点击显示 收起
1 From the Food Research Program, Agriculture and Agri-Food Canada, Guelph, Ontario, Canada (JKGK, CC-H, and JZ); the Department of Food Science, University of Nanchang, Nanchang, China (ZD); the Institute of Nutrition, Friedrich Schiller University, Jena, Germany (GJ); and the Lacombe Research Center, Agriculture and Agri-Food Canada, Lacombe, Alberta, Canada (MERD)
2 Presented at the workshop "The Role of Conjugated Linoleic Acid in Human Health," held in Winnipeg, Canada, March 13–15, 2003. 3 Supported by a research grant from the Dairy Farmers of Ontario (to JKGK and CC-H). Milk samples were supplied from the fish meal studies of BW McBride, Department of Animal Science, University of Guelph (contribution number S146 from the Food Research Program, Agriculture and Agri-Food Canada). 4 Address reprint requests to JKG Kramer, Food Research Program, Agriculture and Agri-Food Canada, 93 Stone Road West, Guelph, Ontario, Canada, N1G 5C9. E-mail: kramerj{at}agr.gc.ca.
ABSTRACT
The chemistry of conjugated fatty acids, specifically octadecadienoic acids (18:2; commonly referred to as conjugated linoleic acid, or CLA), has provided many challenges to lipid analysts because of their unique physical properties and the many possible positional and geometric isomers. After the acid-labile properties of CLAs during analytic procedures were overcome, it became evident that natural products, specifically dairy fats, contain one dominant (c9,t11-CLA), 3 intermediate (t7,c9-, t9,c11-, and t11,c13-CLA), and up to 20 more minor CLA isomers. The best analytic techniques to date include a combination of gas chromatography that uses 100-m highly polar capillary columns, silver ion-HPLC, and a combination of silver ion–thin-layer chromatography and gas chromatography to analyze the CLA and trans 18:1 isomers, because some of them serve as precursors of CLA in biological systems. These analytic techniques have assisted commercial suppliers to prepare pure CLA isomers and have permitted the evaluation of individual CLA isomers for their nutritional and biological activity in animal and human systems. It is increasingly evident that different CLA isomers have distinctly different physiologic and biochemical properties. These techniques are essential to evaluate dairy fats for their CLA content, to design experimental diets to increase the amount of CLA in dairy fats, and to determine the CLA profile in these CLA-enriched dairy fats. These improved techniques are used to evaluate the CLA profile in pork products from pigs fed different commercial CLA mixtures.
Key Words: Conjugated linoleic acid • trans 18:1 • gas chromatography • silver ion–thin-layer chromatography • silver ion-HPLC • animal fats
INTRODUCTION
The history of conjugated linoleic acids (CLAs) dates back to the 1930s when it was found that dairy fats contained compounds that absorbed in UV radiation at about 230 nm, and this absorption increased after extensive alkali saponification (1). In the 1950s it was shown that the cis/trans structure of the conjugated double bonds in these fatty acids (FAs) had a unique doublet in the trans infrared region (2), which has subsequently been used for their specific detection. Both of these properties are still being used today in CLA analysis. However, the complexity of the CLA mixtures has only become apparent with the use of gas chromatography (GC) since the introduction of the very long highly polar capillary columns (3-5). The recent findings that silver iron (Ag+) HPLC columns resolved both geometric as well as positional CLA isomers (3, 6-8) have made possible a powerful analytic tool to complement GC for the complete analyses of the CLA isomer profile.
However, the current upsurge in the interest of CLAs started after the discovery that these FAs were associated with anticarcinogenic activity (9). Several additional physiologic and pathologic responses have since been ascribed to CLA that include effects on several different kinds of cancers, metastases, atherosclerosis, diabetes, immunity, and body fat/protein composition; [reviewed in Pariza et al (10), Kritchevsky and Czarnecki (11), Banni (12), and Belury (13)]. Further research is needed to elucidate the biochemical mechanism(s) of the different CLA isomers and to definitively verify CLA-related responses.
