 0.1425 0.1337 0.1600 0.1371 0.009 0.49
    TBA, 5 mo 0.1550 0.1550 0.1625 0.1475 0.1875 0.1557 0.012 0.32

View Full Table | Close Full ViewTable 7.

Proximate analysis and cooking yields of bacon from pigs fed various fat sources

    Raw bacon 36.34 35.90 42.16 38.06 38.07 38.64 2.31 0.19
    Fried bacon 34.15 35.89 33.33 36.33 33.13 37.90 1.51 0.35
    Microwave bacon 35.61 35.10 36.14 33.68 34.13 35.52 2.07 0.48
    Raw bacon 47.02 46.35 41.89 45.31 45.13 44.70 1.96 0.21
    Fried bacon 16.27 11.79 12.30 13.66 12.83 11.30 2.75 0.33
    Microwave bacon 12.66 12.34 12.39 14.29 13.85 16.36 3.48 0.87
    Raw bacon 4.84 7.40 4.38 6.28 3.96 7.33 1.98 0.28
Cooking yield, % loss
    Fried bacon 66.38 68.89 69.46 68.08 69.18 69.72 1.46 0.44
    Microwave bacon 68.17 66.71 68.76 66.61 68.54 63.95 2.28 0.62

View Full Table | Close Full ViewTable 8.

Effects of dietary CLA on fatty acid profiles of bacon from pigs fed various fat sources

Fatty acids, %
    16:0 (palmitic) 25.27a 29.63b 28.76c 26.71d 26.87d 0.77
    18:0 (stearic) 13.23a 16.39b 15.32 14.70 14.40 0.65
    16:1 (palmitoleic) 2.76a 3.12b 3.09c 2.62d 3.11c 0.12
    18:1 (oleic) 39.63a 31.77b 35.19 34.47 37.43 1.37
    18:2 (linoleic) 13.46a 11.19b 10.87c 15.14d 10.96c 0.73
    18:3 (linolenic) 2.84 2.90 2.89c 2.44d 3.29e 0.12
    20:4 (arachidonic) 1.39a 1.14b 1.20 1.29 1.30 0.06
    Total SFA 38.49a 46.01b 44.08 41.41 41.27 1.37
    Total PUFA 60.09a 50.11b 53.24 55.96 56.09 1.58
    Iodine value 67.44a 57.25b 60.21c 65.30d 63.36d 1.14
CLA isomers
    trans-9, trans-11 0a 0.050b 0.030 0.027 0.018 0.03
    cis-9, trans-11 0a 1.19b 0.611 0.592 0.586 0.12
    trans-10, cis-12 0a 0.930b 0.475 0.467 0.453 0.09
    n 24 23 16 16 15

View Full Table | Close Full ViewTable 9.

Sensory panel results of cooked bacon from pigs fed various fat sources1

Aroma 5.18 4.77 4.91 5.02 5.12 4.69 0.209 0.34
Lean color8 5.36 5.10 5.47 5.35 5.06 5.29 0.16 0.53
Flavor 5.08 5.63 5.55 5.34 5.76 5.32 0.24 0.33
Off-flavor 1.87 2.16 1.98 2.15 2.28 2.06 0.17 0.41
Brittleness 4.56 4.97 4.52 4.31 4.99 4.81 0.27 0.22