|
|
||||||||
Animal and Dairy Science Department, University of Georgia, Athens 30602
| Abstract |
|---|
|
|
|---|
Key Words: Adipocytes Conjugated Linoleic Acid Gene Expression Omega-3 Fatty Acids
| Introduction |
|---|
|
|
|---|
| Use of Dietary Fat to Create Value-Added Products with Altered Fatty Acid Profiles |
|---|
|
|
|---|
Omega-3 Fatty Acids
The primary omega-3 fatty acids that have been of interest are a-linolenic (ALA, 18:3
9, 12, 15), eicosapentaenoic (EPA, 20:5
5, 8, 11, 14, 17), and docosahexaenoic (DHA, 22:6
7, 10, 13, 16, 19) acids. This series of fatty acids are metabolized to the Group 3 series of eicosanoids. In contrast, n-6 fatty acids, which include linoleic (18:2
9, 12) and arachidonic acids (20:4
5, 8, 11, 14), are metabolized to the Group 1 and 2 series of eicosanoids (Chapkin, 2000
). Interest in increasing the intake of n-3 fatty acids is associated with health benefits of these compounds or their metabolites (Chapkin, 2000
). In general, the eicosanoids produced from n-3 fatty acids, particularly EPA and DHA, are less inflammatory, cause vasodilation, and inhibit platelet aggregation, compared with those produced from n-6 fatty acids. Rodents fed fish oil have reduced adipose tissue mass, although these effects may be gender and strain specific (Belzung et al., 1993
; Fickova et al., 2002
).
Studies of primitive diets indicate that humans likely evolved eating a diet with an n-6:n-3 ratio of 1:1 to 4:1 (Eaton et al., 1996
). Due to the high intake of corn and soy oil, the current diet ratio in developed countries is greater than 10:1. Thus, future dietary guidelines in the US will likely not only have recommendations for total fat and polyunsaturated:saturated ratio, but also a recommendation for an n-6:n-3 ratio. Such guidelines exist in the United Kingdom (Scollan et al., 2001b
).
-Linolenic acid can be elongated and further desaturated to give rise to EPA and DHA and is particularly abundant in linseed oil (ALA, 53%; n-6:n-3 ratio = 0.26). Feeding linseed oil or full-fat flaxseed results in increased tissue content of ALA in pigs (Romans et al., 1995a
,b
; Van Oeckel et al., 1996
) and cattle (Scollan et al., 2001a
,b
). Canola or rapeseed oil contains approximately 9% ALA and has an n-6:n-3 ratio of 2.5) and also increases the ALA content of pork when fed to pigs (Leskanich, et al., 1997
). Feeding 10% flaxseed (37% crude fat in seed) for 25 d before slaughter, resulted in increased ALA (control = 0.19, 15% flaxseed = 0.75 mg of ALA/g of tissue) and EPA (control = 0.033, 15% flaxseed = 0.055 mg of EPA/g of tissue), but no change in DHA in muscle (Romans et al., 1995a
).
Because EPA and DHA are more direct precursors for the Group 3 eicosanoids, it would seem more efficient to feed them directly rather than depending on conversion of ALA. In general, EPA and DHA account for 15 to 20% of the fatty acids in most fish oils. Feeding fish oil to nonruminants results in increased tissue content of these fatty acids. Irie and Sakimoto (1992)
reported that feeding diets with 6% fish oil to pigs for 4 wk resulted in a fivefold increase in EPA and a 10-fold increase in DHA content of pork.
There are, however, potential negative effects associated with an increase in n-3 fatty acid content. Because n-3 fatty acids are liquid at room temperature, one of the common observations in products enriched with these fatty acids is that they are oily and often unacceptable to consumers. Although not observed in all studies, there is potential for reduced product shelf life and increased susceptibility to oxidative damage (Wood and Enser, 1997
). Some studies report increased off-flavors in n-3-enriched products. These effects are less likely with ALA than with the fish oils. However, it is estimated that the conversion of ALA to EPA is only 15% efficient (Nettleton, 1991
; Winters et al., 1994
). In a study in cattle comparing the effects of linseed oil and fish oil with a control group, feeding linseed oil resulted in a twofold increase in the ALA content of muscle phospholipids and a 50% increase in EPA (no significant change in DHA). In contrast, feeding fish oil increases ALA by approximately 30%, and increases EPA and DHA by 150 and 112%, respectively (Scollan et al., 2001b
).