CLA and monounsaturated fatty acids (MUFAs) are produced as intermediates of polyunsaturated fatty acids (PUFAs), specifically linoleic (18:2n–6) and linolenic acids (18:3n–3), by rumen bacteria (14, 15). Two major groups of rumen bacteria have been identified that isomerize either the c12 bond to t11, eg, Butyrivibrio fibrisolvens (14-16), or the c9 bond to t10, eg, Megasphaera elsdenii (17). The cascade of possible FAs from 18:2n–6 and 18:3n–3 by these 2 groups of rumen bacteria is shown in Figure 1. There are several dietary factors that influence rumen bacterial population to produce either t11 or t10 containing FAs. The feeding of high-concentrate diets (18, 19) or the addition of vegetable oils, crushed oilseeds, or fish oil to the diet of ruminants is generally associated with reduced milk fat (20-26) and increased t10-18:1 and t10,c12-CLA (18-20, 26). In some cases there is a further decrease in milk production. For a more comprehensive review see Bauman and Griinari (27). The feeding of high-concentrate diets is a general practice when finishing beef cattle before slaughter which may result in increased levels of t10-18:1 and t10,c12-CLA, although a detailed analysis of the individual isomers has to be confirmed. Of interest, higher levels of t10,c12-CLA were reported in calf and horse sera (28), which could be due to the feeding of high-concentrate diets. The major trans 18:1 isomer is vaccenic acid (t11-18:1) under normal rumen conditions and t10-18:1 during conditions of rumen dysfunction (27, 29, 30) (Figure 1). However, ruminants that are exclusively pasture fed have been found to contain very high levels of t11-containing FAs, specifically c9,t11-CLA, t11,c13-CLA, and t11-18:1 (31; analyses of alpine cheese in this supplement).
FIGURE 1.. Possible metabolic intermediates of linoleic and linolenic acid produced by rumen bacteria. Isomerization followed by biohydrogenation (Biohyd.) in the normal rumen produces mainly t11-containing fatty acids, whereas during dysfunctional states mainly t10 fatty acids are produced (see text). Metabolites produced in the rumen can pass through the blood into tissues, including milk fat; the transfer of selected fatty acids is shown by dotted arrows. t11-18:1 is desaturated to c9,t11-18:2 by 9-desaturase, whereas t10-18:1 is not converted to t10,c12-18:2 in the tissue. The underlined trans double bond indicates the common trans double bond formed by the respective rumen bacteria.
The observation that several CLA isomers are synthesized in appreciable amounts in mammary and other tissues from trans 18:1 precursors in ruminants, in addition to those products in the rumen (19, 27, 32, 33), has forced analysts to consider both 18:1 and CLA isomers in their analysis. Although 9-desaturation activity was reported to occur in mammalian tissues some time ago (34, 35), the full extent of its involvement in CLA synthesis was only recently recognized (19, 31, 32). Evidence suggests that 9-desaturation could also play an important role in providing humans with increased CLA levels derived from dietary trans 18:1 precursors by the action of 9-desaturase (36-43). In fact, Banni et al (41) demonstrated that the feeding of t11-18:1 to rats reduced premalignant lesions in the mammary glands, a property previously associated with CLA, suggesting that t11-18:1 was converted to c9,t11-CLA by 9-desaturase.
Many CLA isomers have been identified in natural products, mainly in fats from ruminants, ranging from 7,9- to 12,14-CLA, through use of a combination of GC and Ag+-HPLC techniques. Each of the positional isomers occurs as 4 geometric isomers (cis,trans, trans,cis, cis,cis, and trans,trans) for a total of 24 (7, 8, 44-46). The most abundant CLA isomer in normal dairy and beef fats is c9,t11-CLA with smaller levels of t7,c9-CLA (44) and t11,c13-CLA (31), depending on the diet; the remaining CLA isomers are generally present at low concentrations.
It is increasingly evident that the different CLA isomers have different biological and pathologic effects; for detail see other papers in this supplement. With the availability of individual CLA isomers a thorough evaluation of the effects of each CLA isomer will become possible. Differences in the physiologic and pathologic responses of the CLA isomers are not surprising, considering that different CLA isomers have different physical and chemical properties. For example, geometric CLA isomers for a given positional CLA isomer have different melting points (t,t > c/t > c,c) (47) and oxidative rate (c,c > c/t > t,t) (48).
Several combinations of methods are presented to provide a complete analysis of the CLA and trans 18:1 isomers composition of different matrices. The need for a complimentary analysis that uses GC and Ag+-HPLC is mandatory for a complete analysis of CLA isomers, whereas a complementary analysis that uses GC and Ag+–thin-layer chromatography (TLC)/low-temperature GC is mandatory for the complete resolution of the trans 18:1 isomers in dairy fats and animal tissues. The examples considered in this report are commercial CLA preparations, dairy fats, and tissue lipids of animals fed synthetic CLA preparations.
ANALYSIS OF CONJUGATED LINOLEIC ACID PREPARATIONS
CLA is chemically synthesized by alkali isomerization from 18:2n–6 or oils rich in this FA (ie, sunflower or safflower oils) with the use of different solvents and conditions (49-51). The first products are c9,t11- and t10,c12-CLA in about equal proportions, followed by t8,c10- and c11,t13-CLA during excessive heating (3, 50-53) by way of sigmatropic rearrangement (54). In addition, these CLA preparations contain smaller amounts of corresponding cis,cis and trans,trans isomers from each positional isomer. The older CLA preparations generally contained all 4 positional isomers from 8,10 to 11,13 (Figure 2 A and Figure 3 A) that are now useful only for analytic purposes (4, 5) and are not recommended in feeding trials. Most of the CLA preparations currently available are composed of 2 major CLA isomers (c9,t11- and t10,c12-CLA), but every effort should be made to analyze the purchased CLA products to ensure purity. Analyses by TLC that use the developing solvent hexane:diethyl ether:acetic acid (85:15:1) is recommended to assess the chemical composition and overall purity of these products (55). Two commercially available CLA preparations are shown in Figures 2 and 3. Panels A and B in each of those 2 figures indicate the variation in CLA isomer distribution that one might encounter with use of GC and Ag+-HPLC, respectively. One CLA preparation contained as much as 4% of each one of the other 2 CLA isomers (Figures 2B and 3B), whereas the other is <0.5% (Figures 2C and 3C). However, that CLA preparation (Figure 2C) contained 2 unknown peaks that appear to be 8,10- and 12,14-CLA, although that identification has not been confirmed. The CLA prepared by dehydration of ricinoleic acid, and shown in Figures 2D and 3D, contained numerous additional CLA isomers, including large amounts of t,t-CLA. A note of caution: Minor compounds in CLA preparations should not be ignored because their potential biological activity could be significant. For further detailed isomer identification by GC and Ag+-HPLC, see related publications (3-5, 7, 8, 56, 57).
FIGURE 2.. Partial gas chromatographic separation of the conjugated linoleic acid (CLA) region. Mixture of 4 (A) and 2 (B and C) positional CLA isomers; B and C differed in the amount of t8,c10- and c11,t13-CLA present. CLA produced by dehydration of ricinoleic acid is shown in (D). For gas chromatography column and conditions, see the text. Sample A is a CLA preparation from Nu-Chek Prep (Elysian, MN); samples B to D are commercial CLA preparations provided by Derek Cornelius (Syntrax Innovations Inc, Cape Girardeau, MO).
FIGURE 3.. Silver ion-HPLC separation of the same 4 samples as in Figure 2. Operating condition: 3 silver ion columns in series; the mobile phase hexane:diethyl ether:acetonitrile (99.4:0.5:0.1) operated isocratically.