Thus, although it is possible to alter the fatty acid profile of animal products to enrich their n-3 content, negative effects, such as decreased firmness, reduced shelf life, and off flavors, are likely (Wood and Enser, 1997
). There are also considerable inefficiencies in the conversion of these fatty acids from the diet to product. Feeding a diet with 6% fish oil (approximately 1% EPA + DHA) resulted in pork containing approximately 2 mg of EPA and DHA per 100 mg of total fat (Irie and Sakimoto, 1993). A 100-g serving of pork containing 3 to 5% fat would contribute 60 to 100 mg of these fatty acids. Although this is significant, the current recommendations are for an average daily intake of 1 to 4 g. Assuming a feed intake of 3 kg/d, the pig would have consumed approximately 5 kg of fish oil over the 4-wk period. Finally, the costs associated with keeping these potentially value-added carcasses segregated from other product during processing are a further consideration. Direct consumption of fish is probably the most efficient way for humans to increase their intake of EPA and DHA.
Fatty acid profile of poultry products can also be influenced by diet. Sanz et al (1999a)
reported that whereas feeding polyunsaturated fats resulted in marked changes in the consistency of abdominal fat consistency, it resulted in only moderately higher susceptibility to lipid oxidation as compared to fat from birds fed more saturated fat sources. Zollitsch et al. (1997)
reported that feeding polyunsaturated fatty acids (soy or rapeseed oil) improved growth performance without negative effects on carcass characteristics. They attributed this effect to improved nutrient digestibility with unsaturated fatty acids relative to more saturated forms. Although this is a possibility, it does not explain the reduced fat pad weights that a number of investigators report with dietary PUFA in birds (e.g., Newman et al., 2002
).
It is much more difficult to manipulate the tissue fatty acid profile of ruminant animals (Wood and Enser, 1997
). Scollan et al (2001b)
examined diets containing approximately 6% oil from a saturated source, linseed oil, fish oil, or a combination of linseed and fish oil. They reported that whereas EPA and DHA contents doubled with fish oil feeding relative to the control group, the contribution as a dietary source of n-3 fatty acids was small (control = 13 mg; fish oil, 20:5 + 22:6 = 28 mg from a 100-g serving of beef). Hydrogenation rates for EPA and DHA were approximately 90% (Scollan et al., 2001a
). Management systems seem to have the potential to change fatty acid composition of beef more than dietary fat source. For example, grass-fed beef contains significantly more n-3 fatty acids than would be found in products from animals fed concentrate or feed lot diets (Duckett et al., 1993
).
Conjugated Linoleic Acid
Conjugated linoleic acid represents a group of positional and geometric isomers of linoleic acid. There are a number of reviews that summarize the history of CLA and its potential mechanism of action (Pariza, 1997
; Pariza et al., 2001
; Belury, 2002
). Conjugated linoleic acid is formed under anaerobic conditions in the rumen and large intestine by anaerobic bacteria (Ha et al., 1989
). The predominant isomer in these natural sources is the cis-9, trans-11 version, which has anticarcinogenic properties (Ip et al., 1994
). Synthetic forms of CLA are produced from oils high in linoleic acid oils, such as safflower oil, and these contain predominantly the cis-9, trans-11 and trans-10, cis-12 isomers. With the availability of the synthetic forms of CLA came the observation that CLA also had antiobesity effects (Park et al., 1997
). It has subsequently been shown that these can be attributed to the trans-10, cis-12 isomer (Park et al., 1999
).