Generally, a lack of pure CLA isomers exists for identification purposes and to conduct experiments to evaluate specific responses. A few pure CLA isomers (c9,t11, c9,c11, t9,t11, and t10,c12) are available from Matreya Inc (Pleasant Gap, PA) besides the common CLA mixture containing mainly c9,t11- and t10,c12-CLA. Synthesis of additional CLA isomers for analytic purposes from 7,9- to 12,14-18:2 were recently reported through use of a combination of sigmatropic rearrangement followed by selenium-catalyzed geometric isomerization of known CLA isomers (58) or by a combination of partial hydrazine reduction of known PUFAs followed by alkali isomerization, isolation of products, and further iodine-catalyzed geometric isomerization (59, 60). Several individual trans 18:1 (t6, t7, t9, t11, t12, t13, and t15) and cis 18:1 (c6, c7, c9, c11, c12, c13, and c15) isomers are available from Sigma Chemical Co (St Louis). However, to obtain a complete profile of all trans 18:1 that are well recognized by their relative abundance from t4- to t16-18:1, the isolation of the trans fraction by Ag+-TLC of total milk fat fatty acid methyl esters (FAMEs) is highly recommended (4, 5); details to follow.
ANALYSIS OF CONJUGATED LINOLEIC ACID ISOMERS IN DAIRY FATS
The analysis of CLA isomers is demonstrated through use of 3 samples of milk fat from cows fed different diets and by comparing the GC (Figure 4) and Ag+-HPLC (Figure 5) results. The CLA mixture containing 4 positional CLA isomers served as a GC standard, and it was "spiked" with methyl heneicosanoate (21:0), because this FA is also present in milk fats (4, 5, 57). Total milk fat was converted to their FAMEs through use of sodium methoxide as catalyst (55, 61-63) and analyzed by GC with use of a 100-m CP Sil 88 fused silica capillary column (Varian Inc, Mississauga, ON) (4, 5). The milk fat from cows fed a normal total mixed ration consisted mainly of c9,t11-CLA and contained a considerable amount of t7,c9-CLA. This information could only be provided by the Ag+-HPLC separation (compare Figure 5B, C, and D with Figure 4B, C, and D). Small amounts of t10,c12- and several t,t-CLA isomers were present. The methyl ester of 21:0 generally causes interference among the minor CLA isomers with use of these GC columns (CP Sil 88 or SP 2560) and conditions (4, 53, 57, 64). Therefore, for identification purposes the addition of 21:0 FAME to the CLA mixture is recommended (Figure 4A). The addition of fish meal to the dairy ration increased the content of t7,c9-, t9,c11-, and t10,c12-CLA, whereas c9,t11-CLA remained constant (Figures 4C and 5C) (29). [Please note that the Ag+-HPLC separations represent relative concentration of CLA isomers, and the Ag+-HPLC chromatograms shown are intentionally enlarged to more clearly show minor CLA isomers that distort the relations of the CLA isomers somewhat.] Cheese prepared from the milk of cows grazed at high altitudes in the Alps showed a high content of the t11-containing CLA isomers c9,t11-, t11,c13-, and t11,c13-CLA (Figures 4D and 5D), whereas the amount of t7,c9-CLA was less compared with milk fats from cows reared in commercial operations (Figures 4 and 5, B and C). These results are consistent with results recently reported for the milk fat of cows grazed under similar alpine conditions (31).
FIGURE 4.. A partial gas chromatogram of the conjugated linoleic acid (CLA) region. The CLA mixture containing 4 positional CLA isomers spiked with 21:0 fatty acid methyl ester (FAME) as the standard (A). Also shown is the CLA region of total FAMEs prepared from milk fat from control cows (B) and fish meal–fed cows (C) and cheese obtained from the milk of cows grazed at high altitudes in the Alps (D). Sample A, CLA and 21:0 available from Nu-Chek Pep (Elysian, MN); samples B and C are from a study to evaluate fish meal addition in dairy rations (C Cruz-Hernandez, JKG Kramer, AR Hill, and BW McBride, unpublished data, 2004); sample D is cheese kindly provided by G Jahreis.
FIGURE 5.. Silver ion-HPLC separation of the same 4 samples shown in Figure 4.
CALCULATION OF CONJUGATED LINOLEIC ACID ISOMER COMPOSITION WITH USE OF GAS CHROMATOGRAPHY AND SILVER ION-HPLC
Analysis of the CLA region requires the combination of GC and Ag+-HPLC, shown in Figures 4 and 5, respectively. The CLA peaks are quantitated by GC analyses of total FAMEs. Three GC peaks in the CLA region could not be resolved (Figure 5): the "c9,t11-CLA" peak that could contain t7,c9-, c9,t11-, and t8,c10-CLA, the "c9,c11-CLA" peak that could also contain t11,c13-CLA, and the "t,t-CLA" peak that could contain several t,t-CLA isomers from 7,9- to 10,12-CLA. However, several CLA isomers are well resolved by GC, such as t9,c11-, t10,c12-, c10,c12-, c11,c13-, and t11,t13-CLA. The identification of 21:0 poses a challenge, because it generally occurs at similar concentrations as the minor CLA isomers and elutes anywhere between c11,t13- and c10,c12-CLA, depending on the GC column and the temperature program used (53, 57, 64). Partial GC chromatograms of the CLA region with use of the 21:0 spiked CLA standard from several CP Sil 88 columns and different times of the life of the column over the past 2 y in our laboratory are shown in Figure 6. The elution of 21:0 FAME was from c11,t13- to t10,c12-CLA. Therefore, the addition of 21:0 FAME and the 4 positional CLA isomer mixture (Figure 4A) to a GC standard such as #463 from Nu-Chek Prep (Elysian, MN) is recommended (4, 5).
FIGURE 6.. Representative gas chromatography separations of the conjugated linoleic acid (CLA) standard mixture spiked with methyl heneicosanoate (21:0). These separations were observed in our laboratory over the past 2 y with use of 3 separate 100-m CP Sil 88 columns (Varian Inc, Mississauga, ON) at different times throughout the life of these columns. The ratio of the CLA mixture and 21:0 will be slightly different because it represents several preparations of CLA and 21:0.
Ag+-HPLC clearly resolves each of the CLA isomers present, including the unresolved CLA isomers in the 3 GC regions (Figure 5). The relative concentrations obtained by Ag+-HPLC are used to calculate the unresolved peaks in the GC chromatogram. The t7,c9-, c9,t11-, and t8,c10-CLA isomers are resolved by Ag+-HPLC, but now the geometric CLA isomers of 9,11-CLA (t9,c11- and c9,t11-CLA) coelute. The t8,c10-CLA isomer is generally present at low levels in natural products, except when CLA preparations are used that are composed of a mixture of 4 positional CLA isomers. The coeluting pair of c9,c11- and t11,c13-CLA by GC is resolved by Ag+-HPLC. Finally, all the t,t-CLA isomers are well resolved by Ag+-HPLC. The content of the t10,c12-CLA should be carefully confirmed through use of the GC and Ag+-HPLC results, because 21:0 FAME could cause interference.
ANALYSIS OF trans 18:1 ISOMERS IN DAIRY FATS
The following GC temperature program was used to resolve all the FAMEs present in dairy fats, including the short-chain FAME from 4:0 to the long-chain PUFA, and took 86 min. The temperature program was 45 °C (held for 4 min), 13 °C/min to 175 °C (held for 27 min), 4 °C/min to 215 °C (held for 35 min) (4, 5). Sample loads were chosen such that near baseline resolution was obtained between t11- and t12-18:1. The sample load required for a good resolution of the 18:1 isomers generally proved insufficient for the analyses of many minor long-chain PUFAs as well as the minor CLA isomers. Therefore, 2 different dilutions were generally analyzed for total milk fat FAMEs (4, 5).
However, despite the use of 100-m highly polar capillary columns and selected temperature programs, the region between 18:0 and 18:2n–6 by GC is a complex mixture of many overlapping trans 18:1, cis 18:1, cis/trans 18:2 isomers and saturated FAs that are only partially resolved at best. Typical separations are shown in Figure 7 of the FAMEs in this region from milk fat of cows fed a control (Figure 7A) or a fish meal–containing diet (Figure 7B). The separation of FAMEs prepared from the fat extracted from a cheese sample that was prepared from the milk of cows that grazed at high altitudes in the Alps (cheese sample courtesy of G Jahreis) is shown in Figure 7C. The 18:1 region is further complicated by wide differences in the content and the relative abundance of trans 18:1 FAME isomers that one could encounter between diets and different samples. In the case of feeding fish meal, t10-18:1 obscured the cluster of the other trans 18:1 isomers from t6- to t11-18:1 (Figure 7B), whereas in the cheese fat t11-18:1 masked this region (Figure 7C).