The potential health benefits of CLA have contributed to an interest in identifying means to increase dietary intake. Developing animal products with increased amounts of CLA represents one means by which to accomplish this goal. Meat products from nonruminant animals contain low concentrations of CLA, primarily the cis-9, trans-11 isomer (Chin et al., 1992
). Numerous studies in pigs and poultry demonstrate that feeding synthetic forms of CLA results in increased tissue content (see Azain, 2003). For example, feeding 1.25% CLA to growing-finishing pigs for 29 d prior to slaughter resulted in an increase in the CLA content of loin muscle from undetectable levels in the control to 5.8 mg/g of fatty acid (Wiegand et al., 2002
). A 100-g serving of pork (assume 3 to 5% fat) from a pig fed CLA would provide 15 to 25 mg of CLA. In younger pigs fed CLA, the relative percentages incorporated into muscle were in the range of 2 to 5% of the fatty acids (Ramsay et al., 2001
). This would extrapolate to dietary levels of 60 to 250 mg/100-g serving. Based on extrapolation from rodent studies, a daily intake of 3 to 5 g of CLA would be needed to result in dietary levels similar to those that have anticarcinogenic and antiobesity effects in lab animals. Thus, as with n-3 fatty acids, it is possible to increase CLA content, but the cost effectiveness and practicality of doing so must be questioned.
Unlike n-3 fatty acids, feeding CLA does not have negative effects on carcass quality. One of the metabolic effects of CLA is that it inhibits
-9 desaturase (Lee et al., 1998
), the enzyme responsible for converting palmitate and stearate to palmitoleic and oleic acids, respectively. The net effect of this inhibition is an increase in saturated fatty acid content and a reduction in iodine value (Averette-Gatlin et al, 2002a
), which results in firmer carcass fat characteristics. Although an increase in saturation is contrary to current dietary recommendations, it should be pointed out that the calculated atherogenic index (Ulbricht and Southgate, 1991
) of CLA-enriched pork remains less than that of milk. A decrease in the degree of unsaturation in animal products due to CLA feeding also has the potential to increase product shelf life, although this has yet to be consistently demonstrated. At least two groups have reported that marbling is increased in CLA-fed pigs (Averette-Gatlin, 2002a
; Wiegand et al., 2002
). This has not been consistently observed, however (Dugan et al., 1997
; Eggert et al., 2001
; Tischendorf et al., 2002
).
| Dietary Fat and Adipose Tissue Development |
|---|
|
|
|---|
Effects of Dietary Fat on Cell Number
Preadipocytes represent a pool of cells that have the potential to either proliferate or differentiate into adipocytes. Differentiation is characterized by expression of adipocyte specific proteins such as lipoprotein lipase (Cornelius et al., 1994
) and is regulated by transcription factors such as CCAATT/enhancer binding protein-
(C/EBP-
) and peroxisome proliferator activated receptor-
(PPAR-
; MacDougald and Lane, 1995
; Shao and Lazar, 1997
). Studies in cell culture systems clearly show the effects that fatty acids have on proliferation and differentiation of these cells (Amri et al., 1994
). The mechanism by which this occurs involves the fatty acids or their metabolites (Duplus et al., 2000
, 2001
). In cases where gene expression is affected, it is likely that fatty acids are working through transcription factors such as the peroxisome proliferator activated receptors or PPAR. There are numerous reviews that demonstrate that the fatty acids or their metabolites such as the prostaglandins are ligands for PPAR (Kliewer et al., 1997
; Hertzel and Bernlohr, 1998
; Sessler and Ntambi, 1998
). In rodents, PPAR-
is highly expressed in liver, kidney, heart, and brown adipose tissue, but is not detected in white adipose tissue (Schoonjans et al., 1996
). Stimulation of PPAR-
by fatty acids or specific ligands such as the fibrates, results in increased fatty acid oxidation (Kliewer et al., 1997
; Wolfrum et al., 2001
). PPAR-
2 is highly expressed in white adipose tissue (and at low or undetectable levels in liver) and its stimulation by fatty acids or specific ligands such as the thiazolidinediones results in expression of markers associated with preadipocyte differentiation (Tontonoz et al., 1994
; Schoonjans et al., 1996a
,b
; Gregoire et al., 1998
; Thuilier et al., 1998
).
In contrast to what has been well documented in the rodent, both PPAR-
and -
are expressed in porcine adipose tissue (Houseknecht et al., 1998
; Ding et al., 2001b
; Spurlock et al., 2002
). The significance of this is not yet clear, but emphasizes the danger in extrapolating data from rodent or cell culture systems to large animals. Similarly, both isoforms (
-1 and
-2) of PPAR-
are expressed in bovine adipose tissue (Sundvold et al., 1997
). Expression of PPAR-
has not been reported in the bovine. However, Kawada et al. (1998) reported that Wy14,643, a selective ligand for PPAR-
, stimulated differentiation of bovine fibroblast-like cells into adipocytes in vivo. This response was less potent than that seen with the thiazolidinediones, which are selective for PPAR-
, suggesting a more important role for this isoform in differentiation.