FIGURE 7.. A partial gas chromatogram of the 18:0 to 18:2n–6 region showing the separation of the 18:1 and 18:2 fatty acid methyl ester isomers from control milk fat (A), milk fat from cows fed fish meal (B), and fat extracted from a cheese sample obtained from milk produced by alpine-fed cows (C). The extent of the trans- and cis-18:1 and the cis/trans-18:2 isomers is indicated. Samples A to C are the same milk fats used in Figures 4 and 5 (samples B to D).
A reliable analyses of all the trans and cis 18:1 isomers is only possible by a prior separation of the FAMEs by Ag+-TLC. Total milk FAMEs were fractionated into saturates, mono-trans and mono-cis FAMEs plus CLA with use of Ag+-TLC, as described elsewhere (4, 5, 65, 66). The fractions were dissolved in hexane and analyzed by GC with use of a stepwise isothermal temperature program, starting at 120 °C (4, 5). The cis and trans fractions were analyzed at high- and low-sample loads to resolve the minor and major FAME constituents.
Typical isothermal GC separations of the trans 18:1 FAME at low temperatures, starting at 120 °C, that yielded a complete resolution of all the trans 18:1 isomers from t4- to t16-18:1, except t6 to t8 that coelute, are shown in Figure 8. These low-temperature GC resolutions proved absolutely essential for the resolution of the trans 18:1 isomer mixtures of milk fats in which specific isomers dominated, such as t10 when fish meal was fed (Figure 8B) or t11 when analyzing the cheese fat from grazing cows in the Alpine region (Figure 8C). The t10 was clearly the predominant trans 18:1 isomer in the milk fat of cows fed fish meal (Figure 8B), whereas t11 isomer was the major isomer in milk fats from grazing cows (Figure 8C).
FIGURE 8.. A partial gas chromatogram of the 18:1 isomer region of the isolated trans band by using silver ion–thin-layer chromatography. The samples were the same as in Figure 7: control milk fat (A), milk fat from cows fed fish meal (B), and cheese prepared from the milk of cows that grazed in the Alps (C). A 100-m CP Sil 88 GC column (Varian Inc, Mississauga, ON) and a stepwise isothermal temperature program starting at 120 °C were used (4, 5).
The total trans 18:1 content of dairy fats was then calculated by multiplying the total concentration of t4- to t11-18:1 obtained by GC analysis of total FAMEs, by the ratio of the area response (t4- to t16-18:1) to (t4- to t11-18:1) obtained from the low-temperature isothermal GC analyses. The individual trans 18:1 isomers were calculated through use of the relative concentrations of the GC separation at low temperature of the trans fraction obtained by Ag+-TLC.
Several milk fats have been investigated with use of the combined Ag+-TLC/GC technique, including milk fat from cows (18, 19, 31, 64, 66-68), goats (68, 69), ewes (68), camel (70), and human milk (71, 72). This method has also been applied to the analysis of the alkyl chain of plasmalogenic lipids from sheep heart, which showed a similar profile to the trans 18:1 FA isomers present in sheep fat (73).
ANALYSIS OF CONJUGATED LINOLEIC ACID ISOMERS IN ANIMAL TISSUES
All tissues should be excised quickly, rinsed, frozen immediately between blocks of dry ice (or dropped into liquid N2), and stored at –70 °C until analyzed. To inhibit the action of lipases and phospholipases during homogenization of tissues, samples should be pulverized at dry ice temperature as described previously (74, 75). In addition to total tissue lipid, analysis of the individual lipid classes should be considered, because the incorporation of the CLA isomers into the individual lipid classes might be different, and the different lipid classes often have different functional properties (53, 76, 77).