In porcine preadipocyte cultures, fatty acids have been shown to regulate gene expression of both transcription factors and markers of adipocyte differentiation (Ding et al., 2001a
,b
). However, a clear relationship of fatty acids and transcription factor expression in relation to adiposity in vivo has not been established (Spurlock et al., 2000
, 2002
; Ding et al., 2003
). Based on patterns of transcription factor and adipocyte specific protein, McNeel et al. (2000)
concluded that use of cell culture systems with porcine stromal vascular cells as a source of preadipocytes was a valid method for investigation of adipose development in the pig. However, they also noted that the relative pattern and magnitude of change for various transcripts differed from the in vivo situation. Recently, this group (Ding et al., 2003
) reported that despite strong in vitro effects of fatty acids on transcription, there were no changes in vivo in response to feeding fat. This may be due to the presence of mixed cell populations in vivo, competing fatty acid metabolism, time of sampling, or various other factors.
Effects of CLA on Adipocyte Differentiation
The CLA isomers are also ligands for PPAR (Belury et al., 2002
; Ding et al., 2000
) and have been shown to affect adipose development and gene expression in vitro. In vitro, CLA has been shown to inhibit proliferation of 3T3-L1 cells, a cell line commonly used for the study of adipocyte development (Brodie et al., 1999
; Satory and Smith, 1999
). A potential application of this observation is that exposure of an animal to CLA at the time when fat cell number is being determined might ultimately allow for control of adiposity. We have tested this possibility in both the pig and rat. In the rat study (Poulos et al., 2001
), pregnant females were fed diets containing 0 or 0.5% CLA beginning on d 7 of gestation and continuing through d 21 of lactation. Progeny were monitored at 11 wk of age. There was no effect of maternal diet on adipocyte size or number. It should be noted that whereas it was our expectation that feeding CLA during gestation would result in accumulation in the fetus, we were unable to detect CLA in the pups at birth. CLA was detected at 11 and 21 d of age, so it is likely that it is transferred in milk.
In the pig study (Poulos et al., 2000
), sows were fed 0 or 0.5% CLA beginning on d 40 or 70 of gestation and continuing through d 21 of lactation (weaning). Carcass composition of selected progeny was determined by ultrasound at 100 kg of BW. As in the rat study (Poulos et al., 2001
), there was no influence of maternal diet on body composition at market weight and we were unable to detect CLA in pigs at birth. Sow milk samples collected on d 7 and 21 of lactation showed significant increases in CLA in response to feeding. Conjugated linoleic acid also caused a depression in milk fat.
These studies fail to confirm that CLA can affect fat cell number as was reported in vitro (Brodie et al., 1999
; Satory and Smith, 1999
). In many cases, effects of CLA in culture have been investigated with CLA as the only lipid in the media. This is not physiological. There is a need to conduct additional work with mixed fatty acids as would be typical in vivo.
Effects of Dietary Fat on Cell Size.
In comparison with diets with no added fat, there are characteristic changes in performance associated with diets supplemented with fat. In the pig, as well as in other livestock, these include reduced intake, improved gain and feed efficiency, and increased carcass fat. The increase in carcass fat is most likely the result of increased fat cell size (Steffen et al., 1978
). Addition of fat to the diet results in an increase in the energy density of the diet, which reduces intake (NRC, 1987
). The improved growth performance is likely a combination of the effects of dietary fat on reducing gut passage rate and increasing digestibility of other nutrients, and the metabolic effects that result in increased net energy availability. Pettigrew and Moser (1991)
summarized the effects of dietary fat on performance and carcass characteristics from over 90 studies and concluded that the increase in carcass fat was independent of whether the calorie:protein ratio in the diet was maintained. Earlier work (Allee, 1985
) had suggested that the effects of dietary fat can be offset if the calorie:protein ratio is maintained. However, the extra-caloric and extra-metabolic effects of dietary fat result in greater efficiency of digestion and energy retention, which most likely account for the increased carcass fat despite this adjustment.