A partial GC chromatogram of phosphatidylethanolamine isolated from liver lipids of pigs fed commercial mixtures that contained either 2 or 4 positional CLA isomers is shown in Figure 9 (53; Kramer and Dugan, unpublished data, 2004) The 4 CLA isomers fed could be easily recognized by GC analyses, including some of the minor c,c and t,t isomers, except for the unresolved peaks c9,t11-/t8,c10-, c9,c11-/t11,c13-, and t,t-CLA. All the CLA isomers were separated by Ag+-HPLC, and the CLA isomer composition of the major liver and heart lipids are shown in Figure 10. Several differences in CLA isomer composition depend on the lipid class, the tissue, and the CLA isomer mixture fed, suggesting possible differences in metabolism and activity. The greater accumulation of the c11,t13-CLA in the diphosphatidylglycerol (or cardiolipin) fraction in heart lipids was of concern (53) and led to the subsequent commercial preparations of CLA mixtures that contained mainly 2 CLA isomers (50). The results from Figure 8 would appear to suggest that a good GC analysis might be sufficient if only a 2-CLA isomeric mixture (or a single isomer) was fed to monogastric animals or humans. However, on the basis of our experience, the CLA profile is generally more complex than simple incorporation of the CLA isomer ingested. The complimentary Ag+-HPLC analysis often provides valuable information and confirmatory evidence in many different matrices. The analyses of the CLA isomers will require the results from both GC and Ag+-HPLC as described earlier. The analysis of the alk1-enyl moiety of the plasmalogens from sheep heart lipids is an excellent demonstration of the power of these techniques (73).
FIGURE 9.. A partial gas chromatogram from linoleic (18:2n–6) to arachidonic acid (20:4n–6) showing the accumulation of the conjugated linoleic acid (CLA) isomers in the phosphatidylethanolamine fraction of liver lipids from pigs fed 2 different CLA preparations, consisting of 2 or 4 positional isomers. A 100 SP 2560 GC column (Varian Inc, Mississauga, ON) was used as described in reference 62. The gas chromatography tracing from 4 positional isomers was selected from a previous study (53), whereas the results of 2 positional isomers is not published (Dugan and Kramer, 2004).
FIGURE 10.. Conjugated linoleic acid (CLA) isomer distribution in the major heart and liver lipid classes in pigs fed 2 different CLA preparations consisting of 2 or 4 positional isomers. Data were summarized from the 4 (53) and the 2 (Dugan and Kramer, unpublished data, 2004) CLA isomer studies. TG, triacylglycerol; PC, phosphatidylcholine; PE, phosphatidylethanolamine; DPG, diphosphatidylglycerol.
SUMMARY
This review concentrated on presenting the best analytic methods available to date for the complete analysis of CLA and trans 18:1 isomers with use of GC, Ag+-HPLC, and a combination of Ag+-TLC/GC. Complete information of the CLA isomeric composition is essential to ensure the purity and composition of commercial CLA preparations and individual CLA isomers. Methods to resolve individual CLA isomers are particularly important in nutritional or biochemical studies, because several CLA isomers were shown to have different biological effects or responses. In ruminant studies knowledge of both the trans 18:1 and CLA isomers is essential to evaluate dietary manipulation designed to enhance the CLA content in dairy and beef fats or inadvertently to alter the isomer composition to t10-containing FAs by feeding high-concentrate diets. Most of the previous feeding studies in ruminants will need to be reevaluated for isomer-specific changes of CLA and trans 18:1, much the same as was performed in the past (18, 19, 26). A detailed knowledge of the CLA isomer content and composition is also essential to evaluate the CLA isomers present in the meat products of monogastric (ie, pigs, chicken) and ruminant animals when fed mixed CLA preparations in attempts to enrich these food products with CLA.
This review does not include the important analytic methodologies required for the analyses of the CLA metabolites, ie, the elongated, desaturated, and chain-shortened products. Analyses of these metabolites require unique applications of reversed-phase HPLC (12, 78, 79) and complementary GC and mass spectrometry techniques (78-81), because of the diversity, low concentration, and decreasing stability of these metabolites.
ACKNOWLEDGMENTS
None of the authors had a conflict of interest.
REFERENCES