When fat is included in the diet, there is an inhibition of de novo lipogenesis, and the fatty acids that are deposited in adipose tissue switch from being of endogenous origin to that characteristic of the exogenous fat source. There are important species differences in the primary site of lipogenesis and in the effect of exogenous fatty acids on the rate of lipogenesis. In rodents, lipogenesis occurs in both liver and adipose tissue, whereas in pigs and ruminants it is primarily in adipose tissue. The liver is the main site of lipogenesis in the bird (Allee et al., 1971a
,b
, 1972
; Steffen et al., 1978
; Donaldson, 1985
).
The literature on the effects of the type of dietary fat on lipogenesis is dominated by studies in rodent liver tissue demonstrating the effects of the degree of unsaturation on inhibition of fatty acid synthase (e.g., Clarke, 2001
). In fact, the effects of degree of unsaturation are specific to liver. There is no difference in the degree of inhibition by saturated and unsaturated fatty acids in rodent adipose tissue (Shillabeer et al., 1990
). This suggests that the promoter regions of fatty acid synthase differ between tissues.
Lipogenic enzyme activity and overall lipogenesis are low in the liver of pigs (Lee et al., 2000
; Ding et al., 2001
; Gondret et al., 2001
). There are reports from three independent laboratories that inhibition of lipogenesis in adipose tissue of the pig is stronger with saturated fat sources than with unsaturated sources (Allee et al., 1972
; Camara et al., 1996
, Smith et al., 1996
). Dietary fat inhibits lipogenesis in ruminants (Deeth and Christie, 1979
; Emery, 1979
; Page et al., 1997
), but it is not clear if there are differences in potency for different degrees of unsaturation.
Several studies report that feeding polyunsaturated fatty acids to chickens results in reduced abdominal and total carcass fat as compared to that in birds fed saturated fatty acid sources (Sanz et al., 1999a
,b
, 2000
; Crespo and Esteve-Garcia, 2002a
,b
; Newman et al., 2002
). The calculated contribution of endogenous synthesis of fat to depot fat indicated that there was less endogenous contribution when birds were fed saturated fat (Crespo and Esteve-Garcia, 2002b
). This was unexpected given that other investigators reported that, as with rodents, lipogenesis was inhibited to a greater extent in chicken hepatocytes with unsaturated vs. saturated fatty acids (Lien et al., 2000
; Sanz et al., 2000
.).
It seems that in species where the liver is the primary site of lipogenesis, unsaturated fatty acids are more inhibitory than saturated ones. Such is the case in rodents and poultry. In species where adipose tissue is the primary site of lipogenesis, saturated fatty acids are equivalent (or more potent) than unsaturated. The regulation of hepatic expression of lipogenic enzymes such as fatty acid synthase has been well characterized, but there has been little investigation of the regulation of the gene in adipose tissue in rodents or other species. In particular, there is a need to characterize the potential differences in the promoter regions of these genes in various tissues. Such a characterization may help to explain species and depot differences in the anti-lipogenic response to unsaturated fatty acids.
There are limited studies on the specific effects of n-3 fatty acids on fat cell size in livestock or poultry. In rats, there is evidence that animals fed fish oil have reduced cell size and reduced fat pad weights (Belzung et al., 1993
; Su and Jones, 1993
; Fickova et al., 2002
). This effect seems to be depot, gender, and strain dependent, and there are no corresponding observations in livestock. At least part of these effects may be accounted for by the potent effects of fish oil fatty acids on lipogenesis in the liver (Herzberg and Rogerson, 1988
).
Effects of CLA on Cell Size
With reports of the antiobesity effects of CLA in rodents, there was interest in the potential to use CLA as a repartitioning agent in meat animals. In rats fed CLA, the antiobesity effect is accounted for by decreased lipid filling of adipocytes, and thus, decreased cell size (Azain et al., 2000
). In livestock, the effects of dietary CLA on body fat have been less than that observed in rodents and have been inconsistent. The effects dietary CLA in finishing pigs has recently been reviewed (Azain, 2003). Two factors that appear to account for the variation in carcass fat response to dietary CLA are the level of fat in the diet and the amount of carcass fat in the control group. Conjugated linoleic acid decreases subcutaneous fat thickness when there is little or no other fat added to the diet and when the level of 10th-rib fat in the control group is greater than 23 mm. The relevance of fat thickness is confounded by gender, in that barrows are generally fatter than gilts. There are a few specific studies with appropriate controls that support this explanation for the variation in response.
When CLA was added to diets with 2% added fat, there was an 11% (P < 0.05) decrease in subcutaneous fat thickness. However, at 5% added fat, there was no significant CLA effect (Dugan et al, 2001
). The gender and fat thickness interaction is illustrated in another study (Tischendorf et al., 2002
). In barrows, the control group had 26 mm of 10th-rib fat thickness and CLA reduced fat thickness 11% (P < 0.01). In this same study (Tischendorf et al., 2002
), however, there was no effect of CLA in gilts that had 20 mm of fat. To further illustrate this point, Averette-Gatlin et al. (2002)
examined the effects of CLA on barrows and gilts with and without 4% added fat. They observed no effect of CLA in any of the groups. The key here is that both barrows and gilts had less than 15 mm of fat, well below the proposed threshold of 23 mm. Thus, studies in rodents suggesting that the effects of CLA are independent of diet and background level of carcass fat (Park et al., 1997
; DeLany et al., 1999
) are not necessarily applicable to other species. It would appear that other than alterations in fatty acid profile, which results in increased carcass or belly firmness, there is not a consistent advantage to feeding CLA to pigs.
| Summary and Future Direction |
|---|
|
|
|---|
Cell culture studies clearly demonstrate that fatty acids can modify gene expression and affect adipose development. In vivo studies are less clear. The lack of confirmation of in vitro results may be due to the artificial environments used in cell culture studies or to complexities of in vivo systems. Much of what is known about fatty acid metabolism and effects on gene regulation is based on information gathered in experiments with laboratory animals and from cell lines in culture. There are important species differences that must be considered in applying this information to livestock and poultry (Table 1
). In many cases, species-specific details for nonlaboratory animals are not known. There is a need for investigation of the differences in the regulation of key enzymes and expression of regulatory proteins such as PPAR. The emphasis over the last few decades on reduction of the total amount of adipose tissue needs to be refocused on identification of factors that allow control over the development of specific adipose depots, such as between subcutaneous and intramuscular sites.
|
| Implications |
|---|
|
|
|---|
| Footnotes |
|---|
2 Presented at the ASAS Symposium: Alternative Aspects of Adipocyte Function, Phoenix, AZ. The author wishes to express his appreciation to S. Duckett, G. Hausman, H. Mersmann, and S. Poulos for their input and assistance in the preparation of this manuscript. ![]()
3 Correspondencephone: 706-542-0963; fax: 706-542-0399; e-mail: mazain{at}arches.uga.edu.
Received for publication July 10, 2003. Accepted for publication October 11, 2003.
| Literature Cited |
|---|
|
|
|---|
). Nutr. Res. 22:817824.
). Gene 225:8996.[Medline]
but not obese gene expression by dietary fat supplementation. J. Nutr. Biochem. 11:260266.[Medline]
1 and
2: Genetic mapping and differential expression of the two isoforms. Biochem. Biophys. Res. Commun. 239:857861.[Medline]
- and
-mediated gene expression via liver fatty acid binding protein: A signaling path to the nucleus. Proc. Natl. Acad. Sci. USA 98:23232328.This article has been cited by other articles:
![]() |
S. de Ferranti and D. Mozaffarian The Perfect Storm: Obesity, Adipocyte Dysfunction, and Metabolic Consequences Clin. Chem., June 1, 2008; 54(6): 945 - 955. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. W. Hess, G. E. Moss, and D. C. Rule A decade of developments in the area of fat supplementation research with beef cattle and sheep J Anim Sci, April 1, 2008; 86(14_suppl): E188 - E204. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Guo, S. Wong, W. Xie, T. Lei, and Z. Luo Palmitate modulates intracellular signaling, induces endoplasmic reticulum stress, and causes apoptosis in mouse 3T3-L1 and rat primary preadipocytes Am J Physiol Endocrinol Metab, August 1, 2007; 293(2): E576 - E586. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||